AI-Native vs. AI-Infused Products

What’s the difference between AI native products and AI infused products?

And what are some examples?

I think this is really important for your business, and today, we’re gonna get to the bottom of it.

Hi, I’m Courtney Baker, and this is AI Knowhow from Knownwell, helping you reimagine your business in the AI era.

As always, I’m joined by Knownwell CEO, David DeWolf, Chief Product Officer and Chief Technology Officer, Mohan Rao, and Chief Strategy Officer, Pete Buer.

We also have a discussion with Peter Armaly about how AI can impact the ever important area of customer success.

But first, here’s my recent conversation with David and Mohan about AI native products versus AI infused products.

David, Mohan, there’s something I need some help on because I keep seeing this out in the market and I need some help for myself being able to distinguish between these two things.

I think a lot of executives need help as well because these marketers, I’m telling you, they are good.

Okay.

So right now, everything, every SaaS product, everything, is all of a sudden got like an AI sticker slapped on it.

You know, it’s like everything at the store with a sales sticker.

It’s all got AI slapped on it.

And it’s the hot thing.

But there’s actually a really big difference between an AI native product and an AI infused product.

And I would love both of your help breaking those apart and helping us understand, how do we tell when we’re looking at different products or platforms?

Is this just like something that AI has been, you know, they stuck the marketers, the smart marketers, they stuck that sticker on there?

And what’s the difference?

How do we do this?

So I would love for you two to train us up on how to tell the difference.

And then how do we think about those differently in our businesses?

So I want to give the very direct answer to the question, but at a very high level.

And then I want to hand it over to Mohan, because he’s going to tease us apart for us.

AI Native means that from the ground up, from the very, very beginning, the product has been built to leverage, depend upon, and take as a primary aspect of that product, the value proposition of artificial intelligence.

And what you get when you start with AI and you start to build the product from the ground up that way is a lack of the kind of false boundaries, the constraints you may have, if you’re simply adding it on to an existing product.

You can rethink and you really have a blank canvas from which to draw.

So it really promotes more innovation and a kind of bigger degree of the ability to really embrace this brand new technology.

And so that’s the advantage is you don’t have those constraints and you have that blank canvas.

Okay.

So that’s the high level.

But Mohan, pull that thread and tell us why it really matters as you’re building the product.

The best way to think of it is, are algorithms and data a core part of the value creation, or is some other part of the product the core part of the value creation, right?

So just thinking along those lines will lead to the right answers here.

In an AI infused product, as Courtney, you called it, this is taking an existing product, perhaps calling OpenAI or Anthropic via an API, coming up with some conclusions and displaying it.

That is an example of an AI infused product.

If I can take a broad swipe, every product that existed before, let’s say, 2022 that now claims to be AI is an AI infused product, unless they went and core dug it up and rebuilt the product.

You can just draw a line that way, roughly speaking.

In an AI native product, the core experience, AI is baked into the core experience and is not an add-on.

That is the core differentiation between the two, and our algorithms and data are core part of the value creation, as I said.

We can tease it apart further.

We can talk about the challenges, but really, what is the center of gravity of the product or the system that you’ve built is the question here.

The other piece of that that I like to push into, you talked about the algorithms and data.

One of the things that data algorithms can do is with artificial intelligence, we can actually create new value, not just streamline a process, not just report on value that’s already existed.

It’s actually creating data out of natural language would be an example of this, of where we can’t put BI tools on top of natural language right now and get any sort of meaning out of them.

We actually can.

We can use these large language models, we can use AI to create brand new value that we haven’t seen before.

I think those are the types of things that start to open your mind up and say, if we build the product from the ground up, relying on algorithms data like you were just talking about, Mohan, all of a sudden, we can capture brand new value we haven’t even thought of before, versus just adding a new feature to an existing tool.

I want you to dig into that a little bit more.

You just gave an example there, but I would love to go even further with that.

If I’m in the market to find a piece of technology to solve a problem in my business, how should my expectations be different for an AI native product versus an AI infused, or as I so eloquently put it, a SaaS product with a sticker on it?

What should my expectations be for those two?

Do you still buy software off the shelf at Office Depot, Courtney?

I didn’t know that was you.

Yes, actually those boxes, those are still for me now.

