[00:00:00] Chirag Mehta: Me being a Chief Product Officer, I used to say that Chief Product Officers are chief [00:00:05] backlog officers. They sit on a backlog. That's what [00:00:10] used to happen in the past. And because you could not deliver it enough. Yeah. And [00:00:15] there were always more things to do. And I think now you're introducing [00:00:20] this concept, which is you cannot get something right without experimentation.[00:00:25] 

[00:00:25] Chirag Mehta: And you need to have skills to, to be able to experiment. You just don't have to build features and outcomes and [00:00:30] whatnot, but you really have to understand what is working and what is not working. 

[00:00:34] Trisha Price: [00:00:35] If you build software or lead people who do, then you're in the right place. [00:00:40] This is Hard Calls. Real decisions, real leaders, real outcomes.[00:00:45] 

[00:00:45] Trisha Price: Hi everyone, welcome back to Hard Calls, the podcast that highlights the [00:00:50] best product leaders from across the globe. If you're new to the show, I'd like to invite you to hit [00:00:55] follow or subscribe so you can stay up to date with all the latest episodes. Today's [00:01:00] episode is a conversation I had with Chirag Mehta from Constellation Research. [00:01:05] We discussed the growing shift from being feature factories to become outcome [00:01:10] driven product leaders. And here's why I wanted you to hear this. We're at this [00:01:15] inflection point where AI is compressing product life cycles in ways I have never [00:01:20] seen before from years to days. In many cases. You can literally ship something the [00:01:25] same day you start building it, but that also means you can ship the wrong thing faster [00:01:30] than ever.

[00:01:31] Trisha Price: In this conversation, we talk about what it really means to run your [00:01:35] product team like a research lab. Why your North Star metrics matter more than your feature [00:01:40] count. And how the ratio of product managers to engineering is fundamentally changing. [00:01:45] We also get into something, I don't think enough people are talking about how to make sure your [00:01:50] good decisions actually get rewarded and not buried in backlogs.

[00:01:54] Trisha Price: If [00:01:55] you've ever felt like a Chief Backlog Officer instead of a Chief Product Officer, or if you're [00:02:00] trying to figure out how to experiment without burning through your budget, this one's for you. [00:02:05] 

[00:02:05] Chirag Mehta: So we're gonna have a fun talking about. Anything and everything about [00:02:10] product and experience lifecycle. I have been a Chief Product Officer in the past [00:02:15] and now I serve Chief Product Officer as buyers and as end users.

[00:02:19] Chirag Mehta: So this is gonna be a [00:02:20] fun conversation. So let me share, let me [00:02:25] start with something what I'm seeing, and there's been a lot of chatter [00:02:30] around features versus outcomes, what we have been hearing. [00:02:35] That features are great, but features won't take you anywhere. Anyone can build a feature in [00:02:40] a short period of time.

[00:02:41] Chirag Mehta: You can ship a product in a day if you want to, but it [00:02:45] might not do anything. So you might end up building and shipping something, but that might not [00:02:50] help achieve a business outcome. So we are seeing this transition. From this [00:02:55] feature driven software development to outcome driven product development. [00:03:00] Do you see the same thing?

[00:03:01] Chirag Mehta: Maybe? I mean, that's a good starting point. 

[00:03:02] Trisha Price: Yeah, it's a great starting point. [00:03:05] Chirag, and this is gonna be so fun, both of us serving product people, but being product people. [00:03:10] So I'm really looking forward to the discussion. I do, I, I think that [00:03:15] even prior to aI momentum that we've seen over the [00:03:20] last year.

[00:03:20] Trisha Price: There really has been a shift in focus from feature factory to [00:03:25] outcomes because we can spend a lot of money and celebrate that we shift a new [00:03:30] product or a new feature, but so what? If people don't use it? If it doesn't drive [00:03:35] ARR, if it doesn't reduce support costs, who cares that we shipped it because we [00:03:40] shipped the wrong thing?

[00:03:41] Trisha Price: Well, fast forward to today, and I know [00:03:45] we'll spend more time on this throughout, AI has come along. The ability to use [00:03:50] incredible prototyping tools and even get to real [00:03:55] product is easier than ever and quicker than ever, right? And [00:04:00] you've got tools like Lovable and Bolt and Cursor and all these tools are incredible and they [00:04:05] get product out there even quicker.

