August 7, 2024 • 41 mins
In this episode of the Restaurant Technology Guys Podcast, host Jeremy welcomes Anan, a long-time listener and an AI expert with significant experience in restaurant tech and operations. The discussion dives deep into the challenges and solutions associated with delivery in the restaurant industry, focusing on Loop AI—a platform designed to help restaurants understand and improve their delivery profitability. Anan shares his journey from AI research to working at Uber and starting his own restaurants, and eventually creating Loop AI to address the financial and operational struggles of delivery services. The conversation covers the intricacies of delivery, taxation, profitability, and marketing spend, emphasizing the importance of data and technology in making informed business decisions for delivery channels.
00:00 Introduction and Welcome
00:13 Guest Introduction: Anan's Journey
01:34 The Rise of Delivery in Restaurants
02:26 Challenges in Delivery Profitability
03:16 Loop AI: Revolutionizing Restaurant Delivery
06:02 Understanding Financial Complexities
11:34 Marketing and Consumer Behavior
17:02 Taxation and Reconciliation
38:10 Engaging with Loop AI
41:00 Conclusion and Call to Action
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:02):
This is the RestaurantTechnology Guys podcast.
Helping you run your restaurantbetter.
Jeremy (00:10):
back to the restaurant technology guys podcast.
I thank everyone out there forjoining us.
As I say, each and every timeyou guys listen, it is an honor
to come on the air and talk toyou guys today.
We are talking about deliveryand all it entails.
And so I'm going to introduceour guest here for just a second
and let him share a little bitabout how he got into restaurant
tech.
And so I'm sure I'm going toscrew this up, even though I
(00:31):
just asked him how to say hisname.
it is add non.
No, I screwed it up.
Why don't you introduce yourselfreal quick to our listeners?
And then we can talk a littlebit about what you get to do and
how you got into restauranttech.
Anand Tumuluru (00:43):
No worries, Jeremy.
It's a hard name.
It's Anan.
yeah, it's Anan.
it's a pleasure to be on theshow, Jeremy.
I'm a long time listener.
and, a quick background aboutmyself.
I started my career, as an AIresearcher.
Worked on a previous generationof, machine translation or think
of it as a previous generationof chat.
and, later worked, build atFlipkart building accounting
(01:07):
receivables, over there andlater, transport management
capabilities and also onflipkart's Cloud.
and most recently I was at Uber,through the pandemic.
Saw the growth of delivery frominside the marketplace.
concurrently, I'm also arestauranteur, have a, have a
couple of restaurant locations.
it's a passion project, but,just a lot of us, from the world
(01:28):
of restaurants.
It's a world, where once you arein, you're locked in and you
want to keep going.
So it's a small side projectthat we, that I take seriously
and draw a lot of, a lot of, mypersonal, Pleasure and,
indulgence from
Jeremy (01:43):
I love that.
and, I know that today reallythe topic of conversation is all
wrapped around delivery and someof its intricacies and how, both
operationally and financiallyand all of those kind of things.
And so your most recent projectand why you're here is to talk
about loop AI.
So why don't you give everybodyan intro to what is loop AI and
where did it come from?
and, I'm excited actually totalk about it because I think
(02:06):
there's a lot of people that arestruggling with these problems
and likely they don't know thatthere's solutions out there like
what you and your team have beenable to put together.
Anand Tumuluru (02:13):
absolutely showed me.
So quick context is that as arestaurant or the total amount
of software spend, that we madea zero.
I know what it means to, as arestaurant or to basically look
for every single penny, and makesure that the operation is
profitable.
But at the same time, it'sreally hard to understand.
where the cash flow is and howdo you plug the holes?
(02:33):
And how do you maximize yourrevenue?
coming from that world and goingthrough the pandemic and sign
the seeing the growth of,delivery over, over such a short
span.
one of the big, one of the bigquestions that sort of all
restaurants, were strugglingwith was how do we make delivery
profitable?
(02:53):
And, right around the time ofpandemic, when we would start,
when we're starting to get outthere, it was okay, socially
okay to be out there, a lot ofus were helping our favorite
restauranteurs, our favoritecommunity shops, and mom and
pops out and, just like all ofus, we were out there talking to
some of our favorite restaurantsand.
And literally all of them justsaid, Hey, delivery is our
(03:15):
primary source of revenue.
And we were struggling to, andthey were struggling to
understand how to, how to makeit work.
So that's sort of, where loop,started.
We basically worked with verysmall mom and pops restaurants,
trying to understand what thebig problems they had with
(03:35):
delivery work.
And surprisingly.
with a ton of our partners andfriends today, order aggregation
was not really a problem, right?
So they were able to get ordersfrom Dodo.
I should be scrub hub.
They were able to inject it intotheir, into the T.
O.
S.
Or, maybe they had one tablet,to rule them all.
And that world was working fine.
(03:55):
The problem was.
They weren't really sure, howprofitable delivery was and,
whether they were making money.
30 percent sounded like a lot ofcommissions and they were
struggling to understand whatthey can do to get it right.
And, as data scientists andengineers, one of the first
things that we did was, Hey,this is the time for.
(04:16):
an extremely sophisticated menu,right?
So let's use data to figure outwhat the optimal menu is, what
the layout is, let's optimizeimages.
And maybe there is 20 cents ofpricing that we can change and
did a whole ton of, whole lot ofminor and major changes to their
menus and to their, to thedelivery applications.
And at the end of, at the end ofthe month, we would see these,
(04:40):
Graphs go up in the DoorDash,Ubers like, the revenue would
raise.
but when we go back to theoperators and we tell them, Hey,
have you seen all of thisadditional money come into your
bank account?
their response was, what money?
What are you talking about?
What did you do?
And.
We were shocked.
(05:00):
We were surprised and shockedbecause in the world of Uber,
which is a company that's drivenby extremely large data sets and
large scale experimentation, ifyou made a change that caused a
1% increase in any metric, it'sa companywide celebration.
Like it's a massive event.
in these restaurants, we weremaking like five, 10, 20 percent
(05:22):
difference to their revenue andto their orders.
And based on what they weretelling us, they couldn't see
any difference to their cashflows, to their sort of profit
and loss.
And that was so stunning to usbecause, we just couldn't even
reason with them in terms of howmuch they're making.
And, and that sounded like amuch bigger problem.
Like, how are you operating awhole line of business?
(05:44):
Without even knowing how muchmoney you were walking away at
the end of the month with andthe restaurant Owners basically
said hey, why don't you go talkto our accountants?
you know we go talk to theaccountants and they're like You
know feel free to look at thequickbooks and we followed that
rabbit hole and at the end ofthe day What we realized was
that restaurant accounting is amess it's very hard to
(06:05):
understand your revenue bychannel your profitability most
of them just operate accountingjust so that they can file taxes
Jeremy (06:12):
Mhm.
Anand Tumuluru (06:12):
sales tax liability thing.
It's payroll taxes, stuff likethat.
there isn't like seriousattribution to a specific
channel and amidst thisextremely broken world of
restaurant finances, you have acompletely new business line,
which is DeliBee.
Jeremy (06:26):
Mhm.
Anand Tumuluru (06:27):
it's a whole new financial and operating model.
You have a 30 percent cut interms of commissions.
But you also have a significant,addition in revenue, right?
and is it incremental?
Is it not?
How much cost do you amortizebetween food versus operations?
how would you think of,increasing prices on these apps?
(06:48):
If you increase too much,nobody's going to come to your
store.
If you don't increase too much,you're probably just giving away
free food.
marketing is really effective.
how much do you want to market?
There's just so many questionsthere and you cannot answer
those questions without a clearUnderstanding of your finances.
You cannot make these decisionswithout data and if you
especially fast forward thisentire evolution of delivery
(07:12):
assuming delivery continues togrow
Jeremy (07:15):
Mhm.
Anand Tumuluru (07:16):
And this revenue channel the operations and
finance the complexity withoperations and finances
continues to grow It's anobvious idea in retrospect You
10 years down the line, youcannot imagine, you cannot
visualize a world where thesedecisions are being made without
Jeremy (07:32):
Without understanding them.
Anand Tumuluru (07:34):
without understanding the data.
so for us, what we are buildingis fundamentally a financial
foundation, and a bunch ofoperational levers on top of
them so that restaurants can notonly understand the finances by
this delivery channel, butexecute the operational levers
so that they can make itprofitable.
And especially over the nextdecade.
as delivery grows, as it cementsitself as a more, stronger
(07:56):
channel with, with all kinds ofrestaurants, certainly with fast
casual and the more, take awayspecific formats, like off
premise formats, but definitelywith all restaurants, we think,
it's a no brainer that, asoftware like loop AI, and, a
data, first, orientation towardsdelivery.
it's just going to be an obviousthing that all, every brand must
(08:16):
have to do.
Jeremy (08:17):
Yeah.
And so there's so many differenttrains of thought that I have,
after the description.
So first and foremost, you're100 percent right.
I have an eight year old.
I remember, my wife got aDoorDash gift card, a digital
gift card eight, nine years agowhen, right after she was born.
And quite frankly, I was in thespace and I had no idea.
And so the rapid adoption ofDoorDash, UberEats, GrubHub,
(08:38):
just all of the different TSPsthat have been out there has
been One of the most rapidchanges to restaurants prior to
any of these DSPs being outthere, it was, you could only
get pizza for delivery.
And if you were in certain geos,maybe Chinese food, maybe Indian
food, but you had to be in amajor city, you couldn't likely
get it in the suburbs unless youwanted pizza hut or dominoes.
(08:58):
Now, I would say close to 90percent of restaurants have some
form of delivery.
You talked a little bit aboutkind of aggregation.
We've got multiple aggregatorsthat have been on the show.
So from that perspective, Ithink everybody's figured out
how to get it from thosecompanies into the point of
sale.
what they haven't necessarilydone, and I would love to talk a
little bit about this and howyou guys see this as is taxation
and pricing.
(09:19):
And out of stocks and all ofthat, those are problems.
Those are problems that a lot ofpeople are dealing with.