Can you imagine?

What a different world we were in.

Totally.

I think of this, if I was buying software and I was looking at AI native versus a AI infused product, I would really slice it up.

Is the business process that I am trying, or the pain that I’m trying to solve with this, is this a problem where I need a proven, reliable solution?

I want a conservative approach, or am I trying to future proof?

Do I want to buy something that may be ahead of the curve?

I’m looking for more upside.

It’s an upside versus downside equation.

There’s proven solutions that will leverage AI in order to get some more legs out of it to add more value.

But it’s still accounting software.

It’s just going to do a few more entries for you versus the AI native solution is going to take a totally different approach for that.

With your accounting, you may not want something on the bleeding edge.

But if you’re doing marketing and you’re trying to reach people in a new way and you’re looking for that upside and that growth, you may actually want to take that bet and say, you know what, what we’re looking to do here is differentiate.

I’m going to look for the brand new way of doing something that’s going to give me value other people don’t have versus the conservative approach.

Yeah, I agree with that.

You know, from a buyer perspective, largely it should not matter as much, right?

So if what you’re buying is something that fits your purpose, fits as long as your vision can go, either one can work.

But one of the other differentiators between the two is that generally speaking, the AI infused products work within the bounds of the data that already exists in that SaaS application, right?

And if that is good enough for you, that is fine, right?

So just buy it.

It’s, you know, it’s providing…

Do you think that’s true?

Yeah, tell me, tell me why it’s not.

You’re like, obviously, that’s why I said it out loud.

No, no.

It’s, you know, for example, if you look at any of the traditional SaaS products that we use, CRM applications, ERP and others, they take the walls of what’s in the databases of that particular application and are building AI around it.

Right.

So now that can definitely lead to value creation.

It can automate certain tasks.

It can, but it is still working with the data that exists within that SaaS application.

Is it important now to be with a AI native company because the way that that company is going to be able to grow into the future versus choosing a technology that’s built before 2020 and maybe not having the room to grow and change into the future?

I think the other side of that equation though, and this is where you need to be doing kind of active risk-reward analysis, is I think that that theory is true.

But in this world of AI native, where AI native is brand new, there are going to be a lot of companies that make the wrong bets and get it wrong.

Yeah.

So you may be riding the wrong horse that doesn’t actually make it to that next level.

I think that’s part of the risk equation, is just being AI native doesn’t mean that you are going to be viable in the future.

It means you have a better chance of being viable in the future and not running at a runway, but it also means you may have taken a totally wrong bet and you’re not there anymore.

I think it’s a bigger order question than just future proofing on its own.

You two are helping me so much.

I love this and I think the people listening are, okay, here’s what I have so far.

I’ve been taking notes.

I have AI infused.

Listen, it’s like AD and BC.

We got before 2022 and after 2022.

Somebody come up with something creative, we’ll trademark it.

Okay.

If it’s standard, repeatable, proven, it’s probably an AI infused product because it’s working off of algorithms.

For AI native, you may be looking to future-proof your business, or future-proof maybe the growth or looking for cutting edge technology.

What else am I missing?

You know what I think might be fun and maybe isn’t what’s missing but it might be missing from the conversation is, let’s take some thought exercises.

When I think AI infused, Photoshop is an example.

Adobe has done a great job of embedding AI into Photoshop.

When I think AI native, I think mid-journey.

Both image manipulation, image creation tools, one with an old paradigm, one with the new.

Right now, would you say they solve the same problem?

Maybe, maybe not.

Do I believe that over time, they will converge and one may have a leg up over the other?

Yeah, I think so.

So I think that’s the different paradigm.

I don’t know, Mohan, what’s another example of products like that where you have a front and center existing dominant player in the market that is AI infused, and then you have a new upstart that is AI native?

You can think of use cases in transcription.

So there have been transcription businesses all along.

So you send your video recording out to whoever they listen to and they get it out in nice language and it comes back.

It’s a 24-hour process.

Now we’ve got our AI native real-time that translating.

They may not be as accurate as the tried and tested transcription services and language translation services as well, but that is an example that is more AI centered approach for transcription.

So if I’m a legal firm and needing to transcribe some audio from a, what’s it called, a deposition, I’m going with the first option.

