March 7, 2024 • 38 mins
Unlock the secrets of "Fusion Strategy," the groundbreaking approach that's rewriting the playbook for industrial giants. On our latest episode, we sit down with Venkat Venkatraman, author of the upcoming book "Fusion Strategy: How Real-Time Data and AI Will Power the Industrial Future" who, along with co-author Vijay Govindarajan, is steering companies through a landscape where AI and real-time data are as critical as nuts and bolts. Venkat maps out for our listeners the fascinating terrain where digital and physical realms converge, elucidating how heavyweights in sectors like automotive and agriculture can not only survive but thrive by adopting innovative digital tools to gain actionable insights and drive strategic decisions.
Navigating the future of industry requires a skillful blend of technology and vision, and in this conversation, we dissect the role of AI in transforming traditional industries from the inside out. We scrutinize the "computer on wheels" phenomenon and discuss how companies that fail to integrate AI, data, and cloud connectivity might find themselves in the technological rearview. Moreover, Venkat shares his unique perspective on 'backcasting' and the imperative for a harmonized approach among company leaders to overhaul business strategies, ensuring longevity in an ever-evolving digital age. Join us for an episode that promises not just to inform but to equip you with the strategic foresight needed for the dynamic road ahead.
"Fusion Strategy: How Real-Time Data and AI Will Power the Industrial Future" comes out March 12, 2024 and available for preorder now.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
J.P. Matychak (00:20):
Greetings everyone and welcome to another
episode of the Insights atQuestrom podcast.
I'm JP Matychak and alongsideme is my co-host, Shannon Light.
Shannon, how are you?
Great?
Thanks, JP.
It helps if I bring you up whenthe volume there you doing well
, I'm doing well, great, well,today is another installment in
a segment we like to call theQuestrom Book Club, where we
(00:42):
feature recent and upcomingbooks authored by Questrom
School of Business Faculty, andtoday we welcome Venkat
Venkatraman, the David JMcGrath, Jr.
Professor of Management inInformation Systems.
He's considered one of theforemost authorities on how
companies develop strategies towin with digital technologies,
widely published in leadingacademic and practitioner
(01:04):
journals, he has been recognizedas a top-sided researcher by
Google, scholar and StanfordUniversity.
He's lectured and conducteddigital strategy workshops all
over the world with companiessuch as IBM, merck, ericsson, bp
and many more.
He's also the author of the"Digital Matrix New Rules for
Business Transformation throughtechnology.
And today we're discussing hisupcoming book Fusion Strategy
(01:27):
how Real-Time Data in AI WillPower the Industrial Future,
co-authored with world-renownedinnovation guru Vijay
Govindarajan, a playbook thatwill help industrial companies
combine what they do best tocreate physical products with
what digital do best.
Use algorithms in AI to parseexpansive interconnected data
sets to make strategicconnections that would otherwise
(01:49):
be impossible.
Venkat, welcome to the show.
Venkat Venkatraman (01:53):
Thank you very much.
J.P. Matychak (01:54):
So let's start by setting the stage a little bit.
It's hard not to read anythingtoday and not see news about
data, big data, AI and for themost part, I think people see
those two things as onlyaffecting the digital economy.
But judging by the title of thebook, it sounds like we should
(02:14):
be preparing for data and AI toimpact the entire industrial
economy.
Venkat Venkatraman (02:20):
You're absolutely right, J.
P.
If you think about the timethat I wrote the digital matrix
in 2017, the big question atthat time was how will digital
giants and the startups start toinfluence traditional
industries as they becomedigital?
And most of the examples that Istudied at that time were
(02:44):
digital economy companies music,media, advertising, telecom,
financial services.
These are asset light sectorswhere the analog products could
become digital and we don't needto actually have an analog form
of those products.
(03:06):
When I finished the book and Istarted discussing the ideas
with my colleague VG atDartmouth, we realized that the
industrial sectors have largelybeen untouched by digital
technologies for so long,because you still have to make
cars, you still have to makeaircraft engines, you still have
to make aircrafts and buildingsand tractors and so on.
(03:30):
But we asked the questionthere's a $75 trillion of GDP
that's the industrial sectorthat's largely been untouched.
What would happen if wedigitize products?
What would happen if weactually have real-time insights
into how your car is driven onthe road?
What will happen if you havereal-time insights to your
(03:53):
tractors as they go aboutplowing in different farms in
the world?
And so we started connectingthe digital economy to the
physical economy and came upwith the idea of fusion, which
is the fusion not in the senseof energy sector, but fusion in
the sense of infusing physicaland digital.
(04:16):
And so we believe that the nextdecade is really about digital
transformation of asset heavysectors where physical products
are going to become digital,physical products, and we have
already seen it.
Look at today's speakers,yesterday's output devices, that
(04:37):
today's input output devicesthat are essentially digital
physical products, and we can goon and talk about many other
things as we go along in thispodcast.
J.P. Matychak (04:47):
So what inspired this exploration, then, of
thinking about strategy andreal-time decision-making, and
that the impact that AI and datacould have on this aspect of
business?
Venkat Venkatraman (05:02):
Yeah, again, we went back to the industrial
companies and asked ourselves aquestion what made them
successful?
They have historically spent alot of time designing their
product and making sure thatit's economically viable at a
price point that the customerswill pay, and then they had a
very efficient supply chain.
Things that they did inside thecompany extended through the
(05:26):
global supply chains and oncethey delivered the product, they
had very little information onhow the product actually
performed on the field becausewe didn't add telematic
capability.
We didn't have real-timetracking capabilities and if you
take an automobile as a classicexample, where the automobile
is a consumer product but thedata about the product ended at
(05:53):
the level of the dealer, so a GMor a Ford or a Mercedes would
typically know where the car wassold, not to whom the car was
sold often, and they neverreally had data on who actually
drove the car and how the carperformed on the road.
(06:14):
Now, with telematics, we havereal-time capabilities in terms
of what's happening on the field.
So the shift in thinking forindustrial companies and
strategy is to go away fromthinking about your competencies
as products are designed orproducts are designed and
manufactured, but products asdriven or products as used by
(06:38):
the customers, and once I knowwhy your car fails, I can then
understand whether yours is anoutlier or is likely to be a
pattern that we can begin togeneralize so that we can take
proactive corrective action.
We couldn't do that before.
Sensors were expensive,software was something that was
(07:00):
hard-coded into the product, notreal-time Cloud connectivity
the 5G connectivity was notthere.
So when I did the research in2016 for the previous book,
these were ideas that we couldsee happening in the future, and
by 2023-2024, they become real,and by the time we come to the
end of the decade, this willbecome commonplace.
Shannon Light (07:22):
And to that point , how did these emerging
technologies impact fusionstrategy, as we continue to see
how much growth, even just inmonths of AI, yeah, that's
actually a great question If youthink about.
