September 2, 2025 • 59 mins
Text us your thoughts on the episode or the show!
In this episode of Ops Cast by MarketingOps.com, powered by The MO Pros, Michael Hartmann speaks with Ahmed Datoo, CEO and co-founder of Allgood and a self-described recovering CMO.
Ahmed reflects on his time leading marketing teams and shares a candid perspective on why CMOs struggle with attribution, budget allocation for MarketingOps, and the ongoing pressure to justify impact across creative, analytical, and strategic dimensions. The conversation also covers the shift from rules-based to reasoning-based systems and the role of AI in modern marketing teams.
In this episode, you’ll learn
• Why attribution often creates more conflict than clarity
• What makes MarketingOps hard to fund and even harder to scale
• How AI and digital employees like “Mary” can ease team burnout
• The real reason CMOs burn out and what can be done differently
This is a must-listen for marketing and revenue professionals who want to understand how marketing leadership views operations, resourcing, and emerging technologies like AI.
Episode Brought to You By MO Pros
The #1 Community for Marketing Operations Professionals
Visit UTM.io and tell them the Ops Cast team sent you.
Join us at MOps-Apalooza: https://mopsapalooza.com/
Save 10% with code opscast10
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Ahmed Datoo (04:36):
This is a great question.
My understanding andappreciation of marketing
operations has evolved over theyears.
When I first became a CMO, Iwould classify my relationship
with MOPS as just a hugedisappointment, and the reason
(05:00):
for it is I was like, okay,great, I have a team that can be
my right hand, so to speak.
Like for me, I am a CMO that,um, like, I'm driven by data, I
like to have an understanding ofdata and then that helps,
either, helps influence myintuition, um, but I, I like to
(05:21):
be able to analyze data.
So we run a program.
I want to know what was thesuccess of that program.
I want to be able to analyzedata.
So we run a program.
I want to know what was thesuccess of that program.
I want to know, like, whichsets of marketing activities are
driving pipeline.
So to me, I viewed my marketingoperations team as, like, my
strategic hand, helped me pulltogether the kinds of analysis
that is going to help me makedecisions around allocation of
(05:43):
spend and what programs to berunning, who our icps should be,
etc.
Etc.
So that's my expectation goingin right, and what I found was
the marketing ops team was like,so in the weeds about executing
programs like marketingcampaigns and programs managing
the technology stack, all thatand managing the technology
(06:04):
stack, and like doing managingtechnical debt and doing
integrations, and like a newfeature has been released by a
vendor.
It drove me nuts.
I was like why are we wastingtime on this?
And, and now, remind you, myunderstanding of marketing
operations has evolved so.
So this was my going in justfrustration and I said how
(06:29):
difficult can this be?
Like?
Why does it take so long to dosomething like this?
And so I tend to learn best bydoing so.
Of course, I interact with theMOPS team.
They're like you have noappreciation around how
difficult it is.
I'm like how difficult can itbe to build?
At the time, I think we wereusing Eloqua.
(06:50):
So I was like how difficult canit be to build a campaign in
Eloqua?
Turns out it's really difficult.
And so they would walk methrough about, like you know, it
is very complex and blah, blah,blah.
And so I said you know what I'mgoing to build the campaign.
Can I do the next one?
And I got hands on, I figuredout how.
(07:13):
To you know, I read all theinstructions around building
emails and campaigns.
I was like this doesn't looktoo hard.
And I got the campaign brief.
And of course, the campaignbrief didn't map what was the
email template.
There I got the campaign briefand of course the campaign brief
didn't map what was the emailtemplate.
There are all these variations.
So immediately I'm thinking,okay, we should probably have
some guidelines to the team,like stay within the parameters
of this, but I'm like still, itshouldn't be that difficult to
(07:36):
do.
And then it was just likeendless clicks in the system and
sitting and waiting right,which was something that they
didn't communicate to me, whichis just the poor user experience
of the software and the realityof the software.
And then there's the back andforth between the marketer who
requested, saying don't quitelook like the way you've done
(07:56):
this, can we change this to bold?
And that required me loggingback in the system doing all
these changes.
So I began to have anappreciation, having gone
through that process myself, oflike okay, I get it, like I
understand that they are boggeddown in kind of building out
programs and I understand why.
And, um, I think the otherthing I didn't quite appreciate
(08:20):
is how many different systemsthat they would need to touch to
do something as simple as builda mail campaign.
So it wasn't just Eloqua, itwas like there are a variety of
other systems that they weretouching, and so once he, once I
had an appreciation of this.
Then I had empathy, which is OK, I get it.
Now I understand why they're notable to actually do what it is
(08:44):
that I need to do, and soexperimented with a whole bunch
of stuff, like what if I hiredan offshore team and try to get
rid of the repetitive work?
So we tried doing that and Iwas hoping that would free up
time for my MOPS team to do thestrategic important work that I
was really interested in doing.
And it didn't, primarilybecause the offshore team would
(09:10):
make mistakes and invariably.
You know, in your ideal worldyou have enough prep time for a
campaign.
And I'm guilty of this becauseI would be like, can we go run
this?
Now?
It's the middle of the quarter.
We're not going to meet ourpipeline numbers.
We need to spin up a bunch ofnew campaigns and like overnight
we got to spin stuff up andthere's always a lag when you're
(09:31):
dealing with offshore, sothere's a lag in the initial
creation of it, and then therewas a lag when we needed to do
revisions.
It just slowed the process down.
So I never had really goodexperience offshoring, so I
brought it back onshore.
So the the question around like, um, how, how?
Did I think about mops?
(09:52):
I had this, this appreciationof like this is what I want mops
to do, and, by the way, that'swhat mops wanted to do, like
it's not, like I was suggestingsomething that they viewed was
out of their job description.
They just simply didn't havethe time to do it Right.
And so what I ended up doing isbecause I'm a, you know,
technical CMO, I startedbuilding my own reports.
(10:14):
I was in there, I was, Icreated a data warehouse, I
started building my own sets ofreports and and I realized that
worked for me, but that doesn'twork for most cmos.
Like you know, I'm not going tobe the cmo who's going to be
getting into canva and doingdesign, because that's not where
I spike right.
So, but there are cmos would godo that, but would not go and
(10:36):
like, do the the snowflake dataanalysis type stuff that I was
doing.
So it's just this perpetualproblem that I was facing as a
CMO and this desire to be ableto drive that organization, to
help propel marketing, to beable to do great things, and so
(10:58):
that's something that was alwaysin the back of my head.
Yeah, it's interesting becausethe first thing you said really
is, you know what you wanted wasessentially insights, right?
How are things working?
What are the results?
How can we talk about it withthe rest of the organization?
And I suspect, if my experienceis any indicator, right, that
(11:21):
even if you could have figuredout a way to open up time on
that team, there might not havebeen a skill level to really do
it in a way that would have beenhelpful for you, right, because
I think there's a.
I think there's a.
I mean, I could get this really, really general, having kids
here in the college age.
You know that, like the, thewhole, like uh, math curriculum
(11:44):
for has really not changed in 40or 50 years, right, for kids
coming, and I think that's aproblem.
We get people come out and wejust don't know how to really do
data analysis, right, don'tknow how to put it together,
don't know how to interpret it,don't know how to look at it.
Um, you can't really talk aboutthe difference between a
snapshot in time of somethingversus something that's over
time and all those things thatare different ways of thinking
(12:07):
about it, and that's one of theconcerns and one of the things I
bring up all the time here isjust like we need more people
who can do that, and I'm bullishon things like AI tools being
able to enable some of that,whether it's by freeing up time
and capacity or by helping withdoing some of the heavy lifting
of that.
Whether it's by freeing up timeand capacity or by helping with
doing some of the heavy liftingof data, like organizing data
(12:29):
and doing some initial firstpass analysis.
But do you think the teams youhad before would?
Did you ever get, I guess,first, did you ever get to the
point where they could start todo some of that and did you feel
like they had the skills andcapability of doing that?
so definitely had the skills todo it.
So this this comes to you likehow I hire.
(12:51):
So, generally, when I, when Iwas hiring for anyone in any
role within marketing, it waslike intelligence first and
foremost, some form of curiosityand then some degree of domain
knowledge, and it was the domainknowledge in marketing ops that
always killed me, because Iwould find amazing people, but
(13:14):
they were HubSpot people and wewere a Marketo shop and it's
just too steep of a learningcurve, right.
Or I would find incrediblycurious, incredibly smart,
talented people, but they were.
You know, they spiked on acertain set of technologies that
didn't exist in our stack andthat's the critical component of
(13:36):
it, right.
And so you end up making thesecompromises to get in people who
have domain knowledge about thestack, and it's kind of crazy
that we make decisions.
We need to make decisions basedon familiarity with technology
as opposed to just smarts.
But I was fortunate.
(13:57):
The people that I ended uphiring were really good.
Uh, I mean, had experience withthe tech stack, but were really
smart.
So, yeah, I felt like if, ifthey, if given the time and
given a problem, they would gofigure it out.
Right, they may not necessarilyknow data analysis, but I could
probably work with them to say,like, here are the kinds of
reports that I'm looking for,and then be able to go figure it
(14:18):
out.
This is pre-AI right In an AIworld.
This is what's amazing about AI.
Like, I talk about myself andyou know I like colors.
When given something, I canreview it and tell you whether I
like it or not.
I don't necessarily feel likeI've got great design skill, in
(14:41):
fact, um, you know, Ideliberately stay away from.
I told my wife, like you canhandle all the decorations of
the house.
I won't have any sort of review.
The only thing I ask is that Iget to handle all the tech
gadgets like that.
That's going to be my, my,purview, right?
Michael Hartmann (14:58):
um, so I have my request.
My request was just only vetopower on like if really didn't
like something from a designlike I really want to be able
toanche on electronics.
Ahmed Datoo (15:08):
But um, you know, when it came to, when it came to
design, um, I would it's verydifficult, like I'd sit down
(15:33):
with a designer and the designerrightfully so would ask me a
question.
Like you know, give me yourdesign aesthetic.
And early in my my career I waslike I don't know what my
design aesthetic is, I'll knowit when I see it, which is like
the worst thing.
A designer hears Like um, howdo you work with that?
And so early on it'd be likethey'd give me something and I
(15:55):
would just tear it apart andthey're like, okay, you clearly
don't like that design.
Give me a couple of designsthat you do like, and so I'd
start talking about.
You know, here are vendors thatI like.
And so I began to realize Ibegan to speak the language of
the designer.
Michael Hartmann (16:07):
Right.
Ahmed Datoo (16:09):
Fast forward to today.
Now, all of a sudden, I can bea designer not a great one, but
I can iterate on that designprocess a lot faster.
So I can go into mid journeyright now, and do you know, to
mid journey right now.
And do you know, image creationaround concepts.
I can go into all these codingtools and actually create first
(16:30):
iterations of a homepage design,right To get to a point where
it's like, yeah, this is in thegeneral direction of where I
want to go, and then I can handit off to a professional to
either make it better or atleast have a starting point to
say, like this is kind of theaesthetic of what I'm looking
for, right and so.