But if I’m known well and just needing the transcription from this podcast, I’m going with the second version.

Yeah, and I think very different value propositions, right?

I love that Mohan tapped into the accuracy versus the time and the turnaround would be one.

Another one is these AI native transcription tools are providing summaries and to-do lists, and they’re able to, remember how I said create new value?

Yes, yes.

That is creating new value that doesn’t exist.

When I just have an accurate transcription, I have an accurate transcription, right?

When an AI native and I’m thinking, what can this technology do to solve the problem I’m trying to solve from a transcription of a video conference, all of a sudden I get brand new ideas that weren’t even contemplated before.

I think that’s where that example you bring up, Mohan really calls out the difference.

Okay.

Can we also just do ourselves, Knownwell?

It’s almost like you staged that.

Let’s go.

I didn’t.

Yeah.

I think we can look at adjacent markets to what Knownwell is doing.

I think one of our struggles has been, because we are AI native, we have a very different value proposition from the adjacent markets that currently are out there.

It’s hard to even say because the industry is so young, what is the market that we’re reinventing or is it a category that we’re creating?

We don’t actually know yet.

What we know is that there is a series of problems that aren’t really solved just yet in the market.

Let’s look at a couple of adjacencies.

You brought up the customer success platforms.

There are absolutely customer success platforms that, based on your transactional data, are able to predict the retention of your customer base over time.

The problem, a lot of organizations, specifically professional service organizations, don’t have transactional data that you can use to actually fuel those platforms.

Go back to Mohan’s point about the constraints around the boundaries of the data that already exists.

What are we able to do?

Because we’re AI native, we’re able to actually create operational data from this natural language and communication that’s already flowing around your enterprise and your organization, and we can turn that into the operational missing piece and turn that into value.

Another example that we would see in the market is right now, if you look at professional services firms and you take this the opposite direction, the software that they do use is professional service automation tools.

Now, we’re not doing anything that anybody would claim is in the PSA category right now, because what those tools do is automate the timekeeping and the resource management and the unique aspects of our professional services firm, but they’re not even touching the commercial end of the customer relationship.

How do you monetize that relationship?

How do you…

But when you think about the words professional service automation, shouldn’t that be the whole business and especially the core of the business, which is the economic relationship?

So will there be opportunities for us to create new value in this broad sense of professional service automation?

Absolutely.

And so you can start to see how there are some categories that are closer than others.

But I think this is what new waves of technology does, is these waves kind of change the game and they change markets and they change categories where all of a sudden the boundaries start to merge and evolve and morph over time.

An extension of David to what you said also is in traditional software, humans have to do something with the machine, right?

So you’ve got to enter a lead into your CRM.

You have to update the notes.

You have to create an opportunity, whatever they are, right?

So it could be in customer success platform or it could be in customer relationship management platform.

But in an AI native application, the supposition has to be that there are things happening and getting processed in certain way that give you the intelligence for humans to make decisions.

And if you look at it from how you interact with the machine, it is doing something with the machine versus using your machine to make higher order, intelligent decisions.

So that is another line that you can carve out here between the two.

David, Mohan, thank you.

I think this is really helpful.

And I think for everybody listening as you’re looking out there, I do keep talking about these stickers.

That’s because in marketing, we still, you still will hear the word smack dots.

I kid you not, people talk about smack dots.

I’ve never even heard that word in my life.

What does that mean?

I know.

It means the like little like, you know, when you see an ad and it’s like 40% off everything and it’s in that little round circle.

Ah, interesting.

That’s smack dot.

Yeah.

No idea.

Okay.

So, you know, that’s still in our lingo, which I think it comes from those stickers on packages way back in the day.

So, when we see AI slapped on everything, hopefully this gives you a little more context for what they might be adding AI to.

Is it an existing platform and technology driven by algorithms, or is this truly a brand new technology and an AI native platform or product?

David, Mohan, thank you as always.

You’re very smart.

Adios.

Thank you.

Acquiring new customers is more time consuming, costly, and resource intensive than keeping the customers you already have.

That’s why we’re building an AI platform that warns you about at-risk clients and gives you actionable intelligence on what you can do to prevent surprise churn.

Go to knownwell.com today to learn more and sign up to see what we’re building.