Venkat Venkatraman (07:40):
Let me take an example again from Digital
Matrix and connect it.
One of the examples I used inthe digital matrix was IBM
Watson and the ability of IBM tocreate cognitive computing,
which is really a precursor toall the generative AI that we
see today.
But at that time, what IBM hadto do was to go inside hospitals
(08:03):
, collect all the detailed datato digitize their understanding
of cancer cure or heart diseasecure or whatever the case might
have been.
At that time we didn't have thelarge foundation models and
then suddenly chat GPT came withthe foundation models and then
we can now create real-time dataand AI that sit on top of these
(08:28):
foundation models.
So we couldn't have really beenthinking about fusion
strategies in 2020 unlessindividual sectors started to
digitize all their content notjust structured data, but
unstructured data and very fewindustries did that on a
systematic basis.
So today, in 2024, we canreasonably speculate on why the
(08:54):
different industrial sectors aregoing to develop their large
industrial foundation models,and that is going to be
extremely valuable for them tothen develop their fusion
strategies.
So that's just simply onetechnology.
The second one is really cloudand the 5G or 6G connectivity.
(09:17):
So it's now possible to thinkabout putting a 5G or a 6G modem
on remote devices and collectreal-time data on how they're
performing on the field.
We couldn't do that five yearsback.
We didn't have this widespreadcloud capability or the cellular
capability, and then we got,obviously, the software
(09:38):
capabilities and theapplications, the development
capabilities and all that is nowpossible.
So multiple technologies areconverging to make fusion real.
Multiple technologies areconverging to make fusion
economically feasible for manycompanies to explore.
The real battleground is goingto be how companies are going to
(09:59):
invest to accelerate towardsthis fusion future that we
believe will be the definingactivities for many, many
sectors.
Shannon Light (10:10):
Would you say that that adoption of fusion
strategy is likely the biggestchallenge that a lot of these
industries face?
Venkat Venkatraman (10:21):
Yeah, I think again.
When we finished the book aboutsix to eight months back, we
were in presenting the ideas tovery many companies and they're
beginning to see the rationaleof shifting from industrial age
analog thinking towards thisdigital, real-time fusion
(10:43):
thinking.
And as companies begin torealize that if they don't do
that, somebody else is going tocome between them and the
customers, they will put thosesensors, they will reverse
engineer the process, they willdeliver new value-added services
that otherwise legitimatelyshould have gone to the
(11:05):
companies that make the product.
And so once they realize thatyou need to do this, both for
offensive, proactive reasons, aswell as for defensive, reactive
reasons, then the companiesrealize that they need to
actually focus on this much moreseriously than they might have
done in the past.
J.P. Matychak (11:23):
So what do you think are some of the roadblocks
that stand in the way oforganizations transitioning into
this fusion strategy, and howmight they be able to overcome
them?
Venkat Venkatraman (11:32):
Yeah, I think there are many, many
barriers or factors that inhibitcompanies make the transition.
I think the most importantissue is companies believing
that the future success is goingto be very different than their
(11:52):
past success.
If companies have beenextremely successful in their
development, then they look atthe digital as a 10-year journey
and they don't quite see thatexponential shift happening in
their visible horizon.
Then they can say I'll wait.
I'll wait for a few more years,because in technologies the one
(12:16):
area that I've studied for thebetter part of 35 years the one
common equation is the waitingequation.
Do I invest today or do I waitand invest tomorrow?
Because I know that if I waitand invest tomorrow, much of the
uncertainties can be resolved.
Costs of technology will comedown, there'll be more options
(12:39):
for me to play with, but at thesame time I lose the opportunity
to have a first mover advantage.
So the companies are nowbeginning to ask the question
what's the value of waiting andwhat's the risk of waiting?
And so that's the first andforemost barrier.
The second one is really one ofinvestment shifts.
Companies know how to invest inwhat they're good at.
J.P. Matychak (13:03):
Right.
Venkat Venkatraman (13:04):
And successful companies fail
because they overinvest in whatthey're good at today and
underinvest in what they need tobe good at tomorrow.
So who brings this new thinkinginto the company?
Is it the CIO?
The answer typically is not,because the CIO quite doesn't
understand what a product does.
(13:24):
The CIO may understand what aprocess does and what set of
organization changes need tohappen.
So it's got to be themanufacturing, it's got to be
the marketing person, it's gotto be the operations person, who
may have competencies in theirbranches of industrial
engineering and manufacturingengineering.
(13:46):
But here we are really askingthem to take the role of being
the hybrid managers thatunderstand manufacturing
engineering and data sciences,industrial engineering and
information engineering.
So once these hybrid managerssee the potential, then I think
making the investment cases hasbeen easier.
J.P. Matychak (14:06):
And do you see this move to a fusion strategy
and even data and AI and thismay not get touched onto the
book, but it just kind of sparksthe question Do you see this as
having an impact in creating alarger gap in competition
(14:27):
between companies or do youthink that this has the
opportunity to close the gap ofcompetition?
And, to say it another way,I've even noticed, with the
adoption of AI technologies, thechat GPDs, the Google Gemini's
of the world and whatnot smallercompanies are able to process
(14:48):
things that the likes that majororganizations like McKinsey
would be able to do with hoursand hours of people power and
human capital.
Do you see this fusion strategyas a way to create a more level
playing field for competition?
Venkat Venkatraman (15:06):
I think it's difficult to generalize at this
point in time.
But if you now take yourquestion and start asking can a
new company design a tractorthat has got data and AI
capability at the core, whereit's able to show real-time data
of what the tractor is doing onthe field and Then design it,
(15:30):
manufacture it and deploy it atscale over the next ten years?
The answer is yes.
Why?
Because we can look back andTesla mm-hmm as a relatively new
company that had a differentvision of what a car is going to
be.
Mm-hmm right computer on wheelsconnected to the cloud, with
real-time data and AI, and Allthe incumbents laughed at the
(15:54):
company because they saw it asessentially a faster golf cart.
You're right right, becausenobody actually understood the
entertainment center.
Nobody actually understood thatit's possible to have an battery
electric vehicle that has thesame power as an industrial
(16:15):
combustion engine.
But then, once you have that,you can add a lot more things to
that piece of equipment.
Computer on wheels connected tothe cloud.
A tractor, essentially, canbecome computers on wheels
Connected to the cloud, butexcept it's in a more regulated
setting, if you will, not on,you know, wide roads, but in
(16:37):
defined farms.
And If John Deere doesn't do it, if caterpillar doesn't do it,
if Mahindra tractors don't do it, somebody else will begin to do
that right?
So in that sense, it's possiblefor a new entrepreneur to think
about creating this.
But the reason why I said Idon't want to generalize is
because it will require asignificant amount of investment
(16:58):
right, to actually create thatproduct.