And, by the way, I don't.
There are certain things whereI want to go to a designer
(16:52):
because, like the home page, Iwant to nail right.
But if I'm doing a landing pagefor a webinar that we're doing
next week, I don't necessarilyneed to go to my agency to nail
that right.
I could probably use the tool,the AI tools, myself and build
something comparable, because Ialready know what my brand voice
is, the style, whatever.
(17:12):
I can use those tools myself.
That was not possible at allright.
And so your point about peoplewho are not analytical or
quantitative.
Now, all of a sudden, you knowI started the conversation by
saying the CMO job andmarketing's job in general is
tough because you have to spikein all these different things.
I view this concept of now, asan employee, you will have
(17:34):
access to digital employees anda set of tools that allow you to
spike in each of thosecategories.
You'll be able to do designwork.
You can now write amazing copy.
You can do analytical work.
You can actually do operationsoriented work.
You can do strategic orientedwork.
Right like.
All of those are possible, um,and I think what's what's
(17:56):
required in that, in that sortof world, is taste.
So.
So, when you use these tools,you got to know what good looks
like.
Michael Hartmann (18:10):
Yes.
Ahmed Datoo (18:11):
Right.
So, like I know, when I getoutput from the AI tools for
something like an image, or evenwhen I'm working with copy
because I was a former productmarketer and I worked with
designers right Like, I havegotten an understanding of what
is considered good from amessaging standpoint, what is
(18:33):
considered good or good enoughfrom a design standpoint right,
and so I think anyone who hasexperience right now in
marketing can be superchargedwith the use of these tools yeah
, I'm finding the same, if I forthose who are listening, right,
I had a weird look on my facewhen he said that you have that,
what was the word you used?
Michael Hartmann (18:54):
um, taste, you have to have taste.
And I was like, what does hemean by that?
And when you describe, like youknowing what was what good is,
which is also subjective, rightin a lot of these cases.
But, like, for me, one of thethings I've learned I have to do
when I'm using, just keep itdown to LLMs, right, it's like I
, when I have found it to workbest is when I feed it really
good examples of my own firstpass of stuff and it makes it
(19:18):
better, better, right, or itfeeds me something I go, I don't
really like this, right, Ithink it works both ways in
different scenarios, but I think, um, like you, I think it's I
don't know that I would say it'ssupercharged me, but it
definitely has improvedcommunication and things like
that.
So, um, I'm I'm bullish on itfor sure, and it's becoming more
(19:42):
and more of a part of my likedaily routine is to use that.
So I want to go back to the,the whole concept of like budget
allocation, though a little bit.
I mean, one of the things yousaid when we talked before was
that it was easier for you tokind of get in, you know,
allocate budget for programs,than it was for marketing ops.
(20:04):
Tell me more about what youmeant by that and why do you
think that was the case and whatwere the results of that.
Ahmed Datoo (20:15):
Yeah, I mean.
So to answer that question, yougot to back up a second to a
higher order question, which islike, in general for marketing
for CMOs, it was always easierto get program spend than it was
to get headcount spend,regardless of department like
whether it was mops or productmarketing, right, and the reason
(20:37):
why is because there's a directcorrelation between programs
that you run and pipeline thatyou generate, and pipeline that
you generate has a directcorrelation to sales, right.
so if you can basically say forevery dollar I spend in program,
I'm able to generate, you know,three dollars or four dollars
in pipeline and we have a 25percent close rate.
(21:00):
You can, you can have a directline to like every dollar you
spend is going to generate Xdollars in revenue.
And it was very easy to makethat case right.
And what people didn'tappreciate is you can give me
and I actually had a CEO ask methis like what if I gave you
unlimited program spend?
Tell me the number, you tell mewhat you need in terms of
(21:25):
program spend and I'll give itto you.
And I was like, even if yougave me an unlimited amount of
program spend, I couldn't use it.
And he said why not?
I'm like, because someone atthe end of the day, has to
create these programs.
Someone has to.
Like LinkedIn doesn't justmanage itself right.
Like someone's got to work withthe designers and work with the
(21:46):
copywriters and then log on toLinkedIn and upload the images.
And then, by the way, whensomeone registers for something
in LinkedIn, we want to do afollow-up email.
So how do I integrate theLinkedIn stuff back into our
marketing automation system andtrigger off an email?
But, by the way, if I triggeroff an email, someone has to
design that email and then ifsomeone responds to that email,
(22:09):
how do I route that lead to theappropriate sales rep?
And I was going on and on andyou see the glazed look in the
CEO.
I can only imagine it wasreally difficult to get
headcounts.
I can only imagine.
And so it was sort of like soit was really difficult to get
headcounts spent, even when Iwent to say can I get product
(22:30):
marketing whatever?
And so you started thinkingabout where can I allocate
resources and what are the morecritical resources?
And so, depending on the stageof the company, marketing
operations unfortunately wasalways the last to get resourced
right.
That team, that poor team, wasalways overworked right and
trying to juggle many differentsets of things and being heroic
(22:52):
right still trying to get beingheroic yeah
absolutely right.
Like I talked to a cmo, um andwho invested in our tools and
and I asked the question, whydid you buy our solution?
He said I wanted to give myteam their Friday nights back
and I was like that is justbrilliant.
(23:13):
He experienced the same thingthat my team experienced, which
is like this team is justdedicated, they want to do a
great job, they're workinginsane hours and they get it
done, but as it doesn't scale,so what ends up happening is, as
you get larger, you startputting in SLAs like the team is
(23:34):
going to be able to turn arounda program.
It started off when you'rereally small, in like a couple
of days, and then, as you getlarger, it grew to a week days
and then, as you get larger, itgrew to a week and then, as you
got even larger, it grew to likesix weeks or eight weeks, right
, and that's how you end upmanaging the resourcing issue
and um and so this was this waskind of the problem with
(23:58):
staffing.
So that's why I triedoffshoring, I tried a variety of
different things.
I I'm like you know, ultimatelyit's just how do we improve
velocity of this team and Ithink, by the way, in any of
these marketing departments youhave the same sets of issues,
like they're all.
In general, marketingdepartments are understaffed
relative to what you see on thesales side, relative to what you
(24:19):
see on the customer supportside.
Right, it's also part of thereason why you see these CMOs
just burn out and, for thatmatter, a lot of marketers just
burn out.
They're understaffed.
But I think, in the realm of AIand this concept of digital
employees that could be membersof the team, where you can start
(24:41):
offloading some of these tasksto members of these team now, is
going to free up marketers tobe able to do the kinds of stuff
that they weren't previouslyable to do.
Right, like all the kinds ofstrategic work, analytical work
that we were trying to draw orcreative work that we were
trying to drive from our team.
Like they're now going to havethe time to be able to do this.
And I think I think theseorganizations that are thinking
(25:03):
about AI around drivingefficiency I know that's the
discussion at the board Like howdo we use AI to drive
efficiency?
It's the wrong way of thinkingabout this.
Like that CMO who said it'sabout giving my team their
Friday night back as a startingpoint and then this realization
that once they have breathingroom, once they have the ability
to step back and think aboutwhat is it that we're really
(25:26):
trying to achieve, because theynow no longer are running 100
miles an hour.
You're going to unlock,assuming, michael, the question
you asked earlier, which is didyou hire the right people?
Assuming you've hired the rightpeople, they're going to be
able to grow and to be able todo things in your organization
(25:48):
that weren't previously beingable to be done right, and
that's really the unlock to meof ai yeah, I've heard someone
call it the time dividend, right, so that that that's the,
that's the real thing that thisai stuff is going to unlock is a
time dividend.
Michael Hartmann (26:05):
So to your point right, which to me is like
it's that thing, like it freesup people to do things that take
advantage of their particularskills and knowledge and
experience and their you know,whatever their gifts are.
Right, that maybe aren't, theyjust haven't had time to do.
Ahmed Datoo (26:21):
That's right.
Michael Hartmann (26:22):
And I think that's that's huge.
I'm curious a little bit.
So all this makes sense aboutbeing able to part of why I
think why program spend iseasier to do also is because
it's visible.
It's very visible.
The result of it is somethingthat's on your website, or it's
an ad, or it's an event orwhatever.
It's very concrete on yourwebsite, or it's an ad, or it's
(26:47):
an event or whatever.
Right, it's very concrete.
Um, when it came down tomarketing ops right, you
mentioned like getting headcountis harder, like do you, did you
find, sort of within themarketing apps domain also, that
it was easier to get budgetapproval for new tech than
people as well?
Ahmed Datoo (27:00):
um no it wasn't it wasn't and, um, you know, for me
, for me, I, I think it this isa lot of this is very cmo
dependent, okay, um, so I am acmo who's technical, so when
folks came in with and so youknow folks would come in all the
(27:23):
time with, can we go off andget this technology?
And I'll be like what's theproblem you're trying to solve?
And invariably I would get techspeak.
Now I understand tech speak, soI understand what they were
asking for.
But you know, part of the jobof the CMO is to mentor.
So I would say, so what?
(27:44):
Two words?
So what?
Right, help me understand.
Why should I care?
As a business person, Iunderstand what you're trying to
do in terms of trying to driveefficiency or trying to unlock X
, y and Z, but how, at the endof the day, is it going to make
(28:06):
life better?
And I have this framework thatI use, which is I call it the
age framework, which is aroundawareness, growth and execution.
Like, these are the three areasthat I cared about, right?
So, when I was thinking about,like when you pitch a technology
, three areas that I cared about, right?
So, when I was thinking about,like, when you pitch a
technology.
Pitch it in one of these threeways to address my so what
question?
(28:27):
Now, awareness is not what youthink it is.
It's not external awarenesslike are you driving brand
awareness, it's actuallyinternal awareness.
Michael Hartmann (28:35):
Oh, interesting.
Ahmed Datoo (28:35):
So part of the job of marketing is to market
yourself and market the team.
So, and whether we like it ornot, marketing is a service
organization.
We service sales, we servicesupport, we service the
executive staff.
We service a lot of internalconstituents as well as
(28:57):
customers, and I know CMOs loveto talk about.
Like you know, it is about thecustomer and 100%.
I'm not disagreeing, but, likeyou, you would be in a world of
hurt if you didn't realize thatyou're a services organization
and you have to address theseconstituents.
So when I'm talking aboutawareness, I'm talking about
does this technology improvemarketing's awareness?
(29:18):
Internally?
So as something as simple asinternally?
So as something as simple asyou know.
Um, you have sales reps who arecomplaining that marketing is
throwing over junk leads.
Like marketing never throwsover good leads, I'm always
getting junk.
The reality is there's like twoleads out of the hundred that
they just got that were junk andthey fixate on the two, not the
(29:41):
the 98 that were good.
Michael Hartmann (29:42):
Right, right.