And so understanding did theconsumer pay to uber and what
did uber pay the restaurant?
Have you guys figured that pieceout?
Because I, I feel like yourdescription of the problem from
the last five years, you couldhave been sitting in 50 of my
customer meetings because somany customers are like, Yes, I
(09:39):
have to turn on delivery, but Idon't know how profitable it is.
And I know if I don't have it,I'm going to lose this customer.
But I know if I do have it, Idon't know how profitable it is.
And is it incremental?
Is it not incremental?
So I want to tease out each ofthese, but I'd love to talk a
little bit about evenidentifying the order and how
did it come through and whatdoes it look like, I got, a
pizza on this, order and I paid18 for it to Uber Eats.
(10:02):
How much of it's coming through?
Have you guys figured that pieceout?
I guess let's start there.
How much of it was taxed?
How much it was the product.
Anand Tumuluru (10:09):
Absolutely.
Jeremy.
So one of the core pieces thatyou get with loop AI, is you
basically look at an incomestatement for your third party
delivery with your gross profit,assume that a hundred percent of
that revenue is incremental.
What's your gross profit, right?
against that.
So breaking down the differentfee components, understanding
(10:31):
your marketing spend.
Maybe there is the problem ofchargebacks.
Maybe there is a significantamount of unfulfilled sales and
in the name of canceled andmissed orders.
you touched on a very importanttopic in terms of taxes.
how much taxes does therestaurant?
Oh, versus what was themarketplace facilitator?
component to that, we basicallyunderstand that at an order
(10:53):
level
Jeremy (10:54):
Okay.
Anand Tumuluru (10:55):
at a payout level and basically make sure
that the restaurant is veryclear about what their gross
profit is, not just at anaggregate, but by channel, by
Doordash, by Uber Eats, by, evenwithin channels by dash bars,
customers versus non dash bars.
and even within that, how muchof that was driven by marketing?
(11:17):
What's the true incremental ROASon marketing?
and so on and so forth.
So by different spent years ofcommissions or the money that
they're investing into thesechannels.
How much are they making?
And you can also go down all theway down to an order level and
understand what you actuallymade from that order net, all of
(11:38):
these sort of deductions.
Jeremy (11:40):
I love that.
just for our listeners thatmight not understand what row as
is.
I do, but, I think it's, I thinkit's one of those things that
helping them understand you'vementioned marketing a few times,
and this is one of the thingsthat the Uber Eats, and DoorDash
drivers talk about is we'regoing to get you in front of so
many more eyeballs And you'regoing to spend money to get us
there.
So if you can help our listenersto understand what that looks
like or what that even means,because, again, at times I get
(12:04):
people going, you have thisguest on and I had no idea what
they were talking about.
Can you define that for me?
So would you mind doing that forour listeners?
Anand Tumuluru (12:11):
Absolutely.
So from, from a very, from avery first principled sort of
investment and written oninvestment, terminology, if you
basically spend, let's say 30percent off your revenue, you're
expected to get, let's say, ahundred orders from them,
according to the market basis.
(12:32):
If you actually spend 40 percentby, upping your spend on
marketing, let's say you invest10 percent more, you potentially
get 200 orders or 150 orders.
So if you actually spend alittle more than your
commission.
Then your base rate, you canactually make a lot more money,
in terms of a lot moreincremental orders just because
(12:54):
you're capturing customerseyeballs or helping them convert
once they're at your storebecause they look at a promotion
they would otherwise have notchecked out.
But now they want to get thatpromotion and they check out.
So the million dollar questionthat, truly a million dollar
question in the sense of brandspending millions of dollars and
in marketing is.
(13:14):
Hey, my commission base rate is25 percent.
Could I spend three percent morefive percent more or ten percent
more?
In order for me to maximize thatrevenue.
so can I get more, a lot moreorders and at what percentage of
incremental spend at what more,let's say maybe at 5 percent you
can actually get 50 more orders,but at 10 percent you just get
(13:35):
like 60.
Jeremy (13:38):
It's not.
It's really that measurement ofyour ad spend.
What is the return on ad spend?
I think is what a row as standsfor.
And so it's if I spend this,what do I expect to get back?
If I spend X, I get, I get this.
If I spend Y, I get.
this next amount.
And so understanding what thatlooks like and what did it
drive.
So I love that you guys aredoing that.
(13:59):
Cause so few people evenunderstand that they spray and
pray, throw a bunch of money atit and go, I don't know, was it
good?
Was it not good?
So talk me through that.
Anand Tumuluru (14:07):
yeah.
So the, again, the real worldapplication of this is very
straightforward.
the mathematics of it and thescience of it.
can, can get abstracted away,but fundamentally, let's say
you're a brand, I'm just goingto use starboard as an example,
just because we were on the callwith them before.
And my, it's my favorite brandin the Bay area.
and it's, phenomenal, gardenbird salad.
(14:28):
I'd recommend everybody try.
They have a 4.
8, 4.
9 rating on delivery.
They were one of the earliestdelivery, first brands.
They adopted delivery.
They do really well.
All of the finances andoperations as it relates to how
the food flows out off premise.
super low error rates,phenomenal brand visibility.
(14:51):
Everybody knows about starboardthere, their brand.
it doesn't need to spend moreon, let's say, ads, right?
Because everybody knows aboutStarboard.
there isn't, or even if youdon't know about Starboard, if
you come to Doordash andStarboard is within your
serviceable area, you alreadyknow that it's present.
(15:12):
it's around the area.
Contrast that with, maybe even aMcDonald's, right?
That's completely new to thearea, right?
So the customers in thatdelivery radius didn't even know
about this specific brand beingaround.
So it makes a lot of sense forMcDonald's to spend on ads just
to know, just for the customersto know that McDonald's just
(15:35):
showed up.
in that area.
at the end of the day, I thinkthe nuances around the way that
you run these ads or the waythat you measure these, they're
obviously different.
They're different between Googleand Facebook, as they're
different with marketplaces,too.
But I think the principlesremain the same, which is Hey,
if you already have a lot ofvisibility ads is not the way to
(15:57):
go And in fact you want to knowwhere there is headroom for you
to get A lot more consumers tobe aware of your presence in the
first place and maybe you wantto run ads there similarly with
promotions Maybe you want, a setof customers to try out your
brand and you have that, thegood news is that with the third
party platforms, you have somuch demand in terms of
(16:18):
consumers.
You have that ability to pickand choose these levers, at a
store level, at an, by audiencetype, you can basically deploy
these ads.
For the right store for theright audience type to basically
make up for whatever the storeneeds to be doing in your vision
for that brand and that'sbasically where loop.
ai comes through, we are helpingwe are helping these brands get
(16:40):
that nuanced understanding ofhey, where am I lacking?
How do I spend this money?
How do I make sure i'm notgiving away free food while I
deploy the spend?
and at the end of the day, makesure that You Their third party
delivery presence is ideal andis according to their brand's
vision of how they need to beviewed by customers.
Jeremy (17:00):
I love those, those thoughts.
And again, I think they're allvery needed.
And I think they're because somany people got shotgunned into
doing delivery because of thepandemic, or, they never really
thought through these things.
one of the other things that youtalked about is kind of
taxation.
Taxation is a pain in the buttbecause my understanding, and I
guess, correct me if I'm wrong,is that, even if your restaurant
is in one zip code and one taxzone, they're Uber eats because
(17:23):
they're such a big company orwhomever it might be that's
delivering actually has tocharge the tax rate of where
you're delivering it to notnecessarily where the restaurant
is.
If I'm brick and mortar and Iwalk into the restaurant, I go
order food.
It's whatever the tax rate is atthat physical geography, my
understanding, and I guess,correct me if I'm wrong, is the
taxation for that order isactually where it's getting
delivered to based on thecurrent tax laws, which may be
(17:45):
different than the tax fromwhere I ordered the food from,
which creates a discrepancy.
With your data.
Is that a fair assessment andfiguring out the data between
those two is often a challenge.
And now a word from one of oursponsors.
Every restaurant operatorunderstands the chaos of a
Restaurant kitchen during themeal rush restaurant
(18:05):
technologies, oil, total oilmanagement solutions, and end to
end automated oil managementsystem that delivers filters,
monitors, and recycles yourcooking oil, taking the dirtiest
jobs out of your kitchen andletting your employees focus on
more important tasks.
Control the kitchen chaos withrestaurant technologies and make
your kitchen safer.
No upfront costs to learn more,check out rti- inc.
(18:28):
com or call 888 796 4997.
Anand Tumuluru (18:34):
No, that's a totally fair assessment, Jeremy.
And the good news is that,again, just like with marketing,
third party marketplaces have alot of tools, have a lot of data
around this.
So what they do for therestaurant, barring a few
exceptions, there is alwaysexceptions where Michigan passes
a new law and, they're a coupleof months behind, barring sort
(18:55):
of those exceptions.
They do a phenomenal job ofputting estimating the tax
liability, both the taxes thatthey've withheld versus what you
need to be paying, to your stateauthorities very well at an
order level, at a transactionlevel or at a payout level.
The problem is that it's veryhard to actually integrate all
of that data in, in the contextof your point of sale.
(19:18):
Which has its own taxesconfigured and your accounting
software, which again has itsown rules and logic that's
applied to it.
The second thing is that there'sa whole set of decisions that a
brand needs to make, in termsof, pricing, right?
So if they're increasing prices,so what is it getting taxed on?
(19:38):
Or let's say they spend onmarketing.
they need to make sure thatthey're paying the taxes for the
right amount.
And that basically.
Listening to the marketplacesand taking that help in terms of
sizing their liabilitycorrectly.
a lot of them actually end uppaying double taxes, right?
They basically end up, just tobe considerate, just to be safe,
(19:59):
what they do is, Hey, I made 100on third party delivery.
I just need to pay 8%.
I'm just going to pay thatentire 8%.
Turns out 5 percent of that wasalready paid by the marketplace.
You're just supposed to paythree percent.
so what loop.
ai is helping do is not justunderstanding the tax
liabilities well and buildingthe order and audit trail for
(20:21):
you to understand what, what ledto a certain calculation of tax
liability.