And this is where we wanted tocontrast between the digital
transformation of the assetlight sector and Digital
transformation of the assetheavy sector.
In the asset light sector, onceyou create an app, the marginal
cost of producing the next appis near zero, and we had this
(17:22):
Relatively scalableInfrastructure for distribution
of the apps right Apple AppStore and Android Play Store
right.
Here it's going to be much moredifficult, right?
You got to design the product,you got to develop the product.
You got distribute the product.
You got to maintain the product.
So we will see it happening insectors where the industrial
(17:42):
companies are asleep at theswitch, where the new companies
come in.
But the generalization that wewant the listeners to take away
from that is a digitalindustrial company that
understands the power ofInfusing mechanical engineering
with data and AI Will do muchbetter than a mechanical
(18:02):
engineering company that doesn'tunderstand the power of the
data and AI.
And the same argument holdsgood for any human that is able
to take advantage of GenerativeAI will do better than someone
who still thinks that gen AI issimply for the, for the, you
know, for play and not for work.
Right, and so that's where theshift is going to happen.
Shannon Light (18:25):
And I mean all of that seems to play a big role
in just the long-term planningfor a company.
So, to that point, withCompanies were not going to, you
know, be willing to adopt this,do you see the potential of
them?
Just you know that companycould just being completely
(18:48):
unsuccessful, or what's yourtake on?
Venkat Venkatraman (18:51):
that.
Yeah, I you know.
Obviously we are all biased inhow we think about the
prescription coming out of ourown research, I think, our
General prince prescriptioncoming out of our researchers.
If companies believe that thenext five years are going to be
exactly like the last five years, they will find themselves at a
(19:15):
comparative disadvantageCompared to companies that are
prepared to look at the futureas being very different.
And we also write in our bookthis idea of the distinction
between forecasting and backcasting.
Forecasting is projecting fromtoday into the future, where we
typically assume that thingsthat we want to happen In the
(19:38):
future will continue to happenbecause that's a future we like,
because we'll be successful inthat future.
Backcasting really challengesthe company to take a position
in 2030 and say what will thefuture be in 2030?
Not seen from our lens, butfrom the lens of the customers
that are going to use yourproduct.
(19:59):
What will have to be true forthat future to become real?
And In that future, will mycompany be relevant?
Will my competencies berelevant?
Will I still have a role toplay in get a fair share of that
value in that future?
So the moment we take anoutside in perspective, future
(20:20):
backward perspective rather thaninside out and Project from
today.
Companies start to realize thatmuch of the competencies that
they've been good at today Maynot be relevant in the future.
Right, because corecompetencies are not about what
you're good at today.
Core competencies are alwaysabout will customers give you a
(20:41):
premium For you being good atthat?
So I may be good at many thingsand and today I may be able to
extract value from the, from thecustomers, but that doesn't
necessarily guarantee that thecustomers tomorrow will give me
that same value, but even afraction of that value, in the
future.
J.P. Matychak (21:01):
So this obviously has implications up and down an
organizational structure, youknow, especially for those that
are, you know, traditionalproducts and whatnot.
So what can leaders do?
That sort of see the vision ofthis fusion strategy but then
need to build that, thatgroundswell of support within an
(21:23):
organization that may have thethe inertia of change,
resistance against them, toreally kind of shift the focus.
Venkat Venkatraman (21:31):
Yeah, I think the CXO team has got to
take a coordination role ratherthan a delegation role.
Okay, because what has happened?
When we had the first few years, few decades, of digital
transformation, the CEOsrealized that we need to
transform our processes.
With big ERP systems and CRMsystems, you know the big SAP
(21:54):
implementation.
That's three years and fiveyears of implementation.
They made the investment andthen they delegated it down to
the CIO or the chief humanresource officers and chief
operating officers.
Now the fusion strategy isreally challenging the status
quo and, if you believe that thefuture is not an extrapolation
(22:14):
of the past, we now have torethink the data flows inside
the company, the data flows thatextend to my suppliers and to
my partners and to thedistributors, and have to
reimagine what the factory ofthe future will look like.
We got to reimagine what theproduct of the future will look
(22:35):
like.
We have to reimagine thecontrol center, the operation
center, that these data startsto flow back in.
Many of those things arerelatively new ideas.
Right, we saw with the digitaltechnology people started
tracking real-time data ofconversations about our products
(22:57):
in social media.
That was relatively new.
Suddenly, I need to know whatpeople are saying about my
product on, you know, google orFacebook or Twitter or you know
wherever else the case may be.
So they created listeningcenters, social media listening
centers.
The equivalent idea for productcompanies is to understand
real-time data of how myproducts are performing on the
(23:18):
field.
That requires data, not justsomething that the marketing
department does, but themarketing department shares with
the operations department, withthe procurement department and
with the suppliers.
So what we write in the book isthis idea called digital twin.
J.P. Matychak (23:36):
Okay.
Venkat Venkatraman (23:37):
Where the digital twin has been well
understood in terms of designinga product.
We think the digital twin has toextend beyond product design to
manufacturing, but alsoperformance.
Once I designed the digital twinof actual performance of the
product in the field, then I'llknow why the product fails in a
(24:02):
specific situation and byunderstanding across these
different settings I get a muchbetter insight into why my
product fails.
And that data, if it isfunctionally kept siloed, then
the marketing department willthen not share it with the
manufacturing department, ormanufacturing department may not
(24:23):
share it with the productdesign department and the
product design department maynot share it with the suppliers.
The moment we take the CXO andgive them the charter of
end-to-end visibility with thisreal-time data and AI, they
realize that this is a weaksignal that I'm observing in the
field.
That's going to be extremelyvaluable for me to make
(24:45):
proactive corrections ratherthan think of this as what we
call in the book systems ofrecords rather than systems of
data graphs.
And the systems of data graphsare really understanding how my
products perform in the field.
So we are introducing in thebook ideas and mechanisms by
which the senior leaders can putin place organizational
(25:09):
mechanisms by which they cancreate, this infusion of
physical and digital domains.
J.P. Matychak (25:15):
And it sounds like it really requires the true
breaking down of departmentaland functional silos in order to
be successful.
Venkat Venkatraman (25:24):
Yes, because successful companies have
relied on functional excellence.
They're good at R&D, they'regood at manufacturing, they're
good at marketing, andcross-functional excellence was
not something that was expectedexcept at the senior level, but
(25:47):
even then they relied onfunctional expertise to say if
something is an R&D problem, Irely on you, if something is a
marketing problem, I rely onsomebody else.
So now, by using the wordfusion, we are really beginning
to say that it's the infusion ofdisciplinary thinking, infusion
of mechanical engineering withdata, as I said earlier,
(26:10):
infusion of industrialengineering with information
sciences.
These are new ideas that willmake successful companies
differentiate from others thatstill are thinking about
functional excellence ratherthan cross-functional or fusion
thinking.