Ahmed Datoo (29:44):
So if someone came to me and said, hey, I have a
technology don't get into thedetails of what it is which is,
like I can solve this problemaround junk, like I can reduce
the number of junk leadlegitimate junk leads or leads
that are not our persona, likestudents or, you know,
unemployed people or whateverpeople who just simply are not
(30:05):
going to be buyers of ourproduct I can reduce that so
that the signal to noise is much, much better.
The leads that the sales repsare getting are all good.
You know what that sounds likea home run, because that's going
to improve awareness.
All of a sudden, marketing isgenerating good leads.
I'm not seeing crappy leadsfrom marketing.
The reality is we're stillgenerating some crappy leads.
(30:25):
It's just now.
Sales is not seeing them, right, right.
So that's this concept ofawareness.
So if you can pitch awareness,great Growth.
Growth is an easy one, like ifyou can have a direct line to
revenue.
So I am going to go purchasethis particular tool because
it's going to allow me toexecute this type of marketing
campaign that's going togenerate these kinds of leads,
(30:48):
which, ultimately, is going tobuild pipeline.
And I always and I as a CMO, Inever cared about leads.
I thought that was the wrongway of measuring marketing.
I actually thought it was moreimportant to talk about pipeline
, because that allowed us toalign more closely to sales.
So if you came to me and you'relike this tool is going to help
drive a pipeline and it wasbelievable, like I would ask a
(31:11):
couple of questions.
Like you know, is this just thevendor pitch?
Or like, how do you think aboutit actually driving pipeline?
And if it was believable,that's a no brainer, right?
So if you can drive growth withthat tool, spend no brainer.
And the last is around execution.
And the execution is also tiedto internal awareness.
(31:34):
The execution piece is putyourself in the CMO's spot when
she's going to the board andthat board is saying hey, I just
saw our competitor, like theyare at every trade show and and
this, by the way, happened to meum, they are at every trade
show.
Um, I see their ads on everywebsite that I go to.
(31:56):
Um, and their home page isamazing and ours stinks.
Michael Hartmann (32:02):
Yep.
Ahmed Datoo (32:03):
And I went, I did a LinkedIn search and I was like
I looked at the competitor.
It turns out that thatcompetitor had 25 people in
marketing.
We had four yeah.
So how do you out-execute ateam that's got 25 people?
And you've got four?
Yeah, right, got 25 people andyou've got four right.
So if there are tools that aregoing to allow my team to have
(32:24):
leverage to make thatfour-person organization behave
like a 25-person organization,so that I don't get asked that
question in the boardroom likewhy is competitor X doing X, y
and Z, that's worth it to me,right?
So if you can frame in one ofthese three areas, it's a home
run.
Michael Hartmann (32:43):
If you can frame, in one of these three
areas, it's a home run.
Yeah, that's.
I'm seeing you're nodding myhead at every one of those.
I think, the awareness onebeing internal, because that is
not what I expected.
That's a really good insight,because I think most marketers
which is surprising me are notvery good at marketing
(33:03):
themselves internally.
Right, so telling that story,um, just so, yeah, one of the
things I think a lot ofmarketing apps folks would talk
about, and maybe it's tied tothis execution what is it like?
There's a like, a time-saving,efficiency component.
Is that, is that part of yourthought process?
Or like what do you like?
Or is that like a?
Not really what you?
Or like what do you like?
(33:23):
Or is that like a not reallywhat you're trying to get at,
because it feels like it's close, but maybe not really what
you're talking about?
Ahmed Datoo (33:29):
well, to me it is the execution right, but I I
don't necessarily like the wayyou position it to a cmo is not
necessarily time savings.
It's like hey, you know howwe're only able to launch one
campaign every six weeks and ourcompetitors are launching one
every couple of days.
I can now launch one everycouple of days.
Now you're making the sameargument.
(33:50):
Right, like I'm, I'm I'mfreeing up time, productivity,
etc.
Etc.
To allow me to be able to domore with the same amount of
resources.
But I'm just framing itslightly differently, right, in
a language that the cmo caresabout, and then, unfortunately,
like um, it's good, cmos likethe one that I was talking about
previously was like I want togive my team their friday nights
(34:11):
back.
Your people are your assets.
So the cmo um, that is anargument that will resonate with
the cmo around productivity.
But that CMO has to translatethat to a board and to a CEO who
doesn't always care about thewell-being of employees, I'm sad
to say.
(34:31):
Right, they just care about thebottom line.
And so the way I wouldcommunicate it up is like hey,
we're talking about these kindsof constraints.
What if I could improve theproductivity of the team to be
able to generate and the way Iwould talk about of constraints.
What if I could improve theproductivity of the team to be
able to generate?
And the way I would talk about?
It is like if I generate Xnumber of campaigns, you know
I'm generating five morecampaigns a week and I know
(34:51):
every campaign this comes backto the analytics every campaign
I generate generates X dollarsin pipeline.
I should be able to generate anincremental Y dollars in
pipeline.
Michael Hartmann (35:03):
Yeah.
Ahmed Datoo (35:03):
Right, and so now I'm talking in a language that
the board understands and stilltalking about.
At the end of the day, this isproductivity for my team, but
I'm not messaging productivityfor my team when I'm talking
upwards, and so I try toencourage that skill in all my
team members, because at somepoint they're going to rise to
the level of a CMO and they'regoing to have these discussions
(35:26):
at the board and they need to beable to need to communicate
those in the language that isunderstood, and the language at
all levels of an organization isdifferent.
Like yeah the way you explainthings at all languages, at all
levels are going to be differentyou talked to, brought it up a
couple of times.
Michael Hartmann (35:44):
Right, I can, I can do X programs or do
whatever and it's going to bekind of a direct line to
pipeline.
So you know, the on again, offagain debate about attribution
reporting continues to this day.
Right, I'm curious like cause.
Yeah, I haven't heard you talkabout attribution or influence
(36:08):
pipeline revenue.
You talked about pipeline andrevenue, right?
So how do you think aboutattribution versus some other
models for measuring the impactof a marketing team?
Ahmed Datoo (36:19):
Okay, I'll give you a contrarian take.
At least I think it'scontrarian.
I hate attribution.
I don't want to do attribution.
I think attribution is thesingle want to do attribution.
I think attribution is thesingle reason why cmos get fired
.
I also think attribution is thesingle biggest contributor to
um conflict between sales andmarketing, and the reason is
(36:41):
because the moment you starttalking attribution, you start
getting into well, well, wait aminute, did marketing actually
generate that lead?
Like I, as a sales rep, calledthat lead and left voicemails,
like a dozen voicemails, beforethey responded to that marketing
asset.
And then the marketer on myteam would be like, yeah, but
(37:02):
they didn't respond to yours.
But we put out this amazingasset and they were like well,
they wouldn't have learned aboutyou.
And so you have this back andforth.
Michael Hartmann (37:10):
And I was like this is silly, right?
Ahmed Datoo (37:13):
We should not be talking about who's responsible
for generating the lead.
The only thing we should befocused on is like how do we get
that lead into pipeline?
This is why, on my team, myteam felt very uncomfortable
about talking about pipelinebecause they're like we don't
influence pipeline.
What if they don't follow upwith the lead?
And I'm like look, sales, theyare incented on closing business
(37:39):
.
They're not.
They're not lazy as much aswe'd like to think that they're
not following up with leadsbecause they're lazy.
They're not incented to be lazy.
So if you're giving them goodquality leads, they will follow
up.
If you're giving them bad leads, they're not going to follow up
.
So don't fixate on the leads,because if you fixate on any of
(38:01):
the leads, you're going to givethem a lot of junk.
Fixate on the pipeline, theactual high-quality leads, and I
don't care who generates it,right?
I don't care if it wasmarketing-generated, I don't
care if it was sales-generated,I just care do we have enough
pipeline to meet the number forthe quarter?
And that's how I would like tothink.
So I think if I was a CMO and Iwas interviewing, and if the
(38:25):
CEO started asking me questionsaround, how would you measure
marketing attribution?
I would walk, because that, tome, is a CEO who's going to
create this natural tensionbetween sales and marketing
unnecessary tension.
As opposed to, it's a teamsport.
We all have to work inalignment, in concert with each
(38:48):
each other in order to actuallybe able to drive pipeline.
Now I'll get the question likewell then, if you're not
measuring attribution, how doyou know what marketing programs
to run?
How are you figuring out returnon program spend?
And I would actually say youdon't necessarily need
attribution, like this was salesgenerated, this is marketing
(39:10):
generated.
You just need to know like okay, I ran these, it's so simple.
I ran these number of programs,how many of them generated net
new leads and how many of thosenet new leads actually made it
in the pipeline.
I don't care how many touches orwho ended up doing it, but like
are we seeing stuff from ourevents go into pipeline?
I don't care how many touchesor who ended up doing it, but
like are we seeing stuff fromour events go into pipeline and
is it going to pipeline at ahigher rate than some of our
(39:33):
digital campaigns?
And if so, you know what isthat rate?
How many programs do I need tobe running in order to be able
to generate a certain pipelinenumber, right?
So you begin thinking not onlyin terms of leads, but in terms
of conversion rates, and theconversion rates are going to
vary based on channel thatyou're using, and for every
business, the ideal channel isgoing to be different, and so
(39:55):
this notion of a CMO coming inwith a predefined playbook
generally doesn't work right.
So, but this is part of thediscovery, that that gets, gets
to be done.
But if your team is fixated onlike yeah, it was this
particular touch point thatcaused the lead to go from a
lead to an mql to an sql, it'sjust a battle lost yeah, no,
(40:16):
it's, it's my.
Michael Hartmann (40:17):
So I'm curious because my my take on
attribution is not to just throwit out because like, but if I
was to when I talk about it, ifI I think it's been misused I
think it's the short version,right, it's been used to take
through this credit game, um, orto yeah, and I agree with you,
(40:37):
it probably is, at least, if not, the cause is a major part of
why there's friction betweensales and marketing.
But I think there's a place forit still.
I still wouldn't, I would pushreally hard to like don't spend
a lot of money on a tool orresources to do it.
Do one like get as simple asolution as you can, use it
(40:57):
consistently over time so youcan see trends.
But it's used within marketingto help go like what channels,
with what customer you know,target audience or whatever is
working most effectively, andit's a, it's more of a guide.
If you're taking it up to theexecutive team or the board
that's where I think you'restarting to like then you're
losing for sure.
Ahmed Datoo (41:17):
A hundred percent and to your point like that
would be one of the tools.
If my marketing team came andsaid I wanted to go buy an
attribution tool, I would say no, we have UTM parameters that
we're using.
That should be more than enoughto figure out whether a
particular lead came from aparticular campaign that we ran.
I know it's controversial, butit is just like to me.
(41:50):
This is one of these opinionsthat is grounded on just.
Michael Hartmann (41:51):
I think it's bad for culture, and so I've
just seen too many instances ofthis just gone wrong.
It's just not worth it.
Yeah, well, it's funny, though,because I still see regularly
job postings for marketing opsroles that are especially
leadership ones, like to yeah,part of the job is to implement
marketing attribution, part ofit is to do, um, uh, build
marketing dashboards, and, andI'm like those are things that,
to me, I'm like first off, whyare you building attribution,
(42:14):
like what's the, what's the whyand how is it going to be used?