[00:04:06] Trisha Price: But it can also mean we just build more bad [00:04:10] product or more features that still don't [00:04:15] drive the outcomes that we're trying to drive. And it's [00:04:20] really fun watching this sort of evolution in product [00:04:25] that product people are truly strategy and business people. [00:04:30] And accountability is to the same things that the rest of the company has in terms of [00:04:35] business metrics.

[00:04:36] Chirag Mehta: Yeah, so I, I like what you said, strategy [00:04:40] and business people. And if you look at the sort of product manager, [00:04:45] typical product manager and I'm an engineer by background, and [00:04:50] as part of my team I have product manager, so we're engineers and people who are doing engineering experiences. [00:04:55] So we have seen this combination where anyone and everyone could be a product manager.

[00:04:59] Chirag Mehta: You have [00:05:00] to be passionate about technology domain and all that. But you make a point [00:05:05] that things are changing now. There's this idea being [00:05:10] you have to be technical enough. To figure out what is going on. Because once again, [00:05:15] you're, you know, it's, it's all about technology. You are building products, but we, we are [00:05:20] talking about outcomes and outcome is about business and strategy.

[00:05:24] Chirag Mehta: You have to [00:05:25] translate your company's goals into what is happening outside with [00:05:30] respect to your market competition and with respect to your customers. So what does this blend [00:05:35] of technology and business looks like? 

[00:05:39] Trisha Price: I like [00:05:40] to think of it as a combination of inside out thinking and outside in thinking, [00:05:45] and I think engineers, maybe I'm biased because I am one, can make great [00:05:50] product people.

[00:05:50] Trisha Price: They really can because they nail the inside out thinking the [00:05:55] ability to play with technology and think of the art of possible to innovate [00:06:00] something different. And especially in the era of AI, to be able to break a problem [00:06:05] set down, right? To understand here's the outcome I'm trying to drive, here's the problem, and [00:06:10] then what are the pieces?

[00:06:11] Trisha Price: What are the chunks to start shipping and get velocity and [00:06:15] get things out there. And that inside out thinking is very natural for engineers, [00:06:20] but that can lead you to ship a lot of bad product. And so to me, it doesn't [00:06:25] matter, engineer or not. Outside in thinking. Also, some people might call it [00:06:30] product sense, it is all about knowing your market and the [00:06:35] outcome you're trying to drive. And to have very strong product sense. You have to know [00:06:40] those jobs to be done. You have to know your users inside and out, and what [00:06:45] problems are truly painkillers, not vitamins for them. But that still has [00:06:50] to layer up to your own company's outcomes.

[00:06:52] Trisha Price: So like the outside in thinking part [00:06:55] kind of, and the business strategy comes in two flavors, right? The first is. [00:07:00] Knowing your users, the value you're trying to drive for them, the problems you're [00:07:05] trying to solve. And then the second is, what am I trying to do for my company? [00:07:10] Right? Am I trying to get new ARR?

[00:07:12] Trisha Price: Am I trying to create a new product for the same [00:07:15] customer, but solve a different problem and then really get your product North [00:07:20] Star metrics aligned to that. 

[00:07:22] Chirag Mehta: It's interesting you said that you could end up [00:07:25] shipping a wrong product. I think what we are seeing with AI that it really [00:07:30] compresses the lifecycle and as I said just early [00:07:35] on, is you could build something, ship something the same day, [00:07:40] and guess what?

[00:07:40] Chirag Mehta: You ship the wrong thing. So do you see that as, as part [00:07:45] of this strategy, past part of this validation and what, what we, what people [00:07:50] generally refer to as a product sense as, as you know, rightfully described, which is [00:07:55] really know your end users, but now it is giving us this [00:08:00] ability to accelerate these feedback loops?

[00:08:03] Trisha Price: Yeah, 

[00:08:04] Chirag Mehta: We used to wait [00:08:05] for hours, days, weeks, sometimes to get feedback and like, [00:08:10] "Hey man, we, we shipped the wrong thing." We have to go actually go fix it. But now those [00:08:15] feedback loops are much faster. What is, what is your sense, how does, do you see [00:08:20] that feedback loops? Feedback loops are faster?