Peter Armaly is a partner at Valuize and the author of a new book on Mastering Customer Success.

He sat down with Pete Buer recently to talk about AI’s role in enabling and ensuring customer success.

Peter, great to have you on AI Knowhow.

Welcome.

Thank you, Pete.

I really appreciate it.

Could you please give listeners, for the sake of context, a little bit of background on Valuize and your role there?

Sure.

Valuize is a consulting and advisory firm.

It’s focused on helping B2B technology companies create a connected and collaborative customer lifecycle with the aim to accelerate the net dollar retention.

My role as the company’s principal is to spearhead the penetration of the Valuize brand into the B2B enterprise technology market.

I support the acquisition of new strategic clients in the market, and I lead the prescription and delivery of outcomes to those clients.

Awesome.

Thank you.

And you have recently written a book.

It’s called Mastering Customer Success, Tactics to Decrease Churn and Expand Revenue.

Congratulations.

That’s awesome.

Thank you.

What’s the business problem that sits at the center of the customer success manager’s role in the business or ambitions to do great work?

Yeah.

So the business problem they’re trying to solve is really retaining customers, especially in the world of subscriptions.

And I say that because customer success, although it’s largely known as organizational practice that’s in the world of SaaS, it actually has been picked up by outside of SaaS businesses too.

And so regardless of where it is, their main focus really is keeping customers that are signed.

And especially in the subscription business models where the licensing agreements are usually like two to three years, there’s a lot of imperative to work really quickly on having those customers receive good value for their investment.

Because the buy-in for SaaS subscriptions is low, customers don’t have to invest a lot to start working with your product.

And so the vendor, the imperative for them is to retain the customers two to three cycles so that they start making profits off them.

And so customer success has been in the spotlight for some time as the organizational entity that is required to make sure those customers realize value from their investments.

And that’s the problem in a nutshell, really.

I mean, there’s additional kind of things that people like customer success to focus on, which is expanding revenue and all that.

But none of that’s going to work if they can’t keep the customers in the first place.

Is there a burning platform?

Is there something about customer success that’s getting harder or something about the manager job that’s getting harder and therefore critically important to focus on?

Yeah, I would say it’s really challenging to deliver the practice of customer success because you don’t have everything within your control.

What I mean by that is a lot of the work that gets delivered is dependent upon other organizations within the company.

Usually, obviously, support to fix the product but product organizations themselves have to design good products and have to make sure that the features match the needs in the marketplace.

Oftentimes, you need sales to communicate really good business context information about the clients.

You even need marketing to help you out in terms of communicating at scale.

Customer success, as I mentioned, is at the forefront of driving successful outcomes for clients, but they can’t do it alone.

The burning platform is if they can’t figure out how to orchestrate all that, they’ll fail in the mission.

That’s what we see oftentimes.

That’s the criticism that they get.

That’s what we often see in the market now or in the last couple of years.

A lot of companies blowing up their Customer Success Organizations because essentially they failed at that ability to do that on a consistent basis.

Nice.

Can you share some of the nuggets of wisdom about what Customer Success Managers should do to be successful?

What does best practice look like?

Best practice and this is really mainly the theme of the entire book is to focus on improving the fundamental skills required to be successful in the role.

And there’s no secret, a lot of these skills will equip anybody in any kind of professional role in a company to do well.

And that’s communication skills, that’s organizational skills, that’s just being to have a professional demeanor, it’s being collaborative, it’s being empathetic.

Maybe that last one is more specific for Customer Success because they have to actually be thinking about the customer all the time and trying to put themselves in their shoes.

So empathy is a big huge piece of that puzzle for customer success managers.

And so, if you’re a leader and you’re trying to kind of recruit for your team, there are things that you should be looking for so that you’re finding the best kind of personalities that will be most successful in the role of customer success managers.

But that’s, I’ve led teams, I’ve built teams and I’ve worked with clients now that have teams.

And the common kind of success criteria are really those basic fundamentals.

They hire people that are collaborative, that are eager to learn.

There’s so much to learn in the tech world and in customer success because the clients are all different.

You have to understand their business.

You have to have a curiosity and an appetite for appreciating business differences.

And so there’s just a whole kind of host of what I call just basic fundamentals that we can kind of specify and help people understand if you focus on these things and improve them and develop excellence around these things.