How do you like?
And then, on dashboards, I'mlike, again, not a problem with
reporting, but, starting withthe idea that if you need a
dashboard before you need to,you know, did we get the, the
number of people registered forthis webinar where you're
promoting?
Right, which is a, soundspretty basic, but I think a lot
(42:34):
of organizations don't even dothat.
Yeah, on a consistent basis.
Ahmed Datoo (42:40):
And don't get me wrong, like the analyst I'm not.
I'm I'm a big believer inanalytics.
The dashboards 100 go createthem.
But like the dashboards thatare saying here's the leads
generated by sales versusmarketing, I don't want to see
those in my dashboard.
Right?
Michael Hartmann (42:52):
right, right, all right.
So you you've you've mentionedthis a little bit off and on
right this idea of um, getagents, or ai tools, things like
that, and your company isfocused on that.
How is your perspective on allthis stuff tied with AI and all
(43:13):
that changed since you've kindof moved on from CMO roles?
Ahmed Datoo (43:15):
Yeah, I mean.
So I think I have a great dealof empathy for what the
marketing organizations dealwith, clearly, and so when we
went to start Allgood, it wasthis nagging thing that I had in
my head, which is I was neverappropriately a resource, which
is what we've been spending mostof the time talking about.
So I saw the advent of AI and Ijust thought this is going to
(43:39):
do a huge unlock in a lot oforganizations, and I saw the
influx of tools that were comingin right, and there are a lot
of ai tools that are out thereand I'm going to classify them
as builder tools and, um, youknow, there is this talk now
around going off and hiring agtm engineer and you know to be
(44:02):
the, the owner of all things ai,and and I just think it's a
stupid idea.
And the reason why I say it's astupid idea it's like okay,
let's rewind.
Microsoft Word comes out First.
Word processor right.
(44:23):
First word processor right.
How many organizations weresaying we need to go hire, like
a word engineer or like someonewho just specializes in
Microsoft Word, right?
Michael Hartmann (44:33):
And now, anytime you have word, perfect
or not, before you know.
Ahmed Datoo (44:37):
I mean, the tool should be so easy that everyone
should use it Right, and so so I.
I had this view that AI wasgoing to be something that was
going to be like Microsoft Word,like everyone is going to use
it.
And I get these questions fromCMO like should I hire someone
(44:58):
to be, like, in charge of AI?
And I said absolutely not.
No, that's the wrong way ofthinking about this.
Like everyone in theorganization should be using AI.
I happen to believe that to geteffective use of AI, you're
going to need the appropriatecontext, like providing the
system the appropriate context,so integrating your data systems
, and so there's aninfrastructure layer that you're
(45:20):
going to need to build.
And, you know, shout out toMOPS Like this is the real
opportunity for you to drive theAI initiatives, but it
shouldn't be something that isjust controlled by one group.
It literally should be aninfrastructure that's built that
everyone is able to leverageexpertise.
And the notion of, like me,going off and hiring engineers
like I don't as a marketer Imean a technical marketer, but
(45:53):
no marketer wants to be anengineer, or very few marketers
want to be.
If they wanted to be anengineer, they wouldn't be in
marketing, they'd be an engineer, like the marketers that I
hired.
I hired because they werecreative, they were strategic
right, they were analytical.
Those are the.
Those are the assets that I waslooking for.
That's the reason why peopleget in this business not to like
piece together, code and dothis sort of stuff.
So our view is that you're goingto have these digital employees
(46:17):
that come out of the box, right, that come out of the box that
integrate into systems and dothe work Like you're not
building prompts, you're notdoing the integrations Like.
This is just out of the box.
And the way I think about thisis these digital employees act
(46:38):
just like a regular employee.
So our customers, our digitalemployee, we call her Mary.
So Mary has a.
She gets a Workday account.
She gets an Okta account.
She gets, you know, an Asanaaccount.
She gets a Marketo HubSpotaccount.
She may even get a Salesforceaccount.
She gets onboarded like a realemployee.
(47:00):
You can chat with Mary like youwould chat with a real employee
.
Mary may actually come back toyou with hey, I know you want me
to build this email, but I havequestions around X, y and Z and
if you didn't know any better,you would think that you're
interacting with someone on themarketing operations team, right
(47:20):
?
So you have Mary that justout-of-the-box works.
You start looking at these statswhere, like, 95% of these AI
projects go nowhere, and thereason why they go nowhere is
because it requires so much careand lifting to make them work.
Marketing that can out of thebox work and talk to all the
different systems Now all of asudden so I'll go back to the
(47:51):
beginning of the conversation Idon't necessarily need to hire
someone who is familiar withHubSpot, because you know what
Mary's familiar with HubSpot.
I don't necessarily need to gooff and hire someone who has,
you know, lean data experienceor ring lead experience.
You know why?
Because Mary can do all of thatstuff right.
So what I'm hiring for now ispeople who are going to think
about in the instance of the AIworld, like are they curious?
(48:12):
Are they going to be thinkingabout business process?
How are they going to use Maryto reimagine things right?
Like it goes back to theintelligence and curiosity that
you hire for, not systemsrelated, and it just changes the
way you think about buildingorgs.
Michael Hartmann (48:28):
Yeah, I remember we talked about this
before and I think you may haveeven asked me something like how
would I approach onboarding anAI or a digital employee?
Right, I think it was the firsttime I'd heard that term and I
think I said I think I wouldprobably do it very much like I
would with any.
It very much like I would withany new employee, especially a
(48:48):
more junior person.
And the thing that to me, wouldbe different and this is
probably my style is I like toreally handhold's not really the
right word but really try to bethere as a resource early on
and then, over time, givesomeone more and more freedom to
act independently, and I'd wantto do this like, if it, if it
(49:09):
was a digital employee, I'd wantto do something similar.
What I feel, like what I've, theway I've dealt with some of the
tools, is, when I get tosomething like that, right, I
have to define for it, right,what that threshold is in some,
some way that is, uh, fairlyconcrete, right.
So the way I've done that in thepast is like, if I'm having it
you know, this chat dpt like I'mevaluating something like and I
(49:32):
give it the things I care aboutright, come up with a score.
I don't even tell how to comeup with a score, I just say
these are the things I careabout.
Come up with a score from zero,hundred, right, but only show
me the things that are 70 andabove and it's worked Well.
I don't know what's cutting out, so that's like one of the
mysteries.
I guess I could ask that too,but I think I would do it the
(49:55):
same way.
So, as this digital employee, Igot comfortable.
I was like you're good to gounless your confidence level is
below 80 percent, right, yeah.
And then maybe over time, like,let's move it up to 85 percent
or, sorry, move it down, move itup.
Let's start.
Ahmed Datoo (50:12):
Start with like, if you're 95 and move it down to
85 and so on, right, um, butanyway, well I mean, I, I um,
there's so many interestingthings about that comment, like
I think so a couple thingsMistakes people end up making.
(50:33):
Is they think you're going tobring on AI and it's just going
to do everything magically, likeit just works.
A digital employee would belike you hiring an intern and
not managing them.
Yeah, like someone has got toprovide oversight, right.
And so this is where I comeback to the comment I made about
taste.
Michael Hartmann (50:54):
Yeah.
Ahmed Datoo (50:55):
Mary is going to create stuff and Mary is going
to do stuff, but someone has toreview it to say is it good or
not?
Right, because she will gothrough and she will clean up
data, she will classify thingsshe's going to, she speaks 58
languages, she works 24, sevenLike.
She is going to be your mostfavorite teammate, right?
(51:15):
She's going to do this kind ofstuff that you, you, you don't
get around to doing and she'sgoing to make you look good and
um, but it's only going to workif you have the taste and the
tolerance to be able to givemary feedback, because she's not
always going to get it right.
This is the thing about, youknow, reasoning based systems
versus rules and this view oflike how the future is changing,
(51:38):
like.
The advantage right now iswe're entering a, an age of, and
all of the systems in the pastwere highly deterministic,
meaning you would create a rulein HubSpot, a flow, whatever it
may be, in Marketo, and it waslike do this, do this, do this,
do this.
And it worked really well untilit didn't, and there was a lot
(52:00):
of times where it didn't, and soyou spent a lot of time tuning
in.
Now, with LLMs, you justprovided a problem and you tell
it because it's like the list isendless.
You're never going to get allof that.
(52:36):
Yeah right, in a reasoning-basedworld, you tell an LLM I care
about anyone who's in themarketing ops category, that's
it.
I don't need to give examples.
I don't need to give examples,I don't need to do any of that,
and it's amazingly good at doingthat.
Now it will miss things, so itwill classify something as mops
(53:00):
as an example, when it shouldn'tbe.
And so you end up givingfeedback to Mary like, hey, this
is not mops, this should beclassified as this.
And the amazing thing is itknows that, it stores it and the
next time it sees somethinglike this, it learns from it.
Right, and it learns all thevariations as well.
But you've got to spend thetime you know doing the
(53:20):
providing the feedback and thereview.
Michael Hartmann (53:22):
Yeah, I mean, I've got several ongoing things
with ChatGPT and I do where itmakes sense.
I provide feedback, yeah, andit's done a good job of that.
Yeah, I was smiling when youwere describing that because I
was thinking like I builtsomething using Alteryx years
and years ago and it was just athing that I ran every month,
(53:42):
but it was complicated.
I pulled a bunch of data fromdifferent sources but every
month it would break becausesome new value that I hadn't
anticipated would show up in thesource.
To go fix it Like it was likeyou usually go like, oh, fix it
in this one spot and then rerunit and it was fine, but it was a
constant thing I had to do.
Right, run the flow, it wouldbreak.
(54:03):
Go fix it.
Run the flow, oh, run the flow,it would break.
Go fix it.
Run the flow.
Oh, different thing break, fixit.
Right and it's yeah, I meanthat that by itself like the
ability to to do things like job, job function standardization
or job level standardization,banking excluded.
Right, because everybody's a vp, but yeah, um, it's huge.
(54:26):
So, um, I'm a.
I wish we had more time.
I really do like I feel like wejust scratched the surface of
this, right, invite me back.
Ahmed Datoo (54:35):
I'm happy to come back.
Well, you should come to mobspalooza that's what you need to
do actually, I will be there ohgood, good, good.
Michael Hartmann (54:42):
well, I am still on the fence, so I'm still
trying to figure out a way toget there, but I'm hopeful.
If folks want to connect withyou, continue this conversation
or learn more about what you'redoing, what's the best way for
them to do that?
Ahmed Datoo (54:55):
Yeah, they can go to allgoodhqcom.
That's our website.
I'm on LinkedIn.
Just connect with me onLinkedIn.
Michael Hartmann (55:02):
Awesome, that's terrific.
Well, thank you again for yourtime.
This has been a lot of fun, Ithink.
I think, uh, we're gonna havelots of controversial stuff that
we'll be able to use to promotethis, because I think that's
that's going to be a game.