[00:08:23] Chirag Mehta: And if so, you know, how [00:08:25] does product management as a discipline evolving with that? 

[00:08:28] Trisha Price: Yeah, well, I think it's [00:08:30] magical that feedback loops are shortening and cycles are shortening. Because if [00:08:35] you are strategic, if you do know the business outcome, if you do have product [00:08:40] sense and know your users, these quick feedback cycles allow us to do [00:08:45] true experimentation at scale, right?

[00:08:47] Trisha Price: And so if we know we're laser focused [00:08:50] on our north star metric and whatever that might be, could be a time to value [00:08:55] metric. Could be a conversion metric, could be retention, whatever it [00:09:00] is that you're trying to drive. If you understand it now all of a sudden you can [00:09:05] try three different things or a hundred different experiences [00:09:10] to different subsegments of your population, get feedback and [00:09:15] really nail what that product experience is. What the features are that [00:09:20] you need to build even before you engage engineering. [00:09:25] Right? 

[00:09:25] Chirag Mehta: Even before you engage. 

[00:09:26] Chirag Mehta: Interesting.

[00:09:26] Trisha Price: Yeah. I mean, obviously you should be engaging your partners and having [00:09:30] conversations all along, but you don't need hands-on keyboard for engineering to get prototypes [00:09:35] out and do experiments and get feedback.

[00:09:38] Trisha Price: Not with today's tools that [00:09:40] are out there. And what you need engineering, I always think engineering is my gold [00:09:45] resource. It's scarce and I have to use it really wisely. Now, by the time [00:09:50] I'm asking engineering to build something at scale that users love, my [00:09:55] confidence on the right feature, driving the right outcome is super high [00:10:00] because I've done those experiments and feedback loops with prototypes [00:10:05] I can build in product.

[00:10:06] Chirag Mehta: Yeah, I think. I think I, I would, I I [00:10:10] love this thing. I think what you're saying is the, the high, the signal is of very high [00:10:15] quality. One of the challenges what people have seen, and I've been on both sides. [00:10:20] I've been on the product side and engineering side. Engineers would tell you their product managers don't know anything.

[00:10:25] Chirag Mehta: Product managers would go and say, well, engineers are way too demanding. They just don't listen to us. And [00:10:30] they're both right. Yeah. Because what has happened in the past is that we just never had. [00:10:35] Such a high quality signal from customers from market. So you end up [00:10:40] wasting time building something or modifying something that actually no one [00:10:45] wants, and doesn't actually solve any problem.

[00:10:48] Chirag Mehta: So now given [00:10:50] the accelerated prototyping going out there just [00:10:55] going out there, putting it in front of customers, doesn't, they don't have to be super working, [00:11:00] engineered, you know, highly scalable, you know, kind of prototypes. But you actually get [00:11:05] that, that strong signal that it is actually solving the problem or not.

[00:11:09] Chirag Mehta: So [00:11:10] given, given that and, and given, you know, where this, where this, [00:11:15] this dynamics are going, I think we get the outcome part, which [00:11:20] is you can get to the outcome much faster and you can experiment a lot, [00:11:25] but it's still about cost. I think the [00:11:30] cost is, it's still the cost and companies have been looking at and the companies, the clients [00:11:35] that we advise, one of the questions is like, look, I don't want to do endless [00:11:40] experimentation.

[00:11:41] Chirag Mehta: And because it's still expensive, even if you have [00:11:45] to, even with the help of ai, if you're doing that. So what is [00:11:50] your recommendation? What are you seeing on the cost side? How do you make sure that. [00:11:55] You stay within the budget and you still experiment, but still deliver a [00:12:00] great outcome? It's like having a cake and eat it too, but I hope with AI we can do that.

[00:12:04] Trisha Price: Yeah, I think we [00:12:05] can. You know, look, I think it is very easy to give [00:12:10] yourself a false sense of, oh, I did a great job [00:12:15] and I stayed in budget and got the project done. When [00:12:20] you don't focus on outcomes. Hey, yeah, I can get the project done [00:12:25] now is what I built amazing? And I think you can go the flip side, [00:12:30] which is you can do endless experimentation and also not drive [00:12:35] outcomes.