We are actually automatingreconciliation and filing those
sort of tax entries in yourQuickBooks or in your Restaurant
365 directly.
So we help actually create thosejournal entries and we don't
give you another Excel sheet oranother dashboard for you to
(20:41):
figure out and navigate.
We actually go and close yourbooks for you, so that you don't
have to deal with the complexityof trying to break these nuances
down.
Jeremy (20:50):
I, I appreciate that.
And again, I, I've got a leadfor you.
After we get done with the call,I got a customer that I was just
on the phone with the other dayand he's can you help me fix my
tax thing?
So we can talk about thatoffline.
But, another line of thought Ihad that you talked about, I
think it's, Yeah.
Until you get into the data andyou start to look at it is each
of these delivery serviceproviders do things differently.
(21:12):
They give you data differently.
They give you your reportingdifferently.
They charge you differently.
They may have different rates,different times of day,
different rates.
Talk to me a little bit about,for those that just think, oh,
all delivery service providersare created equal, or they're at
a brand right now that only issupporting DoorDash, but they're
considering Uber Eats andGrubhub.
(21:34):
They all do it differently.
And because they all do itdifferently, reconciling them,
you have to build systems or,engage with something like loop
AI to understand what thosereconciliations are.
So can you walk me through that,that each one of them are their
own tech that's been createddifferently with its own set of
rules, its own set of outputs,its own set of dashboards, its
own set of those kinds ofthings.
But I think all too often, if Ihaven't dug into it, or they
(21:55):
say, that's my accountant'sproblem, it's not my problem.
I just need to offer Uber Eats,but those levels of complexity
that exist because every one ofthem is different.
Anand Tumuluru (22:03):
great question, Chadmi.
And I think.
first things first, I think theanswer to these questions, I'm
always afraid to indirectlyimply that my old employer, Uber
or Doordash, I'm not doing agood enough job, off of doing
these reports.
And I just want to preface thisentire conversation by saying, I
(22:24):
think we, we should, we shouldhave the understanding that this
entire industry was created likethree years ago.
Everybody's figuring everythingout, including the market cases,
right?
So they're figuring out theirown sort of profitability.
They're figuring out what theright team structures are, what
the right reporting standardsare.
And by and far.
The way that the big two,Doordash and Uber Eats, do all
(22:46):
of these transactions, payoutreporting, and helping
restaurants understand sort ofthe breakdowns, is, miles ahead
of, everybody else.
So they have, to their credit,invested quite a bit in
simplifying finances andoperations for restaurants.
there is no doubt about that.
Except, it's, It has a long wayto go, right?
(23:07):
there is still a lot of work todo.
The key problem here,Fundamentally Jeremy, is that
from Dotash's standpoint, theyare building their own ideal
standard and, it's perfect andit's beautiful, except a
restaurant is looking at 15 suchstatements.
They're looking at their POS.
They're looking at EasyGator.
They're looking at Grubhub.
And then they're also looking atLunchbox.
(23:28):
And then they're looking at,Olo, Olo and Thanks.
And at the end of the day, they,the amount of just spread that
they have with respect tounderstanding what was the
actual payment that they gotfrom any of these sort of
providers is a complex maze.
And we went from a world of Noteven having cloud based point of
(23:48):
sale to having 15 different sortof, digital ordering channels on
steroids, just pumping, all ofthese orders into the
restaurant.
So we just live in a world whereit is complex and overwhelming
for the restaurant.
Now, in terms of the, in termsof the complexity that exists
with understanding payouts,everybody's doing their own
thing.
(24:08):
First of all, there isn't like astandard or there isn't like a
rule set that says, This is howyou are, how marketplaces are
supposed to report, or these arethe fees that you need to
expect.
Atoms have been made atdifferent cities and, different,
different states and citieswhere they try to say, Hey,
here's how you can charge it.
But.
That only led to an explosion ofmore than items.
(24:29):
at the end of the, at the end ofthe payout.
So I think everybody's trying tofigure out what's the right way
to, what's the right way topresent it, present these
transactions and payouts.
Now, one of the things that Iwould like to point out, Jeremy,
is that in the world of thirdparty delivery.
The complexity is actually a lotmore than what you would have in
(24:50):
store.
for your regular POStransaction.
So for POS transaction, all youhave is you just go buy the food
and you basically have, let'ssay a 2 percent deduction on
your credit card, right?
But in the world of third partydelivery, you have marketing,
right?
Which is super important.
you want the visibility, youwant promotions.
You potentially have, refunds,which are also bundled in, for
(25:12):
what it's worth.
It's significantly larger thanwhat you would have in store,
fundamentally because there's alot more operational complexity
for taking food out.
But at the same time, it'sactually simpler than, handling
chargebacks is simpler than whatyou would have with your POS,
because with POS It's a lot morecomplex with POS, right?
And then you have the taxes,marketplace, withholding taxes.
(25:33):
And on top of all of this, youbasically have your first party,
delivery orders, which are fixedfee based, ordering channels.
So the complexity with respectto the payouts and transactions,
it's warranted, it's necessary.
It represents, The complexity ofthe channel and the way the
consumers are actually changingtheir behavior, right?
(25:54):
So it represents the wide arrayof things that you would do with
a regular e commerce store,right?
It's no different from what youwould do if you were selling
let's say an iPhone With your ecommerce on an e commerce store
or a bag like or Apple so Ithink there is a lot of
parallels to the world of ecommerce and There's a lot of
complexity that necessitatesrunning your digital delivery
(26:17):
channels well, except it'scompletely new.
Each of them is doing it intheir own way.
And at the end of the day, thefundamental question of, hey,
This is all great.
Like at the end of the day, whatis a gross profit?
Like how much am I actually likepaying?
that question becomes harder andharder to, answer for
restaurants without usingtechnology.
If you're basically, deployinganother person to sit down and
(26:38):
pull these reports, or if youare asking your, outsourced
accountant to do it, good luckwith that.
It's just going to take so muchtime, so much effort, and it's
going to be a one time thing.
just trying to figure out whatthese calculations are.
Jeremy (26:50):
and at the end of the day, in most brands, it doesn't
even represent 50 percent oftheir total volume.
And so you're now at, 20 percentof your volume is through DSPs,
but you've got this complexityof having to deal with this.
So without tech, the juiceoftentimes isn't worth the
squeeze.
And which is why I love, I loveaggregators, order aggregators.
We've had multiples on the show.
I love that they allow theorders to get from the web down
(27:12):
into their store.
So operational complexity, Ininside of the four walls is
great.
And now we've got tools likewhat you've got that are going
to help people reconcile.
Am I making money?
How am I making money?
and really even be able to makebusiness decisions.
And so I'd love to take thatpath because again, Not all
deliver service providers arecreated equal.
(27:33):
Not all of them are going to dothe things that you might want
to deliver on your brand promiseto people.
And depending upon what younegotiated and how long your
term is on your contract, youmay be paying more to one of the
delivery service providers, butthey deliver a higher level of
service.
And maybe the guests are willingto pay more, but you don't know
(27:53):
that unless you have the data tobe able to make those decisions.
So talk me through what all ofthat looks like, because I think
all too often people, I ha I hada customer as silly as this
sounds.
I had a customer recently thatsaid, I can charge more on my
third party delivery than I doin store.
And I looked at them and said,absolutely.
You should, because it's ahigher value to that customer.
(28:13):
Yes.
It's the same cheeseburger.
Yes.
It's the same chicken sandwich,but you're delivering your costs
are higher.
And, all of it.
And so across the board, talk methrough how you guys think about
that and how the data that youguys serve up gives people the
capability to deliver on thosepromises in different ways that,
that they hadn't consideredbefore.
Potentially.
Anand Tumuluru (28:32):
Absolutely.
Absolutely.
So if you, there are a couple ofwhat you might think, think of
as ground rules here, Jeremy.
So the first one is that thecustomer that's coming to
marketplaces, those marketplacesworked hard to get them right.
They have billions of dollars ofspent in terms of.
Demand generation on Google, on,on social media, their own
Jeremy (28:57):
they're the ones that are at the top of the list.
When you search for starboard,They show up door dash might
show up before the actualcustomer's website because
they're trying to drive thatorder through there just to,
create a direct correlation ofwhat you're saying.
They're putting the ad in Googlewhen you go search or when you
go search somewhere, they'reputting that at the top of your
search results that you're goingto door dash thinking that you
(29:18):
might be ordering directly fromthat website.
I apologize.
I didn't mean to cut you off,but I think it's, I think it's
important for people tounderstand that.
Yeah.
Anand Tumuluru (29:25):
and I think Google really, is really going
in the right direction there.
again, we're all figuring thisout.
So Google sort of made thisrecent change where they, where
they let the restaurant chooseif that Doordash or, Uber Eats
store is linked.
but a lot of that behavior.
consumer behavior, especiallythe strongest cohorts of
consumers with thesemarketplaces don't even start
(29:46):
from Google.
They are subscription customersof Doordash.
Of Uber Eats, they're openingthat app in order to search for
food in the first place.
So the food search is not goingto Yelp or Google.
It's actually going directly tothe marketplace.
very similar to product searchon Amazon.
There was a time when peopleused to search on Google.
It's Hey, can I buy my new top,best binoculars?
(30:10):
Under 100, right?
That's what you'd search for onGoogle.
People don't do that anymore.
People actually go to Amazondirectly and search for it.
So that behavioral shift ishappening even with food.
especially here in the Bay Areawhere DoDash is super, super
strong.
People just directly search onDoDash for, best Thai cuisine.
Best Thai food nearby.
Best Pad Thai nearby.
and regardless of whether,whether it is through loyalty,
(30:33):
through search, thesemarketplaces are investing in
that consumer and they arebasically keeping that consumer
information close to theirchest.
They are keeping all of theconsumer's history, all of
their, predictions around thecustomer's behavior in the
future.
And then, and guess what?
They're actually doing aphenomenal job at getting the
(30:54):
customers to reorder.
a couple of years ago, I
Jeremy (30:56):
push market to me all the time.