But no, it was.
It was actually really good.
I think a lot of the ourlisteners um will benefit from
getting your perspective as a,as a CMO, about how marketing
ops is viewed and what you careabout and how you have to then
(55:23):
communicate things.
I think there needs to be moreof that.
So, thank you for that.
Thank you to our supporters andlisteners and, soon, viewers.
We really appreciate you.
Thank you for that.
If you have ideas for guests ortopics or want to be a guest,
you can reach out to Naomi, mikeor me, and we'd be happy to get
the ball rolling Until nexttime.
(55:44):
Bye, everybody.
This is a great question.
My understanding andappreciation of marketing
operations has evolved over theyears.
When I first became a CMO, Iwould classify my relationship
with MOPS as just a hugedisappointment, and the reason
(05:00):
for it is I was like, okay,great, I have a team that can be
my right hand, so to speak.
Like for me, I am a CMO that,um, like, I'm driven by data, I
like to have an understanding ofdata and then that helps,
either, helps influence myintuition, um, but I, I like to
(05:21):
be able to analyze data.
So we run a program.
I want to know what was thesuccess of that program.
I want to be able to analyzedata.
So we run a program.
I want to know what was thesuccess of that program.
I want to know, like, whichsets of marketing activities are
driving pipeline.
So to me, I viewed my marketingoperations team as, like, my
strategic hand, helped me pulltogether the kinds of analysis
that is going to help me makedecisions around allocation of
(05:43):
spend and what programs to berunning, who our icps should be,
etc.
Etc.
So that's my expectation goingin right, and what I found was
the marketing ops team was like,so in the weeds about executing
programs like marketingcampaigns and programs managing
the technology stack, all thatand managing the technology
(06:04):
stack, and like doing managingtechnical debt and doing
integrations, and like a newfeature has been released by a
vendor.
It drove me nuts.
I was like why are we wastingtime on this?
And, and now, remind you, myunderstanding of marketing
operations has evolved so.
So this was my going in justfrustration and I said how
(06:29):
difficult can this be?
Like?
Why does it take so long to dosomething like this?
And so I tend to learn best bydoing so.
Of course, I interact with theMOPS team.
They're like you have noappreciation around how
difficult it is.
I'm like how difficult can itbe to build?
At the time, I think we wereusing Eloqua.
(06:50):
So I was like how difficult canit be to build a campaign in
Eloqua?
Turns out it's really difficult.
And so they would walk methrough about, like you know, it
is very complex and blah, blah,blah.
And so I said you know what I'mgoing to build the campaign.
Can I do the next one?
And I got hands on, I figuredout how.
(07:13):
To you know, I read all theinstructions around building
emails and campaigns.
I was like this doesn't looktoo hard.
And I got the campaign brief.
And of course, the campaignbrief didn't map what was the
email template.
There I got the campaign briefand of course the campaign brief
didn't map what was the emailtemplate.
There are all these variations.
So immediately I'm thinking,okay, we should probably have
some guidelines to the team,like stay within the parameters
of this, but I'm like still, itshouldn't be that difficult to
(07:36):
do.
And then it was just likeendless clicks in the system and
sitting and waiting right,which was something that they
didn't communicate to me, whichis just the poor user experience
of the software and the realityof the software.
And then there's the back andforth between the marketer who
requested, saying don't quitelook like the way you've done
(07:56):
this, can we change this to bold?
And that required me loggingback in the system doing all
these changes.
So I began to have anappreciation, having gone
through that process myself, oflike okay, I get it, like I
understand that they are boggeddown in kind of building out
programs and I understand why.
And, um, I think the otherthing I didn't quite appreciate
(08:20):
is how many different systemsthat they would need to touch to
do something as simple as builda mail campaign.
So it wasn't just Eloqua, itwas like there are a variety of
other systems that they weretouching, and so once he, once I
had an appreciation of this.
Then I had empathy, which is OK, I get it.
Now I understand why they're notable to actually do what it is
(08:44):
that I need to do, and soexperimented with a whole bunch
of stuff, like what if I hiredan offshore team and try to get
rid of the repetitive work?
So we tried doing that and Iwas hoping that would free up
time for my MOPS team to do thestrategic important work that I
was really interested in doing.
And it didn't, primarilybecause the offshore team would
(09:10):
make mistakes and invariably.
You know, in your ideal worldyou have enough prep time for a
campaign.
And I'm guilty of this becauseI would be like, can we go run
this?
Now?
It's the middle of the quarter.
We're not going to meet ourpipeline numbers.
We need to spin up a bunch ofnew campaigns and like overnight
we got to spin stuff up andthere's always a lag when you're
(09:31):
dealing with offshore, sothere's a lag in the initial
creation of it, and then therewas a lag when we needed to do
revisions.
It just slowed the process down.
So I never had really goodexperience offshoring, so I
brought it back onshore.
So the the question around like, um, how, how?
Did I think about mops?
(09:52):
I had this, this appreciationof like this is what I want mops
to do, and, by the way, that'swhat mops wanted to do, like
it's not, like I was suggestingsomething that they viewed was
out of their job description.
They just simply didn't havethe time to do it Right.
And so what I ended up doing isbecause I'm a, you know,
technical CMO, I startedbuilding my own reports.
(10:14):
I was in there, I was, Icreated a data warehouse, I
started building my own sets ofreports and and I realized that
worked for me, but that doesn'twork for most cmos.
Like you know, I'm not going tobe the cmo who's going to be
getting into canva and doingdesign, because that's not where
I spike right.
So, but there are cmos would godo that, but would not go and
(10:36):
like, do the the snowflake dataanalysis type stuff that I was
doing.
So it's just this perpetualproblem that I was facing as a
CMO and this desire to be ableto drive that organization, to
help propel marketing, to beable to do great things, and so
(10:58):
that's something that was alwaysin the back of my head.
Yeah, it's interesting becausethe first thing you said really
is, you know what you wanted wasessentially insights, right?
How are things working?
What are the results?
How can we talk about it withthe rest of the organization?
And I suspect, if my experienceis any indicator, right, that
(11:21):
even if you could have figuredout a way to open up time on
that team, there might not havebeen a skill level to really do
it in a way that would have beenhelpful for you, right, because
I think there's a.
I think there's a.
I mean, I could get this really, really general, having kids
here in the college age.
You know that, like the, thewhole, like uh, math curriculum
(11:44):
for has really not changed in 40or 50 years, right, for kids
coming, and I think that's aproblem.
We get people come out and wejust don't know how to really do
data analysis, right, don'tknow how to put it together,
don't know how to interpret it,don't know how to look at it.
Um, you can't really talk aboutthe difference between a
snapshot in time of somethingversus something that's over
time and all those things thatare different ways of thinking
(12:07):
about it, and that's one of theconcerns and one of the things I
bring up all the time here isjust like we need more people
who can do that, and I'm bullishon things like AI tools being
able to enable some of that,whether it's by freeing up time
and capacity or by helping withdoing some of the heavy lifting
of that.
Whether it's by freeing up timeand capacity or by helping with
doing some of the heavy liftingof data, like organizing data
(12:29):
and doing some initial firstpass analysis.
But do you think the teams youhad before would?
Did you ever get, I guess,first, did you ever get to the
point where they could start todo some of that and did you feel
like they had the skills andcapability of doing that?
so definitely had the skills todo it.
So this this comes to you likehow I hire.
(12:51):
So, generally, when I, when Iwas hiring for anyone in any
role within marketing, it waslike intelligence first and
foremost, some form of curiosityand then some degree of domain
knowledge, and it was the domainknowledge in marketing ops that
always killed me, because Iwould find amazing people, but
(13:14):
they were HubSpot people and wewere a Marketo shop and it's
just too steep of a learningcurve, right.
Or I would find incrediblycurious, incredibly smart,
talented people, but they were.
You know, they spiked on acertain set of technologies that
didn't exist in our stack andthat's the critical component of
(13:36):
it, right.
And so you end up making thesecompromises to get in people who
have domain knowledge about thestack, and it's kind of crazy
that we make decisions.
We need to make decisions basedon familiarity with technology
as opposed to just smarts.
But I was fortunate.
(13:57):
The people that I ended uphiring were really good.
Uh, I mean, had experience withthe tech stack, but were really
smart.
So, yeah, I felt like if, ifthey, if given the time and
given a problem, they would gofigure it out.
Right, they may not necessarilyknow data analysis, but I could
probably work with them to say,like, here are the kinds of
reports that I'm looking for,and then be able to go figure it
(14:18):
out.
This is pre-AI right In an AIworld.
This is what's amazing about AI.
Like, I talk about myself andyou know I like colors.
When given something, I canreview it and tell you whether I
like it or not.
I don't necessarily feel likeI've got great design skill, in
(14:41):
fact, um, you know, Ideliberately stay away from.
I told my wife, like you canhandle all the decorations of
the house.
I won't have any sort of review.
The only thing I ask is that Iget to handle all the tech
gadgets like that.
That's going to be my, my,purview, right?
Michael Hartmann (14:58):
um, so I have my request.
My request was just only vetopower on like if really didn't
like something from a designlike I really want to be able
toanche on electronics.
Ahmed Datoo (15:08):
But um, you know, when it came to, when it came to
design, um, I would it's verydifficult, like I'd sit down
(15:33):
with a designer and the designerrightfully so would ask me a
question.
Like you know, give me yourdesign aesthetic.
And early in my my career I waslike I don't know what my
design aesthetic is, I'll knowit when I see it, which is like
the worst thing.
A designer hears Like um, howdo you work with that?
And so early on it'd be likethey'd give me something and I
(15:55):
would just tear it apart andthey're like, okay, you clearly
don't like that design.
Give me a couple of designsthat you do like, and so I'd
start talking about.
You know, here are vendors thatI like.
And so I began to realize Ibegan to speak the language of
the designer.
Michael Hartmann (16:07):
Right.
Ahmed Datoo (16:09):
Fast forward to today.
Now, all of a sudden, I can bea designer not a great one, but
I can iterate on that designprocess a lot faster.
So I can go into mid journeyright now, and do you know, to
mid journey right now.
And do you know, image creationaround concepts.
I can go into all these codingtools and actually create first
(16:30):
iterations of a homepage design,right To get to a point where
it's like, yeah, this is in thegeneral direction of where I
want to go, and then I can handit off to a professional to
either make it better or atleast have a starting point to
say, like this is kind of theaesthetic of what I'm looking
for, right and so.
And, by the way, I don't.
There are certain things whereI want to go to a designer
(16:52):
because, like the home page, Iwant to nail right.
But if I'm doing a landing pagefor a webinar that we're doing
next week, I don't necessarilyneed to go to my agency to nail
that right.
I could probably use the tool,the AI tools, myself and build
something comparable, because Ialready know what my brand voice
is, the style, whatever.
(17:12):
I can use those tools myself.
That was not possible at allright.
And so your point about peoplewho are not analytical or
quantitative.