J.P. Matychak (26:33):
So let's talk about the book itself.
You both decided to write thisas a workbook almost right and a
guidebook for leaders toimplement this new way of
thinking.
Talk about your reasoningbehind that, and you know, in
(26:55):
particular, what strikes me andit's something that you said
earlier on in the conversationwas that, even since you
finished the book, to now lookhow the world has changed.
So you wrote this as aguidebook.
How are you planning on keepingthis updated for executives and
leaders to keep up with thetimes?
(27:17):
Because, I mean, we thoughttechnology was rapidly changing
before.
It seems exponentially faster.
Venkat Venkatraman (27:27):
Yeah, I think it's a limitation of the
book publishing cycle, if youwill right.
J.P. Matychak (27:33):
Right.
Venkat Venkatraman (27:35):
We finished.
The major drafting of the bookwas done by August of last year.
It goes through reviews.
It comes back with somecomments.
We might have revised it andbasically sent it in just around
Thanksgiving, Right, and thebook is coming out in March.
So that's at least six monthsof thinking.
(27:55):
And nothing has happened and youknow that's okay in many, many
disciplines, but it's not okayin a discipline that we are
really talking about.
That is changing pretty muchevery day, right?
So we did two things.
One is the book is coming outin March.
We, hopefully will have theplaybook with the templates that
(28:15):
will allow companies to thinkabout much of the ideas that
we've been talking about duringthis podcast.
Sometime in the AugustSeptember timeframe, the Harvard
Business Review Press willdecide how they want to
distribute the templates thatmake the playbook actionable and
, even though we write aboutmany of the actions in the book,
(28:35):
we'll provide some templates tomake that happen.
We also wanted to do something,which is to practice fusion in
terms of making physical digitalbook work.
Now, if you think about most ofthe books, there is a physical
book and there's a digital book,which is basically a static
(28:57):
version of the book, and thenthere's an audio book, which is
basically an audio version ofthe static book.
The form doesn't change thesubstance of what's in the book
and what we are now trying to dois to update the material since
the book is finished, expandthe material because, by design,
(29:23):
the book is 200 plus pages andwe may refer to a case in two
paragraphs when the richness ofthe research that we might have
done is maybe six to eight pages, but the actual material may be
a lot more than six, seven,eight pages.
So what we are doing is we'reactually creating a fusion
strategy coach.
(29:44):
Oh, interesting which will be aGen AI coach, which you think of
it as a virtual VG and virtualVenkat oh, where a reader can
potentially ask questions and,to the extent that we can train
the material to help you answerthat question.
We have gone beyond just simplygiving you a static book or a
(30:08):
digital book, and so at thispoint in time, we have tested it
out with about 40 executiveswho've had a chance to read the
early draft of the book and wehave asked them to just test out
the coach to see whether thecoach allows them to go beyond
what we have written in the book.
And we're getting good.
(30:29):
We have gotten very goodfeedback on the coach.
We're gonna test it out with agroup of about 100 students at
Tuck who is gonna take thestrategy class in March and
April.
We're gonna do the same thing.
They're gonna read the book,they're gonna use the coach and
then see if the by using theirprompts right we're not gonna
give them the prompts usingtheir prompts whether they're
(30:51):
able to get deeper insight andunderstand the connections
across the concept beyond whatthey would get from the book.
And that will also help usdevelop the playbook with the
templates and the prompts by thetime that comes out sometime
later this particular year.
We believe that the idea of areader posing a question that
(31:20):
the book is able to answer inthe language of the reader is
really where the book should go,and we're basically thinking of
this as the first stage of thistransformation, and what we
haven't quite decided is howfrequently we'll update it and
(31:44):
how will we host it, where willit be hosted, how long this will
be maintained, how to chargefor it.
All these are open questions,but we believe that we should at
least let the initial readers,the initial buyers of the book,
have an opportunity to askquestions that otherwise we
would have answered using anonline forum or using an email
(32:07):
forum.
But here we hope that we canget the Gen AI to provide first
level answers, and that's gonnabe our experiment with the
launch of the book.
Shannon Light (32:18):
Currently, are the those who have tested out
this coach?
Is that living on an onlineplatform where it's similar to
say, something like chat, gpt,where you type in it's at this
point?
Venkat Venkatraman (32:31):
in time, a proprietary platform in which we
give the link and we give thema password so that only they can
access it, because we don'twanna make it open at this point
in time.
It also raises interestingquestions about intellectual
property Because which is mynext question Absolutely the
(32:54):
book is copyrighted by theHarvard Business Tribunal.
So we are still working throughall those issues.
But the executives then had achance to then test out these
ideas.
Right, they could ask abouttheir company.
Their company might not havebeen done a discussion in our
(33:15):
book.
And then we are seeing how wellthe coach answers that question
and the way we have trained itis actually looking back.
It is very simple, but westruggled through it for a very
long time because we wanted tomake it a simple coach, without
too many options given to thereader, and finally we came down
(33:37):
with a simple idea of what I doin the classroom, right when
somebody asked me a question.
What I do in the classroom isconnected back to what I know
and connected back to what thestudents know.
So what we have trained thecoaches to do the following you
ask a question.
It refers back to the coreideas from the book.
(33:57):
It doesn't say chapter five ordoesn't say chapter seven,
because it synthesized acrosschapters.
It writes an answer in about250 words and, in making it
conversational what it does, itends the answer with two
questions for you.
One that goes deeper into thequestion you asked me, which is
(34:22):
what I would do in the classroom.
Right, you're asking me, I'llexpand more on it.
The second question is tolaterally connect across
concepts or across industries.
So, if the question is aboutfarming, one follow-on question
may be how does this idea applyin transportation or in mining
or in construction?
(34:42):
And if you asked me aboutfarming, the second question may
go deeper into the farming howdo the digital twins apply in
farming?
How do we begin to think aboutsystems of data graphs in
farming?
Right, because that's thelogical connection by creating
this Socratic thinking, which iswhat we do in the classroom,
(35:05):
into the Gen AI, we are able togeneralize and have one general
coach that will continue to haveconversation with you till you
find that you've gotten thelevel of answer you can get from
this.
So so far, it's text-based.
It's not simulating my voice.
I'm not trying to create a GenAI version of my answer using my
(35:29):
voice.
It's text-based, but these areeasy extensions we could do.
We could do an AI avatar withmy voice, an AI avatar with my
video, but first we want to makesure that the concepts work,
and that's what we're doing sofar.
J.P. Matychak (35:45):
Wow, truly an opportunity to both transform
the business world and how we dobusiness, but also the
potential to transform how welearn as well.
Incredible.
Venkat Venkatraman is the DavidJ McGrath, Jr.
Professor in InformationSystems at the Boston University
Questrom School of Business.