They're like, Hey, you haven'tordered from this brand in a
certain amount of time, I don'thave the brands app downloaded,
but I have DoorDash downloadedbecause it's part of my dash
pass as part of my chase card.
And so I order off a door.
this is a real life consumerthat's in this space.
I know exactly what they'redoing, but almost every day I
get a push notification sayingyou haven't ordered from X, Y, Z
(31:17):
brand.
It doesn't matter the name ofthe brand.
What would it look like if Igave you 10 percent off an order
today to drive that behavior?
And it now has me thinking aboutthat.
And I might not take that offernow, but the next time I go to
think about it, it's like, Hey,I haven't had Pad Thai in a
while.
Maybe I'll go to DoorDash thisPad Thai.
So these are real life exampleswhere they're in your customer's
(31:39):
inbox.
They're in their.
apps on their phones or in theiremail inbox, oftentimes more
often than you are to try anddrive that behavior.
So I apologize.
I keep cutting you off, but Ilike, I think that it's critical
for people because all too oftenrestaurant owners, when I sit
and listen to them, they don'tuse these apps because they want
to go experience, they gotanother hospital industry, so
(32:01):
they're not the consumer.
But they need to understand thatthe consumers that are out there
that are using these apps Aredoing these things and it looks
different So i'll let you keepgoing because I think it's
critical for them to understanddiscovery And behavior and them
driving that to your brand andwhat that means for you.
Cause you likely wouldn't havegotten that customer without it.
(32:22):
And now a word from oursponsors.
Growing your business can meanbig time logistical questions
like, how am I going to keep upwith all of these local
deliveries?
Let UberDirect offer you ahelping hand.
With UberDirect, you can takeorders directly from your
website, app, or by phone.
Then drivers who are part ofUber's courier network will
deliver them to your customer'sdoorsteps.
Sound too simple to be true?
(32:43):
Check out uberdirect.
com to learn more.
Anand Tumuluru (32:45):
And no, please keep adding your insights and
thoughts too.
I think it's, I think theimportant thing here is that
there is a behavioral shiftthat's going on with consumers.
in the ways that you justdescribed, which is, which is
hard to understand, which ishard to internalize or build an
intuition around, especially ifyou're a restauranteur, if
you're a restauranteur, you'rebuilt to be a person that's
(33:07):
extremely hospitable to a personthat's in person.
You want to give them the bestservice.
You want to give them the bestfood.
You want to make them smile.
You want to give them a greatexperience.
You're not necessarily thinkingin the world of sort of bits and
digits, right?
I think it has to be said morenumber of times to really
internalize The consumerbehavior shifts that are
happening, shifts that arehappening.
(33:28):
Jeremy, just to your point, justto close that, the funnel, so to
speak, with respect to consumerbehavior.
I think, just a couple of yearsago, I think, in the upcoming
Food On Demand conference, thetwo years ago old version of
this, somebody mentioned thatDash fast consumer, consumers
are ordering on DoDash seventimes a month.
I wouldn't be surprised ifthat's eight, nine, or ten now.
(33:51):
Which means consumers areordering on DoorDash two or
three times a week.
My question is, how long beforeit's 21 times a week, basically
what if it's all the three mealsevery day, right?
So what if they're actuallygoing into apps?
What if most of these behaviormost of the behavior for
customers is like fully digitalAnd maybe it's not 21.
(34:15):
How about right?
That's 5x thing of this marketAnd by the way, that's the
default behavior in a lot ofOther, other cities and
countries outside and it's goingto vary a lot by region It's a
growing, it's a growing customerbase and we are seeing that
growth in marketplaces while thegrowth rate has slowed down It
(34:36):
is still growing very stronglyand we are seeing this across
brands too that delivery revenueis growing, right?
Once a customer comes into theapp they can actually place an
order sometimes in one clicksometimes in two clicks You And
they're able to actually requesta refund or seek help in a
couple, a couple of clicks, theycan actually go ahead and.
(35:00):
Look at that history of order.
So let the app recommend thebest meal that they can have in
one or two clicks.
So for them, for these apps, Ithink the reality is that
consumers are sticking reallyhard to these apps.
They are ordering again andrestaurants need to go and meet
them where they are a lot ofbrands Jeremy to your point
(35:21):
earlier ask us the question,right?
So the so when if I answer yourquestion, hey, how you know
What's the profitability is thisrevenue incrementally fee?
If you get an answer to thatquestion, maybe you're hinting
towards how about we just run anoffer towards our first party
app and get all of theseconsumers out and, acquire them
directly on the brand.
(35:42):
We have seen that story play outa few times and the success
rates have been different acrossdifferent brands.
But the reality is thatespecially not with pickup
orders, but with deliveryorders, The apps are so much
more convenient, the marketplaceapps are so much more convenient
that we inevitably see customersgo into the brand's app and then
come back at a later point oftime once we saw offered ease
(36:05):
out.
And especially the subscriptioncustomers that are extremely
sticky, right?
They keep going back to thethird party apps.
So it's this behavior.
it's just consumer behavior.
So the reason I'm giving you thecontext around the consumer
behavior is that the revenuethat marketplaces are generating
for restaurant in the debate ofwhether it's incremental or not,
(36:25):
it's really incremental.
There is just a ton of anecdotalevidence for brands that have
tried to take these customersaway and have seen those
customers go back.
Plus, the subscription behaviorthat's Important for us all to
understand and to know thatinvesting in these channels is
important.
So once you have that, once yourecognize that cc data and
(36:47):
anecdotes and hear from otherpeople's experiences, the next
question is, Hey, how muchshould I invest here?
What's the, how much of my brandshould be present on these third
party apps?
How much of my revenue should Iexpect?
Should it be 15%?
Should it be 30%?
How much should I invest in afirst party channel?
And how much should I drive myrevenue there?
Those are the things we all haveto figure out in the next two,
(37:08):
three years.
Every brand has to figure out.
Should I invest, let's say, 100,000 in an app so that I get at
least 10 percent of thecustomers into a first party
ordering experience?
Should it be limited to pickups?
We see a lot of brands, forinstance, Dave's Hot Chicken,
one of our favorite brands.
They don't do delivery on theirapp.
They just do pickups, right?
They don't deal with thedelivery experience at all
(37:29):
because they know that, thirdparty apps, the way that they
offer that convenience, itworks, right?
It works.
That's where customers aregoing.
They're happy with it, but thenthey have a phenomenal, app for
pickups.
So all of these decisions interms of what should my channel
mix be, In terms of third partyversus first party, pickup
versus delivery.
(37:49):
How much marketing should Ispend?
These are decisions that brandbrands need to make, in, in the
next one, two, three years,
Jeremy (37:56):
I think it's, I think it's ironic cause we'd never put
a New York strip on our menuwithout understanding what my
food cost was and what it was acost to acquire that customer,
but we just make these guesseson what the customers are going
to be able to bear on thirdparty delivery apps until tools
like you guys have.
And it's just, It baffles me.
And to your point, it's only atwo or three year old problem.
So I think as more and more databecomes available, as these DSP
(38:19):
solution providers end up givingmore data back to the
restaurants to understand thesethings, and then there's tools
like what you guys have, we'llcontinue to get better.
even the delivery experience nowversus what it was two years ago
is Phenomenally better.
And it's gonna only continue toincrease.
and so talk me through.
What does it look like to engagewith loop AI?
What?
What's the ideal customer?
How do they engage with you?
(38:39):
How do they learn more about howyou can help them to understand
the profitability and truly makegood science and data based
decisions about their deliveryservice?
Partners and how they're goingto go into market.
Anand Tumuluru (38:52):
if you basically have a delivery revenue
contribution, that's 20 percentor a hundred thousand dollars
that's coming through from thirdparty marketplaces, loop, loop
AI will be a great fit, for youand fundamentally we help, on
onboarding, the brand for us isvery straightforward.
it takes a few minutes for themto sign up, get loop AI set up.
(39:13):
The first thing that we alwaysrecommend restaurants do is look
at their income statement andlook at all of the line items
where the fees are gettingdeducted, where, where they're
making money or.
Where they're not, and have astrategy, have a plan towards
attacking those different lineitems.
And once we set those goals, wesee returns or improvement in
(39:34):
the profitability as early asone week.
And definitely over a span ofthree months, for instance, with
Craveworthy Brands, one of ourfavorite customers, we've
reduced that downtime from 300minutes a month to 100 minutes.
while they were going through aPOS integration again, one of
those things very hard to knowthat your stores are not even
open for customers to place anorder.
it's just a hidden problem, wecan go solve for that.
(39:54):
Another, another example is,Hey, we are mandated to spend.
According to our marketplacecontract, we should spend a
million dollars in marketing.
We are worried we are givingaway free food.
We are worried we are spendingthat money inefficiently.
how can we understand whatcampaigns are working well for
us?
And we, deploy the cash in a waywhere we can actually grow our
customer base.
(40:15):
another area where, you know,Duke AI can help.
And I think the, our favoriteuser base are the finance teams.
CFOs who, who know the pain or,even outsourced, CPAs who know
the pain of actually pulling allof these reports, understanding
the different formats, keepingup with the reporting format
changes and actually, closingtheir books by creating these
adjustments.
(40:35):
All of that's gone.
once you're on board to loop AI,we basically take care of all of
that reconciliation for you.
the, we truly understand theback office complexity of.
Managing third party deliveryand we basically make sure that
we touch all bases with respectto the burden of getting this
revenue and doing delivery,right?
you know for your team
Jeremy (40:56):
thank you for one, for creating the product.
Cause it is a problem thateverybody that has delivery has
this.
And so the fact that you guysare at a place that you guys are
solving for this is fantastic.
and again, I think, as youcontinue to remind me, it's only
a.
Four or five year old problem atthe most, it's this problem that
has been, we all feel like it'sbeen forever.
Cause you know, I don't know,COVID created, everything is
(41:16):
like in dog years nowadays.
at the end of the day, we'redoing so much of this and it's
so painful.