Now, all of a sudden, you knowI started the conversation by
saying the CMO job andmarketing's job in general is
tough because you have to spikein all these different things.
I view this concept of now, asan employee, you will have
(17:34):
access to digital employees anda set of tools that allow you to
spike in each of thosecategories.
You'll be able to do designwork.
You can now write amazing copy.
You can do analytical work.
You can actually do operationsoriented work.
You can do strategic orientedwork.
Right like.
All of those are possible, um,and I think what's what's
(17:56):
required in that, in that sortof world, is taste.
So.
So, when you use these tools,you got to know what good looks
like.
Michael Hartmann (18:10):
Yes.
Ahmed Datoo (18:11):
Right.
So, like I know, when I getoutput from the AI tools for
something like an image, or evenwhen I'm working with copy
because I was a former productmarketer and I worked with
designers right Like, I havegotten an understanding of what
is considered good from amessaging standpoint, what is
(18:33):
considered good or good enoughfrom a design standpoint right,
and so I think anyone who hasexperience right now in
marketing can be superchargedwith the use of these tools yeah
, I'm finding the same, if I forthose who are listening, right,
I had a weird look on my facewhen he said that you have that,
what was the word you used?
Michael Hartmann (18:54):
um, taste, you have to have taste.
And I was like, what does hemean by that?
And when you describe, like youknowing what was what good is,
which is also subjective, rightin a lot of these cases.
But, like, for me, one of thethings I've learned I have to do
when I'm using, just keep itdown to LLMs, right, it's like I
, when I have found it to workbest is when I feed it really
good examples of my own firstpass of stuff and it makes it
(19:18):
better, better, right, or itfeeds me something I go, I don't
really like this, right, Ithink it works both ways in
different scenarios, but I think, um, like you, I think it's I
don't know that I would say it'ssupercharged me, but it
definitely has improvedcommunication and things like
that.
So, um, I'm I'm bullish on itfor sure, and it's becoming more
(19:42):
and more of a part of my likedaily routine is to use that.
So I want to go back to the,the whole concept of like budget
allocation, though a little bit.
I mean, one of the things yousaid when we talked before was
that it was easier for you tokind of get in, you know,
allocate budget for programs,than it was for marketing ops.
(20:04):
Tell me more about what youmeant by that and why do you
think that was the case and whatwere the results of that.
Ahmed Datoo (20:15):
Yeah, I mean.
So to answer that question, yougot to back up a second to a
higher order question, which islike, in general for marketing
for CMOs, it was always easierto get program spend than it was
to get headcount spend,regardless of department like
whether it was mops or productmarketing, right, and the reason
(20:37):
why is because there's a directcorrelation between programs
that you run and pipeline thatyou generate, and pipeline that
you generate has a directcorrelation to sales, right.
so if you can basically say forevery dollar I spend in program,
I'm able to generate, you know,three dollars or four dollars
in pipeline and we have a 25percent close rate.
(21:00):
You can, you can have a directline to like every dollar you
spend is going to generate Xdollars in revenue.
And it was very easy to makethat case right.
And what people didn'tappreciate is you can give me
and I actually had a CEO ask methis like what if I gave you
unlimited program spend?
Tell me the number, you tell mewhat you need in terms of
(21:25):
program spend and I'll give itto you.
And I was like, even if yougave me an unlimited amount of
program spend, I couldn't use it.
And he said why not?
I'm like, because someone atthe end of the day, has to
create these programs.
Someone has to.
Like LinkedIn doesn't justmanage itself right.
Like someone's got to work withthe designers and work with the
(21:46):
copywriters and then log on toLinkedIn and upload the images.
And then, by the way, whensomeone registers for something
in LinkedIn, we want to do afollow-up email.
So how do I integrate theLinkedIn stuff back into our
marketing automation system andtrigger off an email?
But, by the way, if I triggeroff an email, someone has to
design that email and then ifsomeone responds to that email,
(22:09):
how do I route that lead to theappropriate sales rep?
And I was going on and on andyou see the glazed look in the
CEO.
I can only imagine it wasreally difficult to get
headcounts.
I can only imagine.
And so it was sort of like soit was really difficult to get
headcounts spent, even when Iwent to say can I get product
(22:30):
marketing whatever?
And so you started thinkingabout where can I allocate
resources and what are the morecritical resources?
And so, depending on the stageof the company, marketing
operations unfortunately wasalways the last to get resourced
right.
That team, that poor team, wasalways overworked right and
trying to juggle many differentsets of things and being heroic
(22:52):
right still trying to get beingheroic yeah
absolutely right.
Like I talked to a cmo, um andwho invested in our tools and
and I asked the question, whydid you buy our solution?
He said I wanted to give myteam their Friday nights back
and I was like that is justbrilliant.
(23:13):
He experienced the same thingthat my team experienced, which
is like this team is justdedicated, they want to do a
great job, they're workinginsane hours and they get it
done, but as it doesn't scale,so what ends up happening is, as
you get larger, you startputting in SLAs like the team is
(23:34):
going to be able to turn arounda program.
It started off when you'rereally small, in like a couple
of days, and then, as you getlarger, it grew to a week days
and then, as you get larger, itgrew to a week and then, as you
got even larger, it grew to likesix weeks or eight weeks, right
, and that's how you end upmanaging the resourcing issue
and um and so this was this waskind of the problem with
(23:58):
staffing.
So that's why I triedoffshoring, I tried a variety of
different things.
I I'm like you know, ultimatelyit's just how do we improve
velocity of this team and Ithink, by the way, in any of
these marketing departments youhave the same sets of issues,
like they're all.
In general, marketingdepartments are understaffed
relative to what you see on thesales side, relative to what you
(24:19):
see on the customer supportside.
Right, it's also part of thereason why you see these CMOs
just burn out and, for thatmatter, a lot of marketers just
burn out.
They're understaffed.
But I think, in the realm of AIand this concept of digital
employees that could be membersof the team, where you can start
(24:41):
offloading some of these tasksto members of these team now, is
going to free up marketers tobe able to do the kinds of stuff
that they weren't previouslyable to do.
Right, like all the kinds ofstrategic work, analytical work
that we were trying to draw orcreative work that we were
trying to drive from our team.
Like they're now going to havethe time to be able to do this.
And I think I think theseorganizations that are thinking
(25:03):
about AI around drivingefficiency I know that's the
discussion at the board Like howdo we use AI to drive
efficiency?
It's the wrong way of thinkingabout this.
Like that CMO who said it'sabout giving my team their
Friday night back as a startingpoint and then this realization
that once they have breathingroom, once they have the ability
to step back and think aboutwhat is it that we're really
(25:26):
trying to achieve, because theynow no longer are running 100
miles an hour.
You're going to unlock,assuming, michael, the question
you asked earlier, which is didyou hire the right people?
Assuming you've hired the rightpeople, they're going to be
able to grow and to be able todo things in your organization
(25:48):
that weren't previously beingable to be done right, and
that's really the unlock to meof ai yeah, I've heard someone
call it the time dividend, right, so that that that's the,
that's the real thing that thisai stuff is going to unlock is a
time dividend.
Michael Hartmann (26:05):
So to your point right, which to me is like
it's that thing, like it freesup people to do things that take
advantage of their particularskills and knowledge and
experience and their you know,whatever their gifts are.
Right, that maybe aren't, theyjust haven't had time to do.
Ahmed Datoo (26:21):
That's right.
Michael Hartmann (26:22):
And I think that's that's huge.
I'm curious a little bit.
So all this makes sense aboutbeing able to part of why I
think why program spend iseasier to do also is because
it's visible.
It's very visible.
The result of it is somethingthat's on your website, or it's
an ad, or it's an event orwhatever.
It's very concrete on yourwebsite, or it's an ad, or it's
(26:47):
an event or whatever.
Right, it's very concrete.
Um, when it came down tomarketing ops right, you
mentioned like getting headcountis harder, like do you, did you
find, sort of within themarketing apps domain also, that
it was easier to get budgetapproval for new tech than
people as well?
Ahmed Datoo (27:00):
um no it wasn't it wasn't and, um, you know, for me
, for me, I, I think it this isa lot of this is very cmo
dependent, okay, um, so I am acmo who's technical, so when
folks came in with and so youknow folks would come in all the
(27:23):
time with, can we go off andget this technology?
And I'll be like what's theproblem you're trying to solve?
And invariably I would get techspeak.
Now I understand tech speak, soI understand what they were
asking for.
But you know, part of the jobof the CMO is to mentor.
So I would say, so what?
(27:44):
Two words?
So what?
Right, help me understand.
Why should I care?
As a business person, Iunderstand what you're trying to
do in terms of trying to driveefficiency or trying to unlock X
, y and Z, but how, at the endof the day, is it going to make
(28:06):
life better?
And I have this framework thatI use, which is I call it the
age framework, which is aroundawareness, growth and execution.
Like, these are the three areasthat I cared about, right?
So, when I was thinking about,like when you pitch a technology
, three areas that I cared about, right?
So, when I was thinking about,like, when you pitch a
technology.
Pitch it in one of these threeways to address my so what
question?
(28:27):
Now, awareness is not what youthink it is.
It's not external awarenesslike are you driving brand
awareness, it's actuallyinternal awareness.
Michael Hartmann (28:35):
Oh, interesting.
Ahmed Datoo (28:35):
So part of the job of marketing is to market
yourself and market the team.
So, and whether we like it ornot, marketing is a service
organization.
We service sales, we servicesupport, we service the
executive staff.
We service a lot of internalconstituents as well as
(28:57):
customers, and I know CMOs loveto talk about.
Like you know, it is about thecustomer and 100%.
I'm not disagreeing, but, likeyou, you would be in a world of
hurt if you didn't realize thatyou're a services organization
and you have to address theseconstituents.
So when I'm talking aboutawareness, I'm talking about
does this technology improvemarketing's awareness?
(29:18):
Internally?
So as something as simple asinternally?
So as something as simple asyou know.
Um, you have sales reps who arecomplaining that marketing is
throwing over junk leads.
Like marketing never throwsover good leads, I'm always
getting junk.
The reality is there's like twoleads out of the hundred that
they just got that were junk andthey fixate on the two, not the
(29:41):
the 98 that were good.
Michael Hartmann (29:42):
Right, right.
Ahmed Datoo (29:44):
So if someone came to me and said, hey, I have a
technology don't get into thedetails of what it is which is,
like I can solve this problemaround junk, like I can reduce
the number of junk leadlegitimate junk leads or leads
that are not our persona, likestudents or, you know,
unemployed people or whateverpeople who just simply are not
(30:05):
going to be buyers of ourproduct I can reduce that so
that the signal to noise is much, much better.
The leads that the sales repsare getting are all good.
You know what that sounds likea home run, because that's going
to improve awareness.
All of a sudden, marketing isgenerating good leads.
I'm not seeing crappy leadsfrom marketing.
The reality is we're stillgenerating some crappy leads.
(30:25):
It's just now.
Sales is not seeing them, right, right.
So that's this concept ofawareness.