These will set you up for great success, even if customer success is not in your future long term.

If you choose to do something else, those are good skills to have regardless of where you go.

As I think you know, we focus on AI as a topic in serving executives through the podcast.

And we also have a product in development that has a customer management focus.

And so the big question is, where do you see AI playing in the customer success space?

I see it playing a significant role.

And I see it revolutionizing the practice of customer success.

And I say that based on me predicting that the space requires revolutionizing.

For the last decade, I’ve been saying this space needs to be upended.

Because it’s not really serving, it’s not really delivering on its mandate to drive successful outcomes for customers at scale.

And I specify that because, sure, a lot of companies do a good job with their top 20% of customers, the most lucrative ones, the ones that spend the most ARR.

But there’s a whole thousands of companies, customers out there that are not receiving the kind of service they should receive.

AI will change that significantly.

Up till kind of this expansion of AI in the last couple of years, automation helped.

You know, traditional forms of automation allowed customer success organizations to expand their scope of their work that they did and to cover more accounts.

But AI will really revamp that by offering more predictive kind of capabilities, more accurate targeting of information and guidance.

Because ultimately, to be successful in customer success, you have to be able to provide guidance at the time that customers need it.

And ideally, even before they need it, if you understand their needs and what their experience is at the point in the journey that they’re in, you can provide information for them to help them avoid problems that other customers are receiving at that point.

And so, AI will really accelerate this significantly.

And I’m not a doomsayer.

I’m not the kind of person who thinks, okay, this is going to cause wholesale kind of job losses.

I think there’ll be job change.

But I think ultimately, I put myself in customer’s shoes all the time.

I think this will be good for customers.

I think they’ll get better service from companies across the board.

And they will welcome that.

And, you know, logically, that means they’ll probably invest more over time.

With the vendors who do a good job with incorporating AI.

There’s a continuum of practice around whether or how much revenue generation sits in the customer success function.

And there are arguments that go all the way back through time, the different profiles of people for different jobs.

And so maybe it makes sense.

Maybe it doesn’t.

I wonder if AI somehow helps break that compromise.

Can you imagine CS jobs changing, expanding and taking on new forms of value creation?

Yeah, yeah, I can.

You know, there’s a general kind of push, a lot of pressure for CSMs to develop more, let’s call it sales expertise, so that they can take on responsibility to upsell clients, to sell them adjacent products.

You know, in other words, cross-selling with the same customer.

I think that AI will help equip these CSMs with the kind of skills or knowledge anyways that will give them the confidence to have the right conversations at the right time.

So let’s say you’ve been working with a customer for a while and AI is helping you by giving you information about the customer based on what it’s observing of them through the cases that they’re logging or other kind of behaviors that they’re exhibiting on social media and stuff.

And it can tell, it should be able to tell whether they’re having success besides the work that you’re doing as a CSM directly with the client and you’re documenting it.

AI should augment that information with additional information to help you determine and maybe it’ll provide a clue saying, this client is probably ready for a conversation about product ABC.

And here’s the reasons why.

And I think that’s kind of a quick use case.

I could see happening pretty quickly.

And I just think that AI will really improve the ability of CSMs to have more intelligent business conversations with clients.

And finally, hopefully, put to rest the ongoing kind of complaints that CSMs spend too much time talking about product problems and not enough time about solving business challenges for the clients.

Peter, thank you.

It’s been a delight to speak together today and wish you all the best.

Thank you, Peter.

I really appreciate it.

It was a lot of fun too.

Thanks as always for listening and watching.

Don’t forget to give us a review and make sure that you hit that subscribe, follow whatever button it is in your podcast player of choice to get the latest episode each week.

We also like to ask one of the large language models to weigh in on the topic at hand.

Hey Gemini, welcome to the party.

This episode, we’re talking about the difference between AI native products and AI infused products.

So can you break those down for us?

Sure thing.

AI native products are built from the ground up with AI as their core.

Think of them as AI first.

On the other hand, AI infused products are traditional products that have been upgraded with AI features.

It’s like adding a turbocharger to an existing car.

Now you’re in the know.

Thanks as always for listening.

We’ll see you next week with more AI applications, discussions, and experts.