So I'm grateful for tools likeyours and others that are out
there that are helping solvethis problem.
And so thank you for sharing.
to our listeners, guys, like Isaid, at the onset, we love
having you guys listen to theshow, give us some feedback, go
check out loop AI.
If you haven't alreadysubscribed to the show, whether
that be on YouTube or on yourfavorite podcast player, please
(41:39):
do so and make it a great day.
Thanks for listening to theRestaurant Technology Guys
podcast.
Visit www.
RestaurantTechnologyGuys.
com for tips, industry insights,and more to help you run your
restaurant better.
This is the RestaurantTechnology Guys podcast.
Helping you run your restaurantbetter.
Jeremy (00:10):
back to the restaurant technology guys podcast.
I thank everyone out there forjoining us.
As I say, each and every timeyou guys listen, it is an honor
to come on the air and talk toyou guys today.
We are talking about deliveryand all it entails.
And so I'm going to introduceour guest here for just a second
and let him share a little bitabout how he got into restaurant
tech.
And so I'm sure I'm going toscrew this up, even though I
(00:31):
just asked him how to say hisname.
it is add non.
No, I screwed it up.
Why don't you introduce yourselfreal quick to our listeners?
And then we can talk a littlebit about what you get to do and
how you got into restauranttech.
Anand Tumuluru (00:43):
No worries, Jeremy.
It's a hard name.
It's Anan.
yeah, it's Anan.
it's a pleasure to be on theshow, Jeremy.
I'm a long time listener.
and, a quick background aboutmyself.
I started my career, as an AIresearcher.
Worked on a previous generationof, machine translation or think
of it as a previous generationof chat.
and, later worked, build atFlipkart building accounting
(01:07):
receivables, over there andlater, transport management
capabilities and also onflipkart's Cloud.
and most recently I was at Uber,through the pandemic.
Saw the growth of delivery frominside the marketplace.
concurrently, I'm also arestauranteur, have a, have a
couple of restaurant locations.
it's a passion project, but,just a lot of us, from the world
(01:28):
of restaurants.
It's a world, where once you arein, you're locked in and you
want to keep going.
So it's a small side projectthat we, that I take seriously
and draw a lot of, a lot of, mypersonal, Pleasure and,
indulgence from
Jeremy (01:43):
I love that.
and, I know that today reallythe topic of conversation is all
wrapped around delivery and someof its intricacies and how, both
operationally and financiallyand all of those kind of things.
And so your most recent projectand why you're here is to talk
about loop AI.
So why don't you give everybodyan intro to what is loop AI and
where did it come from?
and, I'm excited actually totalk about it because I think
(02:06):
there's a lot of people that arestruggling with these problems
and likely they don't know thatthere's solutions out there like
what you and your team have beenable to put together.
Anand Tumuluru (02:13):
absolutely showed me.
So quick context is that as arestaurant or the total amount
of software spend, that we madea zero.
I know what it means to, as arestaurant or to basically look
for every single penny, and makesure that the operation is
profitable.
But at the same time, it'sreally hard to understand.
where the cash flow is and howdo you plug the holes?
(02:33):
And how do you maximize yourrevenue?
coming from that world and goingthrough the pandemic and sign
the seeing the growth of,delivery over, over such a short
span.
one of the big, one of the bigquestions that sort of all
restaurants, were strugglingwith was how do we make delivery
profitable?
(02:53):
And, right around the time ofpandemic, when we would start,
when we're starting to get outthere, it was okay, socially
okay to be out there, a lot ofus were helping our favorite
restauranteurs, our favoritecommunity shops, and mom and
pops out and, just like all ofus, we were out there talking to
some of our favorite restaurantsand.
And literally all of them justsaid, Hey, delivery is our
(03:15):
primary source of revenue.
And we were struggling to, andthey were struggling to
understand how to, how to makeit work.
So that's sort of, where loop,started.
We basically worked with verysmall mom and pops restaurants,
trying to understand what thebig problems they had with
(03:35):
delivery work.
And surprisingly.
with a ton of our partners andfriends today, order aggregation
was not really a problem, right?
So they were able to get ordersfrom Dodo.
I should be scrub hub.
They were able to inject it intotheir, into the T.
O.
S.
Or, maybe they had one tablet,to rule them all.
And that world was working fine.
(03:55):
The problem was.
They weren't really sure, howprofitable delivery was and,
whether they were making money.
30 percent sounded like a lot ofcommissions and they were
struggling to understand whatthey can do to get it right.
And, as data scientists andengineers, one of the first
things that we did was, Hey,this is the time for.
(04:16):
an extremely sophisticated menu,right?
So let's use data to figure outwhat the optimal menu is, what
the layout is, let's optimizeimages.
And maybe there is 20 cents ofpricing that we can change and
did a whole ton of, whole lot ofminor and major changes to their
menus and to their, to thedelivery applications.
And at the end of, at the end ofthe month, we would see these,
(04:40):
Graphs go up in the DoorDash,Ubers like, the revenue would
raise.
but when we go back to theoperators and we tell them, Hey,
have you seen all of thisadditional money come into your
bank account?
their response was, what money?
What are you talking about?
What did you do?
And.
We were shocked.
(05:00):
We were surprised and shockedbecause in the world of Uber,
which is a company that's drivenby extremely large data sets and
large scale experimentation, ifyou made a change that caused a
1% increase in any metric, it'sa companywide celebration.
Like it's a massive event.
in these restaurants, we weremaking like five, 10, 20 percent
(05:22):
difference to their revenue andto their orders.
And based on what they weretelling us, they couldn't see
any difference to their cashflows, to their sort of profit
and loss.
And that was so stunning to usbecause, we just couldn't even
reason with them in terms of howmuch they're making.
And, and that sounded like amuch bigger problem.
Like, how are you operating awhole line of business?
(05:44):
Without even knowing how muchmoney you were walking away at
the end of the month with andthe restaurant Owners basically
said hey, why don't you go talkto our accountants?
you know we go talk to theaccountants and they're like You
know feel free to look at thequickbooks and we followed that
rabbit hole and at the end ofthe day What we realized was
that restaurant accounting is amess it's very hard to
(06:05):
understand your revenue bychannel your profitability most
of them just operate accountingjust so that they can file taxes
Jeremy (06:12):
Mhm.
Anand Tumuluru (06:12):
sales tax liability thing.
It's payroll taxes, stuff likethat.
there isn't like seriousattribution to a specific
channel and amidst thisextremely broken world of
restaurant finances, you have acompletely new business line,
which is DeliBee.
Jeremy (06:26):
Mhm.
Anand Tumuluru (06:27):
it's a whole new financial and operating model.
You have a 30 percent cut interms of commissions.
But you also have a significant,addition in revenue, right?
and is it incremental?
Is it not?
How much cost do you amortizebetween food versus operations?
how would you think of,increasing prices on these apps?
(06:48):
If you increase too much,nobody's going to come to your
store.
If you don't increase too much,you're probably just giving away
free food.
marketing is really effective.
how much do you want to market?
There's just so many questionsthere and you cannot answer
those questions without a clearUnderstanding of your finances.
You cannot make these decisionswithout data and if you
especially fast forward thisentire evolution of delivery
(07:12):
assuming delivery continues togrow
Jeremy (07:15):
Mhm.
Anand Tumuluru (07:16):
And this revenue channel the operations and
finance the complexity withoperations and finances
continues to grow It's anobvious idea in retrospect You
10 years down the line, youcannot imagine, you cannot
visualize a world where thesedecisions are being made without
Jeremy (07:32):
Without understanding them.
Anand Tumuluru (07:34):
without understanding the data.
so for us, what we are buildingis fundamentally a financial
foundation, and a bunch ofoperational levers on top of
them so that restaurants can notonly understand the finances by
this delivery channel, butexecute the operational levers
so that they can make itprofitable.
And especially over the nextdecade.
as delivery grows, as it cementsitself as a more, stronger
(07:56):
channel with, with all kinds ofrestaurants, certainly with fast
casual and the more, take awayspecific formats, like off
premise formats, but definitelywith all restaurants, we think,
it's a no brainer that, asoftware like loop AI, and, a
data, first, orientation towardsdelivery.
it's just going to be an obviousthing that all, every brand must
(08:16):
have to do.
Jeremy (08:17):
Yeah.
And so there's so many differenttrains of thought that I have,
after the description.
So first and foremost, you're100 percent right.
I have an eight year old.
I remember, my wife got aDoorDash gift card, a digital
gift card eight, nine years agowhen, right after she was born.
And quite frankly, I was in thespace and I had no idea.
And so the rapid adoption ofDoorDash, UberEats, GrubHub,
(08:38):
just all of the different TSPsthat have been out there has
been One of the most rapidchanges to restaurants prior to
any of these DSPs being outthere, it was, you could only
get pizza for delivery.
And if you were in certain geos,maybe Chinese food, maybe Indian
food, but you had to be in amajor city, you couldn't likely
get it in the suburbs unless youwanted pizza hut or dominoes.
(08:58):
Now, I would say close to 90percent of restaurants have some
form of delivery.
You talked a little bit aboutkind of aggregation.
We've got multiple aggregatorsthat have been on the show.
So from that perspective, Ithink everybody's figured out
how to get it from thosecompanies into the point of
sale.
what they haven't necessarilydone, and I would love to talk a
little bit about this and howyou guys see this as is taxation
and pricing.
(09:19):
And out of stocks and all ofthat, those are problems.
Those are problems that a lot ofpeople are dealing with.
And so understanding did theconsumer pay to uber and what
did uber pay the restaurant?
Have you guys figured that pieceout?
Because I, I feel like yourdescription of the problem from
the last five years, you couldhave been sitting in 50 of my
customer meetings because somany customers are like, Yes, I
(09:39):
have to turn on delivery, but Idon't know how profitable it is.
And I know if I don't have it,I'm going to lose this customer.
But I know if I do have it, Idon't know how profitable it is.
And is it incremental?
Is it not incremental?