[00:12:36] Trisha Price: So to me, I don't think it's about, [00:12:40] if you're focused on ROI, which everybody should be, you know, engineering, [00:12:45] technical products, especially in the world of AI, are expensive to build and expensive to [00:12:50] maintain. And so holding yourself and your team as a business person accountable to [00:12:55] ROI is incredibly important.

[00:12:57] Trisha Price: So if you're doing that. And you [00:13:00] say, this is the amount of money that we're gonna spend right now, and this is the amount of time and you've time [00:13:05] boxed something to get something out there. If you're focused on [00:13:10] outcomes, you're not focused on, did it take me 10 experiments or a hundred or two? [00:13:15] You're focused on, I delivered a product that delivers outcomes.

[00:13:18] Trisha Price: And so I think [00:13:20] that's how you hold the team accountable. And I think that your [00:13:25] likelihood of getting an outcome is much higher if you are experimenting than [00:13:30] if you just do a couple of surveys, do a few zooms, talk to a few [00:13:35] important customers, and have a false sense of security. Because it's not [00:13:40] about, I mean, it is about building a product that people will buy [00:13:45] or use, it's often about the details when we talk about product sense [00:13:50] and then product taste. It's about getting the details right that make people wanna [00:13:55] come back and do something over and over again that make the [00:14:00] value. Something they can't live without. And you will never get to that [00:14:05] level of detail without some pretty serious experimentation.

[00:14:08] Trisha Price: And so I think [00:14:10] that I don't think experimentation gets in the way of ROI, [00:14:15] I think it's what delivers ROI. But either way, you're not gonna get it if you don't know your [00:14:20] North Star metric and you're not relentlessly measuring. [00:14:25] 

[00:14:25] Chirag Mehta: So experimentation. Sounds like a [00:14:30] research lab. Hope it's not a physical research lab, but it sounds like... 

[00:14:34] Trisha Price: [00:14:35] yeah.

[00:14:35] Chirag Mehta: ...you know, we are experimenting a lot more than we could have in the past. [00:14:40] And I, I generally joke and, and sort of me being a [00:14:45] Chief Product Officer, I used to say that Chief Product Officers, our chief backlog [00:14:50] officers, they, they, they sit on a backlog. That's what, that's what used to happen in the [00:14:55] past. And because you could not deliver enough.

[00:14:58] Chirag Mehta: And there were [00:15:00] always more things to do, and I think now you're introducing this, this concept, [00:15:05] which is you cannot get something right without experimentation and you need to have [00:15:10] skills to, to be able to experiment. You just don't have to build features and outcomes and whatnot. But [00:15:15] you really have to understand what is working and what is not working.

[00:15:18] Chirag Mehta: So how [00:15:20] do you, how do you think about this concept of research lab or sort of research [00:15:25] driven or experiment driven product management sort of as part of the larger discipline? [00:15:30] 

[00:15:30] Trisha Price: Yeah, I think that one comes back to what I said [00:15:35] before, which is you have to have extreme understanding of the problems you're [00:15:40] trying to solve and empathy and understanding of your users and their jobs to be done.

[00:15:46] Trisha Price: But I think equally you have to understand [00:15:50] how to measure and what success looks like. And it's [00:15:55] not, you know, when I spend time and, and I know you get to too with product leaders [00:16:00] across the world. They know they're trying to drive ARR, right? [00:16:05] They know their new product needs 5 million in ARR, 10 million in ARR, whatever [00:16:10] it's.

[00:16:10] Chirag Mehta: 10 million today and hundred million at the end of the year.

[00:16:13] Chirag Mehta: Yes. 

[00:16:14] Trisha Price: Yeah. They [00:16:15] know that. That's not a question, but that's not a North star metric in [00:16:20] product. That's a lagging indicator, not a leading indicator. 

[00:16:23] Chirag Mehta: That's a lagging indicator. [00:16:25] Okay. Yeah. Interesting. 

[00:16:26] 

[00:16:26] Trisha Price: That takes time, right? I can't tell when I shipped a feature [00:16:30] or a new product this week if that's driving ARR.

[00:16:34] Trisha Price: [00:16:35] That takes so much time. So then you have to start to think, well, this is what's [00:16:40] hard in an, in a research lab, is what is the thing that I should measure? [00:16:45] Right? What is the North star metric that matters that eventually is going to drive [00:16:50] ARR? And so for example, it could be eventually just, are people [00:16:55] logging in?