(36:05):
His upcoming book FusionStrategy how Real Data in AI
Will Power the Industrial Future, is coming out on March 12th
2024.
In is available for pre-ordernow on Amazon.
Venkat, thank you for joiningus.
This has been incrediblyinteresting.
Venkat Venkatraman (36:22):
Thank you very much for having me.
J.P. Matychak (36:26):
Well, that will wrap things up for this episode
of the Insights at QuestromPodcast.
I'd like to thank our guestagain, Venkat Venkatraman.
Remember for more informationon this and previous episodes,
along with other insights fromQuestrom experts, visit us at
insights.
bu.
edu.
For Shannon Light.
I'm JP Matychak, and so long.
Greetings everyone and welcome to another
episode of the Insights atQuestrom podcast.
I'm JP Matychak and alongsideme is my co-host, Shannon Light.
Shannon, how are you?
Great?
Thanks, JP.
It helps if I bring you up whenthe volume there you doing well
, I'm doing well, great, well,today is another installment in
a segment we like to call theQuestrom Book Club, where we
(00:42):
feature recent and upcomingbooks authored by Questrom
School of Business Faculty, andtoday we welcome Venkat
Venkatraman, the David JMcGrath, Jr.
Professor of Management inInformation Systems.
He's considered one of theforemost authorities on how
companies develop strategies towin with digital technologies,
widely published in leadingacademic and practitioner
(01:04):
journals, he has been recognizedas a top-sided researcher by
Google, scholar and StanfordUniversity.
He's lectured and conducteddigital strategy workshops all
over the world with companiessuch as IBM, merck, ericsson, bp
and many more.
He's also the author of the"Digital Matrix New Rules for
Business Transformation throughtechnology.
And today we're discussing hisupcoming book Fusion Strategy
(01:27):
how Real-Time Data in AI WillPower the Industrial Future,
co-authored with world-renownedinnovation guru Vijay
Govindarajan, a playbook thatwill help industrial companies
combine what they do best tocreate physical products with
what digital do best.
Use algorithms in AI to parseexpansive interconnected data
sets to make strategicconnections that would otherwise
(01:49):
be impossible.
Venkat, welcome to the show.
Venkat Venkatraman (01:53):
Thank you very much.
J.P. Matychak (01:54):
So let's start by setting the stage a little bit.
It's hard not to read anythingtoday and not see news about
data, big data, AI and for themost part, I think people see
those two things as onlyaffecting the digital economy.
But judging by the title of thebook, it sounds like we should
(02:14):
be preparing for data and AI toimpact the entire industrial
economy.
Venkat Venkatraman (02:20):
You're absolutely right, J.
P.
If you think about the timethat I wrote the digital matrix
in 2017, the big question atthat time was how will digital
giants and the startups start toinfluence traditional
industries as they becomedigital?
And most of the examples that Istudied at that time were
(02:44):
digital economy companies music,media, advertising, telecom,
financial services.
These are asset light sectorswhere the analog products could
become digital and we don't needto actually have an analog form
of those products.
(03:06):
When I finished the book and Istarted discussing the ideas
with my colleague VG atDartmouth, we realized that the
industrial sectors have largelybeen untouched by digital
technologies for so long,because you still have to make
cars, you still have to makeaircraft engines, you still have
to make aircrafts and buildingsand tractors and so on.
(03:30):
But we asked the questionthere's a $75 trillion of GDP
that's the industrial sectorthat's largely been untouched.
What would happen if wedigitize products?
What would happen if weactually have real-time insights
into how your car is driven onthe road?
What will happen if you havereal-time insights to your
(03:53):
tractors as they go aboutplowing in different farms in
the world?
And so we started connectingthe digital economy to the
physical economy and came upwith the idea of fusion, which
is the fusion not in the senseof energy sector, but fusion in
the sense of infusing physicaland digital.
(04:16):
And so we believe that the nextdecade is really about digital
transformation of asset heavysectors where physical products
are going to become digital,physical products, and we have
already seen it.
Look at today's speakers,yesterday's output devices, that
(04:37):
today's input output devicesthat are essentially digital
physical products, and we can goon and talk about many other
things as we go along in thispodcast.
J.P. Matychak (04:47):
So what inspired this exploration, then, of
thinking about strategy andreal-time decision-making, and
that the impact that AI and datacould have on this aspect of
business?
Venkat Venkatraman (05:02):
Yeah, again, we went back to the industrial
companies and asked ourselves aquestion what made them
successful?
They have historically spent alot of time designing their
product and making sure thatit's economically viable at a
price point that the customerswill pay, and then they had a
very efficient supply chain.
Things that they did inside thecompany extended through the
(05:26):
global supply chains and oncethey delivered the product, they
had very little information onhow the product actually
performed on the field becausewe didn't add telematic
capability.
We didn't have real-timetracking capabilities and if you
take an automobile as a classicexample, where the automobile
is a consumer product but thedata about the product ended at
(05:53):
the level of the dealer, so a GMor a Ford or a Mercedes would
typically know where the car wassold, not to whom the car was
sold often, and they neverreally had data on who actually
drove the car and how the carperformed on the road.
(06:14):
Now, with telematics, we havereal-time capabilities in terms
of what's happening on the field.
So the shift in thinking forindustrial companies and
strategy is to go away fromthinking about your competencies
as products are designed orproducts are designed and
manufactured, but products asdriven or products as used by
(06:38):
the customers, and once I knowwhy your car fails, I can then
understand whether yours is anoutlier or is likely to be a
pattern that we can begin togeneralize so that we can take
proactive corrective action.
We couldn't do that before.
Sensors were expensive,software was something that was
(07:00):
hard-coded into the product, notreal-time Cloud connectivity
the 5G connectivity was notthere.
So when I did the research in2016 for the previous book,
these were ideas that we couldsee happening in the future, and
by 2023-2024, they become real,and by the time we come to the
end of the decade, this willbecome commonplace.
Shannon Light (07:22):
And to that point , how did these emerging
technologies impact fusionstrategy, as we continue to see
how much growth, even just inmonths of AI, yeah, that's
actually a great question If youthink about.
Venkat Venkatraman (07:40):
Let me take an example again from Digital
Matrix and connect it.
One of the examples I used inthe digital matrix was IBM
Watson and the ability of IBM tocreate cognitive computing,
which is really a precursor toall the generative AI that we
see today.
But at that time, what IBM hadto do was to go inside hospitals
(08:03):
, collect all the detailed datato digitize their understanding
of cancer cure or heart diseasecure or whatever the case might
have been.
At that time we didn't have thelarge foundation models and
then suddenly chat GPT came withthe foundation models and then
we can now create real-time dataand AI that sit on top of these
(08:28):
foundation models.