So I want to tease out each ofthese, but I'd love to talk a
little bit about evenidentifying the order and how
did it come through and whatdoes it look like, I got, a
pizza on this, order and I paid18 for it to Uber Eats.
(10:02):
How much of it's coming through?
Have you guys figured that pieceout?
I guess let's start there.
How much of it was taxed?
How much it was the product.
Anand Tumuluru (10:09):
Absolutely.
Jeremy.
So one of the core pieces thatyou get with loop AI, is you
basically look at an incomestatement for your third party
delivery with your gross profit,assume that a hundred percent of
that revenue is incremental.
What's your gross profit, right?
against that.
So breaking down the differentfee components, understanding
(10:31):
your marketing spend.
Maybe there is the problem ofchargebacks.
Maybe there is a significantamount of unfulfilled sales and
in the name of canceled andmissed orders.
you touched on a very importanttopic in terms of taxes.
how much taxes does therestaurant?
Oh, versus what was themarketplace facilitator?
component to that, we basicallyunderstand that at an order
(10:53):
level
Jeremy (10:54):
Okay.
Anand Tumuluru (10:55):
at a payout level and basically make sure
that the restaurant is veryclear about what their gross
profit is, not just at anaggregate, but by channel, by
Doordash, by Uber Eats, by, evenwithin channels by dash bars,
customers versus non dash bars.
and even within that, how muchof that was driven by marketing?
(11:17):
What's the true incremental ROASon marketing?
and so on and so forth.
So by different spent years ofcommissions or the money that
they're investing into thesechannels.
How much are they making?
And you can also go down all theway down to an order level and
understand what you actuallymade from that order net, all of
(11:38):
these sort of deductions.
Jeremy (11:40):
I love that.
just for our listeners thatmight not understand what row as
is.
I do, but, I think it's, I thinkit's one of those things that
helping them understand you'vementioned marketing a few times,
and this is one of the thingsthat the Uber Eats, and DoorDash
drivers talk about is we'regoing to get you in front of so
many more eyeballs And you'regoing to spend money to get us
there.
So if you can help our listenersto understand what that looks
like or what that even means,because, again, at times I get
(12:04):
people going, you have thisguest on and I had no idea what
they were talking about.
Can you define that for me?
So would you mind doing that forour listeners?
Anand Tumuluru (12:11):
Absolutely.
So from, from a very, from avery first principled sort of
investment and written oninvestment, terminology, if you
basically spend, let's say 30percent off your revenue, you're
expected to get, let's say, ahundred orders from them,
according to the market basis.
(12:32):
If you actually spend 40 percentby, upping your spend on
marketing, let's say you invest10 percent more, you potentially
get 200 orders or 150 orders.
So if you actually spend alittle more than your
commission.
Then your base rate, you canactually make a lot more money,
in terms of a lot moreincremental orders just because
(12:54):
you're capturing customerseyeballs or helping them convert
once they're at your storebecause they look at a promotion
they would otherwise have notchecked out.
But now they want to get thatpromotion and they check out.
So the million dollar questionthat, truly a million dollar
question in the sense of brandspending millions of dollars and
in marketing is.
(13:14):
Hey, my commission base rate is25 percent.
Could I spend three percent morefive percent more or ten percent
more?
In order for me to maximize thatrevenue.
so can I get more, a lot moreorders and at what percentage of
incremental spend at what more,let's say maybe at 5 percent you
can actually get 50 more orders,but at 10 percent you just get
(13:35):
like 60.
Jeremy (13:38):
It's not.
It's really that measurement ofyour ad spend.
What is the return on ad spend?
I think is what a row as standsfor.
And so it's if I spend this,what do I expect to get back?
If I spend X, I get, I get this.
If I spend Y, I get.
this next amount.
And so understanding what thatlooks like and what did it
drive.
So I love that you guys aredoing that.
(13:59):
Cause so few people evenunderstand that they spray and
pray, throw a bunch of money atit and go, I don't know, was it
good?
Was it not good?
So talk me through that.
Anand Tumuluru (14:07):
yeah.
So the, again, the real worldapplication of this is very
straightforward.
the mathematics of it and thescience of it.
can, can get abstracted away,but fundamentally, let's say
you're a brand, I'm just goingto use starboard as an example,
just because we were on the callwith them before.
And my, it's my favorite brandin the Bay area.
and it's, phenomenal, gardenbird salad.
(14:28):
I'd recommend everybody try.
They have a 4.
8, 4.
9 rating on delivery.
They were one of the earliestdelivery, first brands.
They adopted delivery.
They do really well.
All of the finances andoperations as it relates to how
the food flows out off premise.
super low error rates,phenomenal brand visibility.
(14:51):
Everybody knows about starboardthere, their brand.
it doesn't need to spend moreon, let's say, ads, right?
Because everybody knows aboutStarboard.
there isn't, or even if youdon't know about Starboard, if
you come to Doordash andStarboard is within your
serviceable area, you alreadyknow that it's present.
(15:12):
it's around the area.
Contrast that with, maybe even aMcDonald's, right?
That's completely new to thearea, right?
So the customers in thatdelivery radius didn't even know
about this specific brand beingaround.
So it makes a lot of sense forMcDonald's to spend on ads just
to know, just for the customersto know that McDonald's just
(15:35):
showed up.
in that area.
at the end of the day, I thinkthe nuances around the way that
you run these ads or the waythat you measure these, they're
obviously different.
They're different between Googleand Facebook, as they're
different with marketplaces,too.
But I think the principlesremain the same, which is Hey,
if you already have a lot ofvisibility ads is not the way to
(15:57):
go And in fact you want to knowwhere there is headroom for you
to get A lot more consumers tobe aware of your presence in the
first place and maybe you wantto run ads there similarly with
promotions Maybe you want, a setof customers to try out your
brand and you have that, thegood news is that with the third
party platforms, you have somuch demand in terms of
(16:18):
consumers.
You have that ability to pickand choose these levers, at a
store level, at an, by audiencetype, you can basically deploy
these ads.
For the right store for theright audience type to basically
make up for whatever the storeneeds to be doing in your vision
for that brand and that'sbasically where loop.
ai comes through, we are helpingwe are helping these brands get
(16:40):
that nuanced understanding ofhey, where am I lacking?
How do I spend this money?
How do I make sure i'm notgiving away free food while I
deploy the spend?
and at the end of the day, makesure that You Their third party
delivery presence is ideal andis according to their brand's
vision of how they need to beviewed by customers.
Jeremy (17:00):
I love those, those thoughts.
And again, I think they're allvery needed.
And I think they're because somany people got shotgunned into
doing delivery because of thepandemic, or, they never really
thought through these things.
one of the other things that youtalked about is kind of
taxation.
Taxation is a pain in the buttbecause my understanding, and I
guess, correct me if I'm wrong,is that, even if your restaurant
is in one zip code and one taxzone, they're Uber eats because
(17:23):
they're such a big company orwhomever it might be that's
delivering actually has tocharge the tax rate of where
you're delivering it to notnecessarily where the restaurant
is.
If I'm brick and mortar and Iwalk into the restaurant, I go
order food.
It's whatever the tax rate is atthat physical geography, my
understanding, and I guess,correct me if I'm wrong, is the
taxation for that order isactually where it's getting
delivered to based on thecurrent tax laws, which may be
(17:45):
different than the tax fromwhere I ordered the food from,
which creates a discrepancy.
With your data.
Is that a fair assessment andfiguring out the data between
those two is often a challenge.
And now a word from one of oursponsors.
Every restaurant operatorunderstands the chaos of a
Restaurant kitchen during themeal rush restaurant
(18:05):
technologies, oil, total oilmanagement solutions, and end to
end automated oil managementsystem that delivers filters,
monitors, and recycles yourcooking oil, taking the dirtiest
jobs out of your kitchen andletting your employees focus on
more important tasks.
Control the kitchen chaos withrestaurant technologies and make
your kitchen safer.
No upfront costs to learn more,check out rti- inc.
(18:28):
com or call 888 796 4997.
Anand Tumuluru (18:34):
No, that's a totally fair assessment, Jeremy.
And the good news is that,again, just like with marketing,
third party marketplaces have alot of tools, have a lot of data
around this.
So what they do for therestaurant, barring a few
exceptions, there is alwaysexceptions where Michigan passes
a new law and, they're a coupleof months behind, barring sort
(18:55):
of those exceptions.
They do a phenomenal job ofputting estimating the tax
liability, both the taxes thatthey've withheld versus what you
need to be paying, to your stateauthorities very well at an
order level, at a transactionlevel or at a payout level.
The problem is that it's veryhard to actually integrate all
of that data in, in the contextof your point of sale.
(19:18):
Which has its own taxesconfigured and your accounting
software, which again has itsown rules and logic that's
applied to it.
The second thing is that there'sa whole set of decisions that a
brand needs to make, in termsof, pricing, right?
So if they're increasing prices,so what is it getting taxed on?
(19:38):
Or let's say they spend onmarketing.
they need to make sure thatthey're paying the taxes for the
right amount.
And that basically.
Listening to the marketplacesand taking that help in terms of
sizing their liabilitycorrectly.
a lot of them actually end uppaying double taxes, right?
They basically end up, just tobe considerate, just to be safe,
(19:59):
what they do is, Hey, I made 100on third party delivery.
I just need to pay 8%.
I'm just going to pay thatentire 8%.
Turns out 5 percent of that wasalready paid by the marketplace.
You're just supposed to paythree percent.
so what loop.
ai is helping do is not justunderstanding the tax
liabilities well and buildingthe order and audit trail for
(20:21):
you to understand what, what ledto a certain calculation of tax
liability.
We are actually automatingreconciliation and filing those
sort of tax entries in yourQuickBooks or in your Restaurant
365 directly.
So we help actually create thosejournal entries and we don't
give you another Excel sheet oranother dashboard for you to
(20:41):
figure out and navigate.
We actually go and close yourbooks for you, so that you don't
have to deal with the complexityof trying to break these nuances
down.
Jeremy (20:50):
I, I appreciate that.
And again, I, I've got a leadfor you.