So if you can pitch awareness,great Growth.
Growth is an easy one, like ifyou can have a direct line to
revenue.
So I am going to go purchasethis particular tool because
it's going to allow me toexecute this type of marketing
campaign that's going togenerate these kinds of leads,
(30:48):
which, ultimately, is going tobuild pipeline.
And I always and I as a CMO, Inever cared about leads.
I thought that was the wrongway of measuring marketing.
I actually thought it was moreimportant to talk about pipeline
, because that allowed us toalign more closely to sales.
So if you came to me and you'relike this tool is going to help
drive a pipeline and it wasbelievable, like I would ask a
(31:11):
couple of questions.
Like you know, is this just thevendor pitch?
Or like, how do you think aboutit actually driving pipeline?
And if it was believable,that's a no brainer, right?
So if you can drive growth withthat tool, spend no brainer.
And the last is around execution.
And the execution is also tiedto internal awareness.
(31:34):
The execution piece is putyourself in the CMO's spot when
she's going to the board andthat board is saying hey, I just
saw our competitor, like theyare at every trade show and and
this, by the way, happened to meum, they are at every trade
show.
Um, I see their ads on everywebsite that I go to.
(31:56):
Um, and their home page isamazing and ours stinks.
Michael Hartmann (32:02):
Yep.
Ahmed Datoo (32:03):
And I went, I did a LinkedIn search and I was like
I looked at the competitor.
It turns out that thatcompetitor had 25 people in
marketing.
We had four yeah.
So how do you out-execute ateam that's got 25 people?
And you've got four?
Yeah, right, got 25 people andyou've got four right.
So if there are tools that aregoing to allow my team to have
(32:24):
leverage to make thatfour-person organization behave
like a 25-person organization,so that I don't get asked that
question in the boardroom likewhy is competitor X doing X, y
and Z, that's worth it to me,right?
So if you can frame in one ofthese three areas, it's a home
run.
Michael Hartmann (32:43):
If you can frame, in one of these three
areas, it's a home run.
Yeah, that's.
I'm seeing you're nodding myhead at every one of those.
I think, the awareness onebeing internal, because that is
not what I expected.
That's a really good insight,because I think most marketers
which is surprising me are notvery good at marketing
(33:03):
themselves internally.
Right, so telling that story,um, just so, yeah, one of the
things I think a lot ofmarketing apps folks would talk
about, and maybe it's tied tothis execution what is it like?
There's a like, a time-saving,efficiency component.
Is that, is that part of yourthought process?
Or like what do you like?
Or is that like a?
Not really what you?
Or like what do you like?
(33:23):
Or is that like a not reallywhat you're trying to get at,
because it feels like it's close, but maybe not really what
you're talking about?
Ahmed Datoo (33:29):
well, to me it is the execution right, but I I
don't necessarily like the wayyou position it to a cmo is not
necessarily time savings.
It's like hey, you know howwe're only able to launch one
campaign every six weeks and ourcompetitors are launching one
every couple of days.
I can now launch one everycouple of days.
Now you're making the sameargument.
(33:50):
Right, like I'm, I'm I'mfreeing up time, productivity,
etc.
Etc.
To allow me to be able to domore with the same amount of
resources.
But I'm just framing itslightly differently, right, in
a language that the cmo caresabout, and then, unfortunately,
like um, it's good, cmos likethe one that I was talking about
previously was like I want togive my team their friday nights
(34:11):
back.
Your people are your assets.
So the cmo um, that is anargument that will resonate with
the cmo around productivity.
But that CMO has to translatethat to a board and to a CEO who
doesn't always care about thewell-being of employees, I'm sad
to say.
(34:31):
Right, they just care about thebottom line.
And so the way I wouldcommunicate it up is like hey,
we're talking about these kindsof constraints.
What if I could improve theproductivity of the team to be
able to generate and the way Iwould talk about of constraints.
What if I could improve theproductivity of the team to be
able to generate?
And the way I would talk about?
It is like if I generate Xnumber of campaigns, you know
I'm generating five morecampaigns a week and I know
(34:51):
every campaign this comes backto the analytics every campaign
I generate generates X dollarsin pipeline.
I should be able to generate anincremental Y dollars in
pipeline.
Michael Hartmann (35:03):
Yeah.
Ahmed Datoo (35:03):
Right, and so now I'm talking in a language that
the board understands and stilltalking about.
At the end of the day, this isproductivity for my team, but
I'm not messaging productivityfor my team when I'm talking
upwards, and so I try toencourage that skill in all my
team members, because at somepoint they're going to rise to
the level of a CMO and they'regoing to have these discussions
(35:26):
at the board and they need to beable to need to communicate
those in the language that isunderstood, and the language at
all levels of an organization isdifferent.
Like yeah the way you explainthings at all languages, at all
levels are going to be differentyou talked to, brought it up a
couple of times.
Michael Hartmann (35:44):
Right, I can, I can do X programs or do
whatever and it's going to bekind of a direct line to
pipeline.
So you know, the on again, offagain debate about attribution
reporting continues to this day.
Right, I'm curious like cause.
Yeah, I haven't heard you talkabout attribution or influence
(36:08):
pipeline revenue.
You talked about pipeline andrevenue, right?
So how do you think aboutattribution versus some other
models for measuring the impactof a marketing team?
Ahmed Datoo (36:19):
Okay, I'll give you a contrarian take.
At least I think it'scontrarian.
I hate attribution.
I don't want to do attribution.
I think attribution is thesingle want to do attribution.
I think attribution is thesingle reason why cmos get fired
.
I also think attribution is thesingle biggest contributor to
um conflict between sales andmarketing, and the reason is
(36:41):
because the moment you starttalking attribution, you start
getting into well, well, wait aminute, did marketing actually
generate that lead?
Like I, as a sales rep, calledthat lead and left voicemails,
like a dozen voicemails, beforethey responded to that marketing
asset.
And then the marketer on myteam would be like, yeah, but
(37:02):
they didn't respond to yours.
But we put out this amazingasset and they were like well,
they wouldn't have learned aboutyou.
And so you have this back andforth.
Michael Hartmann (37:10):
And I was like this is silly, right?
Ahmed Datoo (37:13):
We should not be talking about who's responsible
for generating the lead.
The only thing we should befocused on is like how do we get
that lead into pipeline?
This is why, on my team, myteam felt very uncomfortable
about talking about pipelinebecause they're like we don't
influence pipeline.
What if they don't follow upwith the lead?
And I'm like look, sales, theyare incented on closing business
(37:39):
.
They're not.
They're not lazy as much aswe'd like to think that they're
not following up with leadsbecause they're lazy.
They're not incented to be lazy.
So if you're giving them goodquality leads, they will follow
up.
If you're giving them bad leads, they're not going to follow up
.
So don't fixate on the leads,because if you fixate on any of
(38:01):
the leads, you're going to givethem a lot of junk.
Fixate on the pipeline, theactual high-quality leads, and I
don't care who generates it,right?
I don't care if it wasmarketing-generated, I don't
care if it was sales-generated,I just care do we have enough
pipeline to meet the number forthe quarter?
And that's how I would like tothink.
So I think if I was a CMO and Iwas interviewing, and if the
(38:25):
CEO started asking me questionsaround, how would you measure
marketing attribution?
I would walk, because that, tome, is a CEO who's going to
create this natural tensionbetween sales and marketing
unnecessary tension.
As opposed to, it's a teamsport.
We all have to work inalignment, in concert with each
(38:48):
each other in order to actuallybe able to drive pipeline.
Now I'll get the question likewell then, if you're not
measuring attribution, how doyou know what marketing programs
to run?
How are you figuring out returnon program spend?
And I would actually say youdon't necessarily need
attribution, like this was salesgenerated, this is marketing
(39:10):
generated.
You just need to know like okay, I ran these, it's so simple.
I ran these number of programs,how many of them generated net
new leads and how many of thosenet new leads actually made it
in the pipeline.
I don't care how many touches orwho ended up doing it, but like
are we seeing stuff from ourevents go into pipeline?
I don't care how many touchesor who ended up doing it, but
like are we seeing stuff fromour events go into pipeline and
is it going to pipeline at ahigher rate than some of our
(39:33):
digital campaigns?
And if so, you know what isthat rate?
How many programs do I need tobe running in order to be able
to generate a certain pipelinenumber, right?
So you begin thinking not onlyin terms of leads, but in terms
of conversion rates, and theconversion rates are going to
vary based on channel thatyou're using, and for every
business, the ideal channel isgoing to be different, and so
(39:55):
this notion of a CMO coming inwith a predefined playbook
generally doesn't work right.
So, but this is part of thediscovery, that that gets, gets
to be done.
But if your team is fixated onlike yeah, it was this
particular touch point thatcaused the lead to go from a
lead to an mql to an sql, it'sjust a battle lost yeah, no,
(40:16):
it's, it's my.
Michael Hartmann (40:17):
So I'm curious because my my take on
attribution is not to just throwit out because like, but if I
was to when I talk about it, ifI I think it's been misused I
think it's the short version,right, it's been used to take
through this credit game, um, orto yeah, and I agree with you,
(40:37):
it probably is, at least, if not, the cause is a major part of
why there's friction betweensales and marketing.
But I think there's a place forit still.
I still wouldn't, I would pushreally hard to like don't spend
a lot of money on a tool orresources to do it.
Do one like get as simple asolution as you can, use it
(40:57):
consistently over time so youcan see trends.
But it's used within marketingto help go like what channels,
with what customer you know,target audience or whatever is
working most effectively, andit's a, it's more of a guide.
If you're taking it up to theexecutive team or the board
that's where I think you'restarting to like then you're
losing for sure.
Ahmed Datoo (41:17):
A hundred percent and to your point like that
would be one of the tools.
If my marketing team came andsaid I wanted to go buy an
attribution tool, I would say no, we have UTM parameters that
we're using.
That should be more than enoughto figure out whether a
particular lead came from aparticular campaign that we ran.
I know it's controversial, butit is just like to me.
(41:50):
This is one of these opinionsthat is grounded on just.
Michael Hartmann (41:51):
I think it's bad for culture, and so I've
just seen too many instances ofthis just gone wrong.
It's just not worth it.
Yeah, well, it's funny, though,because I still see regularly
job postings for marketing opsroles that are especially
leadership ones, like to yeah,part of the job is to implement
marketing attribution, part ofit is to do, um, uh, build
marketing dashboards, and, andI'm like those are things that,
to me, I'm like first off, whyare you building attribution,
(42:14):
like what's the, what's the whyand how is it going to be used?
How do you like?
And then, on dashboards, I'mlike, again, not a problem with
reporting, but, starting withthe idea that if you need a
dashboard before you need to,you know, did we get the, the
number of people registered forthis webinar where you're
promoting?
Right, which is a, soundspretty basic, but I think a lot
(42:34):
of organizations don't even dothat.
Yeah, on a consistent basis.
Ahmed Datoo (42:40):
And don't get me wrong, like the analyst I'm not.