So we couldn't have really beenthinking about fusion
strategies in 2020 unlessindividual sectors started to
digitize all their content notjust structured data, but
unstructured data and very fewindustries did that on a
systematic basis.
So today, in 2024, we canreasonably speculate on why the
(08:54):
different industrial sectors aregoing to develop their large
industrial foundation models,and that is going to be
extremely valuable for them tothen develop their fusion
strategies.
So that's just simply onetechnology.
The second one is really cloudand the 5G or 6G connectivity.
(09:17):
So it's now possible to thinkabout putting a 5G or a 6G modem
on remote devices and collectreal-time data on how they're
performing on the field.
We couldn't do that five yearsback.
We didn't have this widespreadcloud capability or the cellular
capability, and then we got,obviously, the software
(09:38):
capabilities and theapplications, the development
capabilities and all that is nowpossible.
So multiple technologies areconverging to make fusion real.
Multiple technologies areconverging to make fusion
economically feasible for manycompanies to explore.
The real battleground is goingto be how companies are going to
(09:59):
invest to accelerate towardsthis fusion future that we
believe will be the definingactivities for many, many
sectors.
Shannon Light (10:10):
Would you say that that adoption of fusion
strategy is likely the biggestchallenge that a lot of these
industries face?
Venkat Venkatraman (10:21):
Yeah, I think again.
When we finished the book aboutsix to eight months back, we
were in presenting the ideas tovery many companies and they're
beginning to see the rationaleof shifting from industrial age
analog thinking towards thisdigital, real-time fusion
(10:43):
thinking.
And as companies begin torealize that if they don't do
that, somebody else is going tocome between them and the
customers, they will put thosesensors, they will reverse
engineer the process, they willdeliver new value-added services
that otherwise legitimatelyshould have gone to the
(11:05):
companies that make the product.
And so once they realize thatyou need to do this, both for
offensive, proactive reasons, aswell as for defensive, reactive
reasons, then the companiesrealize that they need to
actually focus on this much moreseriously than they might have
done in the past.
J.P. Matychak (11:23):
So what do you think are some of the roadblocks
that stand in the way oforganizations transitioning into
this fusion strategy, and howmight they be able to overcome
them?
Venkat Venkatraman (11:32):
Yeah, I think there are many, many
barriers or factors that inhibitcompanies make the transition.
I think the most importantissue is companies believing
that the future success is goingto be very different than their
(11:52):
past success.
If companies have beenextremely successful in their
development, then they look atthe digital as a 10-year journey
and they don't quite see thatexponential shift happening in
their visible horizon.
Then they can say I'll wait.
I'll wait for a few more years,because in technologies the one
(12:16):
area that I've studied for thebetter part of 35 years the one
common equation is the waitingequation.
Do I invest today or do I waitand invest tomorrow?
Because I know that if I waitand invest tomorrow, much of the
uncertainties can be resolved.
Costs of technology will comedown, there'll be more options
(12:39):
for me to play with, but at thesame time I lose the opportunity
to have a first mover advantage.
So the companies are nowbeginning to ask the question
what's the value of waiting andwhat's the risk of waiting?
And so that's the first andforemost barrier.
The second one is really one ofinvestment shifts.
Companies know how to invest inwhat they're good at.
J.P. Matychak (13:03):
Right.
Venkat Venkatraman (13:04):
And successful companies fail
because they overinvest in whatthey're good at today and
underinvest in what they need tobe good at tomorrow.
So who brings this new thinkinginto the company?
Is it the CIO?
The answer typically is not,because the CIO quite doesn't
understand what a product does.
(13:24):
The CIO may understand what aprocess does and what set of
organization changes need tohappen.
So it's got to be themanufacturing, it's got to be
the marketing person, it's gotto be the operations person, who
may have competencies in theirbranches of industrial
engineering and manufacturingengineering.
(13:46):
But here we are really askingthem to take the role of being
the hybrid managers thatunderstand manufacturing
engineering and data sciences,industrial engineering and
information engineering.
So once these hybrid managerssee the potential, then I think
making the investment cases hasbeen easier.
J.P. Matychak (14:06):
And do you see this move to a fusion strategy
and even data and AI and thismay not get touched onto the
book, but it just kind of sparksthe question Do you see this as
having an impact in creating alarger gap in competition
(14:27):
between companies or do youthink that this has the
opportunity to close the gap ofcompetition?
And, to say it another way,I've even noticed, with the
adoption of AI technologies, thechat GPDs, the Google Gemini's
of the world and whatnot smallercompanies are able to process
(14:48):
things that the likes that majororganizations like McKinsey
would be able to do with hoursand hours of people power and
human capital.
Do you see this fusion strategyas a way to create a more level
playing field for competition?
Venkat Venkatraman (15:06):
I think it's difficult to generalize at this
point in time.
But if you now take yourquestion and start asking can a
new company design a tractorthat has got data and AI
capability at the core, whereit's able to show real-time data
of what the tractor is doing onthe field and Then design it,
(15:30):
manufacture it and deploy it atscale over the next ten years?
The answer is yes.
Why?
Because we can look back andTesla mm-hmm as a relatively new
company that had a differentvision of what a car is going to
be.
Mm-hmm right computer on wheelsconnected to the cloud, with
real-time data and AI, and Allthe incumbents laughed at the
(15:54):
company because they saw it asessentially a faster golf cart.
You're right right, becausenobody actually understood the
entertainment center.
Nobody actually understood thatit's possible to have an battery
electric vehicle that has thesame power as an industrial
(16:15):
combustion engine.
But then, once you have that,you can add a lot more things to
that piece of equipment.
Computer on wheels connected tothe cloud.
A tractor, essentially, canbecome computers on wheels
Connected to the cloud, butexcept it's in a more regulated
setting, if you will, not on,you know, wide roads, but in
(16:37):
defined farms.
And If John Deere doesn't do it, if caterpillar doesn't do it,
if Mahindra tractors don't do it, somebody else will begin to do
that right?
So in that sense, it's possiblefor a new entrepreneur to think
about creating this.
But the reason why I said Idon't want to generalize is
because it will require asignificant amount of investment
(16:58):
right, to actually create thatproduct.
And this is where we wanted tocontrast between the digital
transformation of the assetlight sector and Digital
transformation of the assetheavy sector.
In the asset light sector, onceyou create an app, the marginal
cost of producing the next appis near zero, and we had this
(17:22):
Relatively scalableInfrastructure for distribution
of the apps right Apple AppStore and Android Play Store
right.
Here it's going to be much moredifficult, right?
You got to design the product,you got to develop the product.
You got distribute the product.
You got to maintain the product.
So we will see it happening insectors where the industrial
(17:42):
companies are asleep at theswitch, where the new companies
come in.
But the generalization that wewant the listeners to take away
from that is a digitalindustrial company that
understands the power ofInfusing mechanical engineering
with data and AI Will do muchbetter than a mechanical
(18:02):
engineering company that doesn'tunderstand the power of the
data and AI.