After we get done with the call,I got a customer that I was just
on the phone with the other dayand he's can you help me fix my
tax thing?
So we can talk about thatoffline.
But, another line of thought Ihad that you talked about, I
think it's, Yeah.
Until you get into the data andyou start to look at it is each
of these delivery serviceproviders do things differently.
(21:12):
They give you data differently.
They give you your reportingdifferently.
They charge you differently.
They may have different rates,different times of day,
different rates.
Talk to me a little bit about,for those that just think, oh,
all delivery service providersare created equal, or they're at
a brand right now that only issupporting DoorDash, but they're
considering Uber Eats andGrubhub.
(21:34):
They all do it differently.
And because they all do itdifferently, reconciling them,
you have to build systems or,engage with something like loop
AI to understand what thosereconciliations are.
So can you walk me through that,that each one of them are their
own tech that's been createddifferently with its own set of
rules, its own set of outputs,its own set of dashboards, its
own set of those kinds ofthings.
But I think all too often, if Ihaven't dug into it, or they
(21:55):
say, that's my accountant'sproblem, it's not my problem.
I just need to offer Uber Eats,but those levels of complexity
that exist because every one ofthem is different.
Anand Tumuluru (22:03):
great question, Chadmi.
And I think.
first things first, I think theanswer to these questions, I'm
always afraid to indirectlyimply that my old employer, Uber
or Doordash, I'm not doing agood enough job, off of doing
these reports.
And I just want to preface thisentire conversation by saying, I
(22:24):
think we, we should, we shouldhave the understanding that this
entire industry was created likethree years ago.
Everybody's figuring everythingout, including the market cases,
right?
So they're figuring out theirown sort of profitability.
They're figuring out what theright team structures are, what
the right reporting standardsare.
And by and far.
The way that the big two,Doordash and Uber Eats, do all
(22:46):
of these transactions, payoutreporting, and helping
restaurants understand sort ofthe breakdowns, is, miles ahead
of, everybody else.
So they have, to their credit,invested quite a bit in
simplifying finances andoperations for restaurants.
there is no doubt about that.
Except, it's, It has a long wayto go, right?
(23:07):
there is still a lot of work todo.
The key problem here,Fundamentally Jeremy, is that
from Dotash's standpoint, theyare building their own ideal
standard and, it's perfect andit's beautiful, except a
restaurant is looking at 15 suchstatements.
They're looking at their POS.
They're looking at EasyGator.
They're looking at Grubhub.
And then they're also looking atLunchbox.
(23:28):
And then they're looking at,Olo, Olo and Thanks.
And at the end of the day, they,the amount of just spread that
they have with respect tounderstanding what was the
actual payment that they gotfrom any of these sort of
providers is a complex maze.
And we went from a world of Noteven having cloud based point of
(23:48):
sale to having 15 different sortof, digital ordering channels on
steroids, just pumping, all ofthese orders into the
restaurant.
So we just live in a world whereit is complex and overwhelming
for the restaurant.
Now, in terms of the, in termsof the complexity that exists
with understanding payouts,everybody's doing their own
thing.
(24:08):
First of all, there isn't like astandard or there isn't like a
rule set that says, This is howyou are, how marketplaces are
supposed to report, or these arethe fees that you need to
expect.
Atoms have been made atdifferent cities and, different,
different states and citieswhere they try to say, Hey,
here's how you can charge it.
But.
That only led to an explosion ofmore than items.
(24:29):
at the end of the, at the end ofthe payout.
So I think everybody's trying tofigure out what's the right way
to, what's the right way topresent it, present these
transactions and payouts.
Now, one of the things that Iwould like to point out, Jeremy,
is that in the world of thirdparty delivery.
The complexity is actually a lotmore than what you would have in
(24:50):
store.
for your regular POStransaction.
So for POS transaction, all youhave is you just go buy the food
and you basically have, let'ssay a 2 percent deduction on
your credit card, right?
But in the world of third partydelivery, you have marketing,
right?
Which is super important.
you want the visibility, youwant promotions.
You potentially have, refunds,which are also bundled in, for
(25:12):
what it's worth.
It's significantly larger thanwhat you would have in store,
fundamentally because there's alot more operational complexity
for taking food out.
But at the same time, it'sactually simpler than, handling
chargebacks is simpler than whatyou would have with your POS,
because with POS It's a lot morecomplex with POS, right?
And then you have the taxes,marketplace, withholding taxes.
(25:33):
And on top of all of this, youbasically have your first party,
delivery orders, which are fixedfee based, ordering channels.
So the complexity with respectto the payouts and transactions,
it's warranted, it's necessary.
It represents, The complexity ofthe channel and the way the
consumers are actually changingtheir behavior, right?
(25:54):
So it represents the wide arrayof things that you would do with
a regular e commerce store,right?
It's no different from what youwould do if you were selling
let's say an iPhone With your ecommerce on an e commerce store
or a bag like or Apple so Ithink there is a lot of
parallels to the world of ecommerce and There's a lot of
complexity that necessitatesrunning your digital delivery
(26:17):
channels well, except it'scompletely new.
Each of them is doing it intheir own way.
And at the end of the day, thefundamental question of, hey,
This is all great.
Like at the end of the day, whatis a gross profit?
Like how much am I actually likepaying?
that question becomes harder andharder to, answer for
restaurants without usingtechnology.
If you're basically, deployinganother person to sit down and
(26:38):
pull these reports, or if youare asking your, outsourced
accountant to do it, good luckwith that.
It's just going to take so muchtime, so much effort, and it's
going to be a one time thing.
just trying to figure out whatthese calculations are.
Jeremy (26:50):
and at the end of the day, in most brands, it doesn't
even represent 50 percent oftheir total volume.
And so you're now at, 20 percentof your volume is through DSPs,
but you've got this complexityof having to deal with this.
So without tech, the juiceoftentimes isn't worth the
squeeze.
And which is why I love, I loveaggregators, order aggregators.
We've had multiples on the show.
I love that they allow theorders to get from the web down
(27:12):
into their store.
So operational complexity, Ininside of the four walls is
great.
And now we've got tools likewhat you've got that are going
to help people reconcile.
Am I making money?
How am I making money?
and really even be able to makebusiness decisions.
And so I'd love to take thatpath because again, Not all
deliver service providers arecreated equal.
(27:33):
Not all of them are going to dothe things that you might want
to deliver on your brand promiseto people.
And depending upon what younegotiated and how long your
term is on your contract, youmay be paying more to one of the
delivery service providers, butthey deliver a higher level of
service.
And maybe the guests are willingto pay more, but you don't know
(27:53):
that unless you have the data tobe able to make those decisions.
So talk me through what all ofthat looks like, because I think
all too often people, I ha I hada customer as silly as this
sounds.
I had a customer recently thatsaid, I can charge more on my
third party delivery than I doin store.
And I looked at them and said,absolutely.
You should, because it's ahigher value to that customer.
(28:13):
Yes.
It's the same cheeseburger.
Yes.
It's the same chicken sandwich,but you're delivering your costs
are higher.
And, all of it.
And so across the board, talk methrough how you guys think about
that and how the data that youguys serve up gives people the
capability to deliver on thosepromises in different ways that,
that they hadn't consideredbefore.
Potentially.
Anand Tumuluru (28:32):
Absolutely.
Absolutely.
So if you, there are a couple ofwhat you might think, think of
as ground rules here, Jeremy.
So the first one is that thecustomer that's coming to
marketplaces, those marketplacesworked hard to get them right.
They have billions of dollars ofspent in terms of.
Demand generation on Google, on,on social media, their own
Jeremy (28:57):
they're the ones that are at the top of the list.
When you search for starboard,They show up door dash might
show up before the actualcustomer's website because
they're trying to drive thatorder through there just to,
create a direct correlation ofwhat you're saying.
They're putting the ad in Googlewhen you go search or when you
go search somewhere, they'reputting that at the top of your
search results that you're goingto door dash thinking that you
(29:18):
might be ordering directly fromthat website.
I apologize.
I didn't mean to cut you off,but I think it's, I think it's
important for people tounderstand that.
Yeah.
Anand Tumuluru (29:25):
and I think Google really, is really going
in the right direction there.
again, we're all figuring thisout.
So Google sort of made thisrecent change where they, where
they let the restaurant chooseif that Doordash or, Uber Eats
store is linked.
but a lot of that behavior.
consumer behavior, especiallythe strongest cohorts of
consumers with thesemarketplaces don't even start
(29:46):
from Google.
They are subscription customersof Doordash.
Of Uber Eats, they're openingthat app in order to search for
food in the first place.
So the food search is not goingto Yelp or Google.
It's actually going directly tothe marketplace.
very similar to product searchon Amazon.
There was a time when peopleused to search on Google.
It's Hey, can I buy my new top,best binoculars?
(30:10):
Under 100, right?
That's what you'd search for onGoogle.
People don't do that anymore.
People actually go to Amazondirectly and search for it.
So that behavioral shift ishappening even with food.
especially here in the Bay Areawhere DoDash is super, super
strong.
People just directly search onDoDash for, best Thai cuisine.
Best Thai food nearby.
Best Pad Thai nearby.
and regardless of whether,whether it is through loyalty,
(30:33):
through search, thesemarketplaces are investing in
that consumer and they arebasically keeping that consumer
information close to theirchest.
They are keeping all of theconsumer's history, all of
their, predictions around thecustomer's behavior in the
future.
And then, and guess what?
They're actually doing aphenomenal job at getting the
(30:54):
customers to reorder.
a couple of years ago, I
Jeremy (30:56):
push market to me all the time.
They're like, Hey, you haven'tordered from this brand in a
certain amount of time, I don'thave the brands app downloaded,
but I have DoorDash downloadedbecause it's part of my dash
pass as part of my chase card.
And so I order off a door.
this is a real life consumerthat's in this space.
I know exactly what they'redoing, but almost every day I
get a push notification sayingyou haven't ordered from X, Y, Z
(31:17):
brand.
It doesn't matter the name ofthe brand.
What would it look like if Igave you 10 percent off an order
today to drive that behavior?