I'm I'm a big believer inanalytics.
The dashboards 100 go createthem.
But like the dashboards thatare saying here's the leads
generated by sales versusmarketing, I don't want to see
those in my dashboard.
Right?
Michael Hartmann (42:52):
right, right, all right.
So you you've you've mentionedthis a little bit off and on
right this idea of um, getagents, or ai tools, things like
that, and your company isfocused on that.
How is your perspective on allthis stuff tied with AI and all
(43:13):
that changed since you've kindof moved on from CMO roles?
Ahmed Datoo (43:15):
Yeah, I mean.
So I think I have a great dealof empathy for what the
marketing organizations dealwith, clearly, and so when we
went to start Allgood, it wasthis nagging thing that I had in
my head, which is I was neverappropriately a resource, which
is what we've been spending mostof the time talking about.
So I saw the advent of AI and Ijust thought this is going to
(43:39):
do a huge unlock in a lot oforganizations, and I saw the
influx of tools that were comingin right, and there are a lot
of ai tools that are out thereand I'm going to classify them
as builder tools and, um, youknow, there is this talk now
around going off and hiring agtm engineer and you know to be
(44:02):
the, the owner of all things ai,and and I just think it's a
stupid idea.
And the reason why I say it's astupid idea it's like okay,
let's rewind.
Microsoft Word comes out First.
Word processor right.
(44:23):
First word processor right.
How many organizations weresaying we need to go hire, like
a word engineer or like someonewho just specializes in
Microsoft Word, right?
Michael Hartmann (44:33):
And now, anytime you have word, perfect
or not, before you know.
Ahmed Datoo (44:37):
I mean, the tool should be so easy that everyone
should use it Right, and so so I.
I had this view that AI wasgoing to be something that was
going to be like Microsoft Word,like everyone is going to use
it.
And I get these questions fromCMO like should I hire someone
(44:58):
to be, like, in charge of AI?
And I said absolutely not.
No, that's the wrong way ofthinking about this.
Like everyone in theorganization should be using AI.
I happen to believe that to geteffective use of AI, you're
going to need the appropriatecontext, like providing the
system the appropriate context,so integrating your data systems
, and so there's aninfrastructure layer that you're
(45:20):
going to need to build.
And, you know, shout out toMOPS Like this is the real
opportunity for you to drive theAI initiatives, but it
shouldn't be something that isjust controlled by one group.
It literally should be aninfrastructure that's built that
everyone is able to leverageexpertise.
And the notion of, like me,going off and hiring engineers
like I don't as a marketer Imean a technical marketer, but
(45:53):
no marketer wants to be anengineer, or very few marketers
want to be.
If they wanted to be anengineer, they wouldn't be in
marketing, they'd be an engineer, like the marketers that I
hired.
I hired because they werecreative, they were strategic
right, they were analytical.
Those are the.
Those are the assets that I waslooking for.
That's the reason why peopleget in this business not to like
piece together, code and dothis sort of stuff.
So our view is that you're goingto have these digital employees
(46:17):
that come out of the box, right, that come out of the box that
integrate into systems and dothe work Like you're not
building prompts, you're notdoing the integrations Like.
This is just out of the box.
And the way I think about thisis these digital employees act
(46:38):
just like a regular employee.
So our customers, our digitalemployee, we call her Mary.
So Mary has a.
She gets a Workday account.
She gets an Okta account.
She gets, you know, an Asanaaccount.
She gets a Marketo HubSpotaccount.
She may even get a Salesforceaccount.
She gets onboarded like a realemployee.
(47:00):
You can chat with Mary like youwould chat with a real employee
.
Mary may actually come back toyou with hey, I know you want me
to build this email, but I havequestions around X, y and Z and
if you didn't know any better,you would think that you're
interacting with someone on themarketing operations team, right
(47:20):
?
So you have Mary that justout-of-the-box works.
You start looking at these statswhere, like, 95% of these AI
projects go nowhere, and thereason why they go nowhere is
because it requires so much careand lifting to make them work.
Marketing that can out of thebox work and talk to all the
different systems Now all of asudden so I'll go back to the
(47:51):
beginning of the conversation Idon't necessarily need to hire
someone who is familiar withHubSpot, because you know what
Mary's familiar with HubSpot.
I don't necessarily need to gooff and hire someone who has,
you know, lean data experienceor ring lead experience.
You know why?
Because Mary can do all of thatstuff right.
So what I'm hiring for now ispeople who are going to think
about in the instance of the AIworld, like are they curious?
(48:12):
Are they going to be thinkingabout business process?
How are they going to use Maryto reimagine things right?
Like it goes back to theintelligence and curiosity that
you hire for, not systemsrelated, and it just changes the
way you think about buildingorgs.
Michael Hartmann (48:28):
Yeah, I remember we talked about this
before and I think you may haveeven asked me something like how
would I approach onboarding anAI or a digital employee?
Right, I think it was the firsttime I'd heard that term and I
think I said I think I wouldprobably do it very much like I
would with any.
It very much like I would withany new employee, especially a
(48:48):
more junior person.
And the thing that to me, wouldbe different and this is
probably my style is I like toreally handhold's not really the
right word but really try to bethere as a resource early on
and then, over time, givesomeone more and more freedom to
act independently, and I'd wantto do this like, if it, if it
(49:09):
was a digital employee, I'd wantto do something similar.
What I feel, like what I've, theway I've dealt with some of the
tools, is, when I get tosomething like that, right, I
have to define for it, right,what that threshold is in some,
some way that is, uh, fairlyconcrete, right.
So the way I've done that in thepast is like, if I'm having it
you know, this chat dpt like I'mevaluating something like and I
(49:32):
give it the things I care aboutright, come up with a score.
I don't even tell how to comeup with a score, I just say
these are the things I careabout.
Come up with a score from zero,hundred, right, but only show
me the things that are 70 andabove and it's worked Well.
I don't know what's cutting out, so that's like one of the
mysteries.
I guess I could ask that too,but I think I would do it the
(49:55):
same way.
So, as this digital employee, Igot comfortable.
I was like you're good to gounless your confidence level is
below 80 percent, right, yeah.
And then maybe over time, like,let's move it up to 85 percent
or, sorry, move it down, move itup.
Let's start.
Ahmed Datoo (50:12):
Start with like, if you're 95 and move it down to
85 and so on, right, um, butanyway, well I mean, I, I um,
there's so many interestingthings about that comment, like
I think so a couple thingsMistakes people end up making.
(50:33):
Is they think you're going tobring on AI and it's just going
to do everything magically, likeit just works.
A digital employee would belike you hiring an intern and
not managing them.
Yeah, like someone has got toprovide oversight, right.
And so this is where I comeback to the comment I made about
taste.
Michael Hartmann (50:54):
Yeah.
Ahmed Datoo (50:55):
Mary is going to create stuff and Mary is going
to do stuff, but someone has toreview it to say is it good or
not?
Right, because she will gothrough and she will clean up
data, she will classify thingsshe's going to, she speaks 58
languages, she works 24, sevenLike.
She is going to be your mostfavorite teammate, right?
(51:15):
She's going to do this kind ofstuff that you, you, you don't
get around to doing and she'sgoing to make you look good and
um, but it's only going to workif you have the taste and the
tolerance to be able to givemary feedback, because she's not
always going to get it right.
This is the thing about, youknow, reasoning based systems
versus rules and this view oflike how the future is changing,
(51:38):
like.
The advantage right now iswe're entering a, an age of, and
all of the systems in the pastwere highly deterministic,
meaning you would create a rulein HubSpot, a flow, whatever it
may be, in Marketo, and it waslike do this, do this, do this,
do this.
And it worked really well untilit didn't, and there was a lot
(52:00):
of times where it didn't, and soyou spent a lot of time tuning
in.
Now, with LLMs, you justprovided a problem and you tell
it because it's like the list isendless.
You're never going to get allof that.
(52:36):
Yeah right, in a reasoning-basedworld, you tell an LLM I care
about anyone who's in themarketing ops category, that's
it.
I don't need to give examples.
I don't need to give examples,I don't need to do any of that,
and it's amazingly good at doingthat.
Now it will miss things, so itwill classify something as mops
(53:00):
as an example, when it shouldn'tbe.
And so you end up givingfeedback to Mary like, hey, this
is not mops, this should beclassified as this.
And the amazing thing is itknows that, it stores it and the
next time it sees somethinglike this, it learns from it.
Right, and it learns all thevariations as well.
But you've got to spend thetime you know doing the
(53:20):
providing the feedback and thereview.
Michael Hartmann (53:22):
Yeah, I mean, I've got several ongoing things
with ChatGPT and I do where itmakes sense.
I provide feedback, yeah, andit's done a good job of that.
Yeah, I was smiling when youwere describing that because I
was thinking like I builtsomething using Alteryx years
and years ago and it was just athing that I ran every month,
(53:42):
but it was complicated.
I pulled a bunch of data fromdifferent sources but every
month it would break becausesome new value that I hadn't
anticipated would show up in thesource.
To go fix it Like it was likeyou usually go like, oh, fix it
in this one spot and then rerunit and it was fine, but it was a
constant thing I had to do.
Right, run the flow, it wouldbreak.
(54:03):
Go fix it.
Run the flow, oh, run the flow,it would break.
Go fix it.
Run the flow.
Oh, different thing break, fixit.
Right and it's yeah, I meanthat that by itself like the
ability to to do things like job, job function standardization
or job level standardization,banking excluded.
Right, because everybody's a vp, but yeah, um, it's huge.
(54:26):
So, um, I'm a.
I wish we had more time.
I really do like I feel like wejust scratched the surface of
this, right, invite me back.
Ahmed Datoo (54:35):
I'm happy to come back.
Well, you should come to mobspalooza that's what you need to
do actually, I will be there ohgood, good, good.
Michael Hartmann (54:42):
well, I am still on the fence, so I'm still
trying to figure out a way toget there, but I'm hopeful.
If folks want to connect withyou, continue this conversation
or learn more about what you'redoing, what's the best way for
them to do that?
Ahmed Datoo (54:55):
Yeah, they can go to allgoodhqcom.
That's our website.
I'm on LinkedIn.
Just connect with me onLinkedIn.
Michael Hartmann (55:02):
Awesome, that's terrific.
Well, thank you again for yourtime.
This has been a lot of fun, Ithink.
I think, uh, we're gonna havelots of controversial stuff that
we'll be able to use to promotethis, because I think that's
that's going to be a game.
But no, it was.
It was actually really good.
I think a lot of the ourlisteners um will benefit from
getting your perspective as a,as a CMO, about how marketing
ops is viewed and what you careabout and how you have to then
(55:23):
communicate things.
I think there needs to be moreof that.
So, thank you for that.
Thank you to our supporters andlisteners and, soon, viewers.
We really appreciate you.
Thank you for that.
If you have ideas for guests ortopics or want to be a guest,
you can reach out to Naomi, mikeor me, and we'd be happy to get
the ball rolling Until nexttime.
(55:44):
Bye, everybody.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com