And the same argument holdsgood for any human that is able
to take advantage of GenerativeAI will do better than someone
who still thinks that gen AI issimply for the, for the, you
know, for play and not for work.
Right, and so that's where theshift is going to happen.
Shannon Light (18:25):
And I mean all of that seems to play a big role
in just the long-term planningfor a company.
So, to that point, withCompanies were not going to, you
know, be willing to adopt this,do you see the potential of
them?
Just you know that companycould just being completely
(18:48):
unsuccessful, or what's yourtake on?
Venkat Venkatraman (18:51):
that.
Yeah, I you know.
Obviously we are all biased inhow we think about the
prescription coming out of ourown research, I think, our
General prince prescriptioncoming out of our researchers.
If companies believe that thenext five years are going to be
exactly like the last five years, they will find themselves at a
(19:15):
comparative disadvantageCompared to companies that are
prepared to look at the futureas being very different.
And we also write in our bookthis idea of the distinction
between forecasting and backcasting.
Forecasting is projecting fromtoday into the future, where we
typically assume that thingsthat we want to happen In the
(19:38):
future will continue to happenbecause that's a future we like,
because we'll be successful inthat future.
Backcasting really challengesthe company to take a position
in 2030 and say what will thefuture be in 2030?
Not seen from our lens, butfrom the lens of the customers
that are going to use yourproduct.
(19:59):
What will have to be true forthat future to become real?
And In that future, will mycompany be relevant?
Will my competencies berelevant?
Will I still have a role toplay in get a fair share of that
value in that future?
So the moment we take anoutside in perspective, future
(20:20):
backward perspective rather thaninside out and Project from
today.
Companies start to realize thatmuch of the competencies that
they've been good at today Maynot be relevant in the future.
Right, because corecompetencies are not about what
you're good at today.
Core competencies are alwaysabout will customers give you a
(20:41):
premium For you being good atthat?
So I may be good at many thingsand and today I may be able to
extract value from the, from thecustomers, but that doesn't
necessarily guarantee that thecustomers tomorrow will give me
that same value, but even afraction of that value, in the
future.
J.P. Matychak (21:01):
So this obviously has implications up and down an
organizational structure, youknow, especially for those that
are, you know, traditionalproducts and whatnot.
So what can leaders do?
That sort of see the vision ofthis fusion strategy but then
need to build that, thatgroundswell of support within an
(21:23):
organization that may have thethe inertia of change,
resistance against them, toreally kind of shift the focus.
Venkat Venkatraman (21:31):
Yeah, I think the CXO team has got to
take a coordination role ratherthan a delegation role.
Okay, because what has happened?
When we had the first few years, few decades, of digital
transformation, the CEOsrealized that we need to
transform our processes.
With big ERP systems and CRMsystems, you know the big SAP
(21:54):
implementation.
That's three years and fiveyears of implementation.
They made the investment andthen they delegated it down to
the CIO or the chief humanresource officers and chief
operating officers.
Now the fusion strategy isreally challenging the status
quo and, if you believe that thefuture is not an extrapolation
(22:14):
of the past, we now have torethink the data flows inside
the company, the data flows thatextend to my suppliers and to
my partners and to thedistributors, and have to
reimagine what the factory ofthe future will look like.
We got to reimagine what theproduct of the future will look
(22:35):
like.
We have to reimagine thecontrol center, the operation
center, that these data startsto flow back in.
Many of those things arerelatively new ideas.
Right, we saw with the digitaltechnology people started
tracking real-time data ofconversations about our products
(22:57):
in social media.
That was relatively new.
Suddenly, I need to know whatpeople are saying about my
product on, you know, google orFacebook or Twitter or you know
wherever else the case may be.
So they created listeningcenters, social media listening
centers.
The equivalent idea for productcompanies is to understand
real-time data of how myproducts are performing on the
(23:18):
field.
That requires data, not justsomething that the marketing
department does, but themarketing department shares with
the operations department, withthe procurement department and
with the suppliers.
So what we write in the book isthis idea called digital twin.
J.P. Matychak (23:36):
Okay.
Venkat Venkatraman (23:37):
Where the digital twin has been well
understood in terms of designinga product.
We think the digital twin has toextend beyond product design to
manufacturing, but alsoperformance.
Once I designed the digital twinof actual performance of the
product in the field, then I'llknow why the product fails in a
(24:02):
specific situation and byunderstanding across these
different settings I get a muchbetter insight into why my
product fails.
And that data, if it isfunctionally kept siloed, then
the marketing department willthen not share it with the
manufacturing department, ormanufacturing department may not
(24:23):
share it with the productdesign department and the
product design department maynot share it with the suppliers.
The moment we take the CXO andgive them the charter of
end-to-end visibility with thisreal-time data and AI, they
realize that this is a weaksignal that I'm observing in the
field.
That's going to be extremelyvaluable for me to make
(24:45):
proactive corrections ratherthan think of this as what we
call in the book systems ofrecords rather than systems of
data graphs.
And the systems of data graphsare really understanding how my
products perform in the field.
So we are introducing in thebook ideas and mechanisms by
which the senior leaders can putin place organizational
(25:09):
mechanisms by which they cancreate, this infusion of
physical and digital domains.
J.P. Matychak (25:15):
And it sounds like it really requires the true
breaking down of departmentaland functional silos in order to
be successful.
Venkat Venkatraman (25:24):
Yes, because successful companies have
relied on functional excellence.
They're good at R&D, they'regood at manufacturing, they're
good at marketing, andcross-functional excellence was
not something that was expectedexcept at the senior level, but
(25:47):
even then they relied onfunctional expertise to say if
something is an R&D problem, Irely on you, if something is a
marketing problem, I rely onsomebody else.
So now, by using the wordfusion, we are really beginning
to say that it's the infusion ofdisciplinary thinking, infusion
of mechanical engineering withdata, as I said earlier,
(26:10):
infusion of industrialengineering with information
sciences.
These are new ideas that willmake successful companies
differentiate from others thatstill are thinking about
functional excellence ratherthan cross-functional or fusion
thinking.
J.P. Matychak (26:33):
So let's talk about the book itself.
You both decided to write thisas a workbook almost right and a
guidebook for leaders toimplement this new way of
thinking.
Talk about your reasoningbehind that, and you know, in
(26:55):
particular, what strikes me andit's something that you said
earlier on in the conversationwas that, even since you
finished the book, to now lookhow the world has changed.
So you wrote this as aguidebook.
How are you planning on keepingthis updated for executives and
leaders to keep up with thetimes?
(27:17):
Because, I mean, we thoughttechnology was rapidly changing
before.
It seems exponentially faster.
Venkat Venkatraman (27:27):
Yeah, I think it's a limitation of the
book publishing cycle, if youwill right.