And it now has me thinking aboutthat.
And I might not take that offernow, but the next time I go to
think about it, it's like, Hey,I haven't had Pad Thai in a
while.
Maybe I'll go to DoorDash thisPad Thai.
So these are real life exampleswhere they're in your customer's
(31:39):
inbox.
They're in their.
apps on their phones or in theiremail inbox, oftentimes more
often than you are to try anddrive that behavior.
So I apologize.
I keep cutting you off, but Ilike, I think that it's critical
for people because all too oftenrestaurant owners, when I sit
and listen to them, they don'tuse these apps because they want
to go experience, they gotanother hospital industry, so
(32:01):
they're not the consumer.
But they need to understand thatthe consumers that are out there
that are using these apps Aredoing these things and it looks
different So i'll let you keepgoing because I think it's
critical for them to understanddiscovery And behavior and them
driving that to your brand andwhat that means for you.
Cause you likely wouldn't havegotten that customer without it.
(32:22):
And now a word from oursponsors.
Growing your business can meanbig time logistical questions
like, how am I going to keep upwith all of these local
deliveries?
Let UberDirect offer you ahelping hand.
With UberDirect, you can takeorders directly from your
website, app, or by phone.
Then drivers who are part ofUber's courier network will
deliver them to your customer'sdoorsteps.
Sound too simple to be true?
(32:43):
Check out uberdirect.
com to learn more.
Anand Tumuluru (32:45):
And no, please keep adding your insights and
thoughts too.
I think it's, I think theimportant thing here is that
there is a behavioral shiftthat's going on with consumers.
in the ways that you justdescribed, which is, which is
hard to understand, which ishard to internalize or build an
intuition around, especially ifyou're a restauranteur, if
you're a restauranteur, you'rebuilt to be a person that's
(33:07):
extremely hospitable to a personthat's in person.
You want to give them the bestservice.
You want to give them the bestfood.
You want to make them smile.
You want to give them a greatexperience.
You're not necessarily thinkingin the world of sort of bits and
digits, right?
I think it has to be said morenumber of times to really
internalize The consumerbehavior shifts that are
happening, shifts that arehappening.
(33:28):
Jeremy, just to your point, justto close that, the funnel, so to
speak, with respect to consumerbehavior.
I think, just a couple of yearsago, I think, in the upcoming
Food On Demand conference, thetwo years ago old version of
this, somebody mentioned thatDash fast consumer, consumers
are ordering on DoDash seventimes a month.
I wouldn't be surprised ifthat's eight, nine, or ten now.
(33:51):
Which means consumers areordering on DoorDash two or
three times a week.
My question is, how long beforeit's 21 times a week, basically
what if it's all the three mealsevery day, right?
So what if they're actuallygoing into apps?
What if most of these behaviormost of the behavior for
customers is like fully digitalAnd maybe it's not 21.
(34:15):
How about right?
That's 5x thing of this marketAnd by the way, that's the
default behavior in a lot ofOther, other cities and
countries outside and it's goingto vary a lot by region It's a
growing, it's a growing customerbase and we are seeing that
growth in marketplaces while thegrowth rate has slowed down It
(34:36):
is still growing very stronglyand we are seeing this across
brands too that delivery revenueis growing, right?
Once a customer comes into theapp they can actually place an
order sometimes in one clicksometimes in two clicks You And
they're able to actually requesta refund or seek help in a
couple, a couple of clicks, theycan actually go ahead and.
(35:00):
Look at that history of order.
So let the app recommend thebest meal that they can have in
one or two clicks.
So for them, for these apps, Ithink the reality is that
consumers are sticking reallyhard to these apps.
They are ordering again andrestaurants need to go and meet
them where they are a lot ofbrands Jeremy to your point
(35:21):
earlier ask us the question,right?
So the so when if I answer yourquestion, hey, how you know
What's the profitability is thisrevenue incrementally fee?
If you get an answer to thatquestion, maybe you're hinting
towards how about we just run anoffer towards our first party
app and get all of theseconsumers out and, acquire them
directly on the brand.
(35:42):
We have seen that story play outa few times and the success
rates have been different acrossdifferent brands.
But the reality is thatespecially not with pickup
orders, but with deliveryorders, The apps are so much
more convenient, the marketplaceapps are so much more convenient
that we inevitably see customersgo into the brand's app and then
come back at a later point oftime once we saw offered ease
(36:05):
out.
And especially the subscriptioncustomers that are extremely
sticky, right?
They keep going back to thethird party apps.
So it's this behavior.
it's just consumer behavior.
So the reason I'm giving you thecontext around the consumer
behavior is that the revenuethat marketplaces are generating
for restaurant in the debate ofwhether it's incremental or not,
(36:25):
it's really incremental.
There is just a ton of anecdotalevidence for brands that have
tried to take these customersaway and have seen those
customers go back.
Plus, the subscription behaviorthat's Important for us all to
understand and to know thatinvesting in these channels is
important.
So once you have that, once yourecognize that cc data and
(36:47):
anecdotes and hear from otherpeople's experiences, the next
question is, Hey, how muchshould I invest here?
What's the, how much of my brandshould be present on these third
party apps?
How much of my revenue should Iexpect?
Should it be 15%?
Should it be 30%?
How much should I invest in afirst party channel?
And how much should I drive myrevenue there?
Those are the things we all haveto figure out in the next two,
(37:08):
three years.
Every brand has to figure out.
Should I invest, let's say, 100,000 in an app so that I get at
least 10 percent of thecustomers into a first party
ordering experience?
Should it be limited to pickups?
We see a lot of brands, forinstance, Dave's Hot Chicken,
one of our favorite brands.
They don't do delivery on theirapp.
They just do pickups, right?
They don't deal with thedelivery experience at all
(37:29):
because they know that, thirdparty apps, the way that they
offer that convenience, itworks, right?
It works.
That's where customers aregoing.
They're happy with it, but thenthey have a phenomenal, app for
pickups.
So all of these decisions interms of what should my channel
mix be, In terms of third partyversus first party, pickup
versus delivery.
(37:49):
How much marketing should Ispend?
These are decisions that brandbrands need to make, in, in the
next one, two, three years,
Jeremy (37:56):
I think it's, I think it's ironic cause we'd never put
a New York strip on our menuwithout understanding what my
food cost was and what it was acost to acquire that customer,
but we just make these guesseson what the customers are going
to be able to bear on thirdparty delivery apps until tools
like you guys have.
And it's just, It baffles me.
And to your point, it's only atwo or three year old problem.
So I think as more and more databecomes available, as these DSP
(38:19):
solution providers end up givingmore data back to the
restaurants to understand thesethings, and then there's tools
like what you guys have, we'llcontinue to get better.
even the delivery experience nowversus what it was two years ago
is Phenomenally better.
And it's gonna only continue toincrease.
and so talk me through.
What does it look like to engagewith loop AI?
What?
What's the ideal customer?
How do they engage with you?
(38:39):
How do they learn more about howyou can help them to understand
the profitability and truly makegood science and data based
decisions about their deliveryservice?
Partners and how they're goingto go into market.
Anand Tumuluru (38:52):
if you basically have a delivery revenue
contribution, that's 20 percentor a hundred thousand dollars
that's coming through from thirdparty marketplaces, loop, loop
AI will be a great fit, for youand fundamentally we help, on
onboarding, the brand for us isvery straightforward.
it takes a few minutes for themto sign up, get loop AI set up.
(39:13):
The first thing that we alwaysrecommend restaurants do is look
at their income statement andlook at all of the line items
where the fees are gettingdeducted, where, where they're
making money or.
Where they're not, and have astrategy, have a plan towards
attacking those different lineitems.
And once we set those goals, wesee returns or improvement in
(39:34):
the profitability as early asone week.
And definitely over a span ofthree months, for instance, with
Craveworthy Brands, one of ourfavorite customers, we've
reduced that downtime from 300minutes a month to 100 minutes.
while they were going through aPOS integration again, one of
those things very hard to knowthat your stores are not even
open for customers to place anorder.
it's just a hidden problem, wecan go solve for that.
(39:54):
Another, another example is,Hey, we are mandated to spend.
According to our marketplacecontract, we should spend a
million dollars in marketing.
We are worried we are givingaway free food.
We are worried we are spendingthat money inefficiently.
how can we understand whatcampaigns are working well for
us?
And we, deploy the cash in a waywhere we can actually grow our
customer base.
(40:15):
another area where, you know,Duke AI can help.
And I think the, our favoriteuser base are the finance teams.
CFOs who, who know the pain or,even outsourced, CPAs who know
the pain of actually pulling allof these reports, understanding
the different formats, keepingup with the reporting format
changes and actually, closingtheir books by creating these
adjustments.
(40:35):
All of that's gone.
once you're on board to loop AI,we basically take care of all of
that reconciliation for you.
the, we truly understand theback office complexity of.
Managing third party deliveryand we basically make sure that
we touch all bases with respectto the burden of getting this
revenue and doing delivery,right?
you know for your team
Jeremy (40:56):
thank you for one, for creating the product.
Cause it is a problem thateverybody that has delivery has
this.
And so the fact that you guysare at a place that you guys are
solving for this is fantastic.
and again, I think, as youcontinue to remind me, it's only
a.
Four or five year old problem atthe most, it's this problem that
has been, we all feel like it'sbeen forever.
Cause you know, I don't know,COVID created, everything is
(41:16):
like in dog years nowadays.
at the end of the day, we'redoing so much of this and it's
so painful.
So I'm grateful for tools likeyours and others that are out
there that are helping solvethis problem.
And so thank you for sharing.
to our listeners, guys, like Isaid, at the onset, we love
having you guys listen to theshow, give us some feedback, go
check out loop AI.
If you haven't alreadysubscribed to the show, whether
that be on YouTube or on yourfavorite podcast player, please
(41:39):
do so and make it a great day.
Thanks for listening to theRestaurant Technology Guys
podcast.
Visit www.
RestaurantTechnologyGuys.
com for tips, industry insights,and more to help you run your
restaurant better.
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