J.P. Matychak (27:33):
Right.
Venkat Venkatraman (27:35):
We finished.
The major drafting of the bookwas done by August of last year.
It goes through reviews.
It comes back with somecomments.
We might have revised it andbasically sent it in just around
Thanksgiving, Right, and thebook is coming out in March.
So that's at least six monthsof thinking.
(27:55):
And nothing has happened and youknow that's okay in many, many
disciplines, but it's not okayin a discipline that we are
really talking about.
That is changing pretty muchevery day, right?
So we did two things.
One is the book is coming outin March.
We, hopefully will have theplaybook with the templates that
(28:15):
will allow companies to thinkabout much of the ideas that
we've been talking about duringthis podcast.
Sometime in the AugustSeptember timeframe, the Harvard
Business Review Press willdecide how they want to
distribute the templates thatmake the playbook actionable and
, even though we write aboutmany of the actions in the book,
(28:35):
we'll provide some templates tomake that happen.
We also wanted to do something,which is to practice fusion in
terms of making physical digitalbook work.
Now, if you think about most ofthe books, there is a physical
book and there's a digital book,which is basically a static
(28:57):
version of the book, and thenthere's an audio book, which is
basically an audio version ofthe static book.
The form doesn't change thesubstance of what's in the book
and what we are now trying to dois to update the material since
the book is finished, expandthe material because, by design,
(29:23):
the book is 200 plus pages andwe may refer to a case in two
paragraphs when the richness ofthe research that we might have
done is maybe six to eight pages, but the actual material may be
a lot more than six, seven,eight pages.
So what we are doing is we'reactually creating a fusion
strategy coach.
(29:44):
Oh, interesting which will be aGen AI coach, which you think of
it as a virtual VG and virtualVenkat oh, where a reader can
potentially ask questions and,to the extent that we can train
the material to help you answerthat question.
We have gone beyond just simplygiving you a static book or a
(30:08):
digital book, and so at thispoint in time, we have tested it
out with about 40 executiveswho've had a chance to read the
early draft of the book and wehave asked them to just test out
the coach to see whether thecoach allows them to go beyond
what we have written in the book.
And we're getting good.
(30:29):
We have gotten very goodfeedback on the coach.
We're gonna test it out with agroup of about 100 students at
Tuck who is gonna take thestrategy class in March and
April.
We're gonna do the same thing.
They're gonna read the book,they're gonna use the coach and
then see if the by using theirprompts right we're not gonna
give them the prompts usingtheir prompts whether they're
(30:51):
able to get deeper insight andunderstand the connections
across the concept beyond whatthey would get from the book.
And that will also help usdevelop the playbook with the
templates and the prompts by thetime that comes out sometime
later this particular year.
We believe that the idea of areader posing a question that
(31:20):
the book is able to answer inthe language of the reader is
really where the book should go,and we're basically thinking of
this as the first stage of thistransformation, and what we
haven't quite decided is howfrequently we'll update it and
(31:44):
how will we host it, where willit be hosted, how long this will
be maintained, how to chargefor it.
All these are open questions,but we believe that we should at
least let the initial readers,the initial buyers of the book,
have an opportunity to askquestions that otherwise we
would have answered using anonline forum or using an email
(32:07):
forum.
But here we hope that we canget the Gen AI to provide first
level answers, and that's gonnabe our experiment with the
launch of the book.
Shannon Light (32:18):
Currently, are the those who have tested out
this coach?
Is that living on an onlineplatform where it's similar to
say, something like chat, gpt,where you type in it's at this
point?
Venkat Venkatraman (32:31):
in time, a proprietary platform in which we
give the link and we give thema password so that only they can
access it, because we don'twanna make it open at this point
in time.
It also raises interestingquestions about intellectual
property Because which is mynext question Absolutely the
(32:54):
book is copyrighted by theHarvard Business Tribunal.
So we are still working throughall those issues.
But the executives then had achance to then test out these
ideas.
Right, they could ask abouttheir company.
Their company might not havebeen done a discussion in our
(33:15):
book.
And then we are seeing how wellthe coach answers that question
and the way we have trained itis actually looking back.
It is very simple, but westruggled through it for a very
long time because we wanted tomake it a simple coach, without
too many options given to thereader, and finally we came down
(33:37):
with a simple idea of what I doin the classroom, right when
somebody asked me a question.
What I do in the classroom isconnected back to what I know
and connected back to what thestudents know.
So what we have trained thecoaches to do the following you
ask a question.
It refers back to the coreideas from the book.
(33:57):
It doesn't say chapter five ordoesn't say chapter seven,
because it synthesized acrosschapters.
It writes an answer in about250 words and, in making it
conversational what it does, itends the answer with two
questions for you.
One that goes deeper into thequestion you asked me, which is
(34:22):
what I would do in the classroom.
Right, you're asking me, I'llexpand more on it.
The second question is tolaterally connect across
concepts or across industries.
So, if the question is aboutfarming, one follow-on question
may be how does this idea applyin transportation or in mining
or in construction?
(34:42):
And if you asked me aboutfarming, the second question may
go deeper into the farming howdo the digital twins apply in
farming?
How do we begin to think aboutsystems of data graphs in
farming?
Right, because that's thelogical connection by creating
this Socratic thinking, which iswhat we do in the classroom,
(35:05):
into the Gen AI, we are able togeneralize and have one general
coach that will continue to haveconversation with you till you
find that you've gotten thelevel of answer you can get from
this.
So so far, it's text-based.
It's not simulating my voice.
I'm not trying to create a GenAI version of my answer using my
(35:29):
voice.
It's text-based, but these areeasy extensions we could do.
We could do an AI avatar withmy voice, an AI avatar with my
video, but first we want to makesure that the concepts work,
and that's what we're doing sofar.
J.P. Matychak (35:45):
Wow, truly an opportunity to both transform
the business world and how we dobusiness, but also the
potential to transform how welearn as well.
Incredible.
Venkat Venkatraman is the DavidJ McGrath, Jr.
Professor in InformationSystems at the Boston University
Questrom School of Business.
(36:05):
His upcoming book FusionStrategy how Real Data in AI
Will Power the Industrial Future, is coming out on March 12th
2024.
In is available for pre-ordernow on Amazon.
Venkat, thank you for joiningus.
This has been incrediblyinteresting.
Venkat Venkatraman (36:22):
Thank you very much for having me.
J.P. Matychak (36:26):
Well, that will wrap things up for this episode
of the Insights at QuestromPodcast.
I'd like to thank our guestagain, Venkat Venkatraman.
Remember for more informationon this and previous episodes,
along with other insights fromQuestrom experts, visit us at
insights.
bu.
edu.
For Shannon Light.
I'm JP Matychak, and so long.
Popular Podcasts
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
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.