[00:16:55] Trisha Price: Okay, they're logging in. Are they doing the job to be done and [00:17:00] are they getting it done quicker with my product than they were before they bought the product? Then it [00:17:05] could be, well, okay, what's the real they, they may be getting a workflow [00:17:10] done, but. Maybe that's or, or they may be completing [00:17:15] transactions, but maybe the ultimate value, especially in this era of AI, is they don't have to [00:17:20] do anything.

[00:17:21] Trisha Price: So now you start getting really cute and tricky when [00:17:25] you think about analytics and metrics and outcomes, because it [00:17:30] might be that the true value is that the person doesn't have to do anything at all. But it just gets [00:17:35] done. It gets done by your agents that you're building. 

[00:17:37] Chirag Mehta: Yeah. I say that, you know, the best user [00:17:40] interface is no interface.

[00:17:41] Trisha Price: Yeah. Really. I don't wanna do anything in your product. I just want it to [00:17:45] work. Right. So I think that having that mindset that you just shared, right, [00:17:50] the best user interface is me doing nothing at all. And you just building my [00:17:55] confidence that it was done and done correctly. Like that's how we have to think.

[00:17:59] Trisha Price: And [00:18:00] when we're running this research lab, we're gonna try three different things. Which one built [00:18:05] trust? Which one was most efficient, right? Which one was highest quality? [00:18:10] And then how do I put each of those things together to nail this experience? [00:18:15] 

[00:18:15] Chirag Mehta: Trust, efficiency and quality interesting. And you could [00:18:20] prioritize any which way you like.

[00:18:22] Chirag Mehta: Sometimes you don't wanna be efficient if your job is to [00:18:25] just get trust, then you just focus on that one metric. So I think this [00:18:30] leads to the product life cycle because it the, [00:18:35] you are never working in a sort of, in one part [00:18:40] of the life cycle. Life cycle, you're, you're always as a product manager, as product leaders.

[00:18:44] Chirag Mehta: You are building [00:18:45] something new. You are maintaining something that's going on and you are putting out fires. [00:18:50] There's an escalation and you're on the call. That's, that's so, that's a life, you know, that's a life all [00:18:55] product leaders and all product managers actually live. But the thing is the discovery [00:19:00] to delivery, like as part of Pendo and part of, you know, Pendo platform and products, you are [00:19:05] in this like a smack middle of this lifecycle. [00:19:10] What is changing? I know AI is impacting all parts of lifecycle. We have seen [00:19:15] some in a good way. Some not so much in a good way. What are you, what are you seeing? [00:19:20] 

[00:19:20] Trisha Price: Yeah, I definitely think it is changing the product development lifecycle, [00:19:25] where even though it was, it's been iterative for a long time. You know, there was, "Hey, let's [00:19:30] define the business outcome." Let's plan it. Let's do designs. Let's get [00:19:35] feedback, let's build, let's watch, like, you know, let's ship, [00:19:40] let's enable the customers in the field, especially in B2B software. [00:19:45] And then let's, let's measure, learn. And sort of start back [00:19:50] over. It's now that from business outcome definition, [00:19:55] design, build, get feedback is sort of all [00:20:00] meshed together because we can use these prototyping [00:20:05] tools to build something that feels really real and it is real and get [00:20:10] feedback from it.

[00:20:11] Trisha Price: That whole sort of double diamond process [00:20:15] and portion of the lifecycle sort of gets [00:20:20] all smooshed into one and it's like, Hey, I have an outcome I'm trying to drive. [00:20:25] Here's a whole bunch of different experiments to run. And a lot of that's [00:20:30] happening before build. So I think what we used to think of as build is like, [00:20:35] let's get this minimum set of features out there and get feedback on it and iterate.[00:20:40] 

[00:20:40] Trisha Price: Now if build is like true engineering build that happens [00:20:45] after we've done a ton of prototyping and have a much stronger sense [00:20:50] of our product that we're trying to build and what's gonna work. [00:20:55] So it really just sort of really accelerates that first part of [00:21:00] the lifecycle. 

[00:21:01] Chirag Mehta: I was chatting with one of the buyers and [00:21:05] large financial company and what they shared with me [00:21:10] they use traditional sprints. This two weeks this concept of [00:21:15] sprint zero, which is you get your, you know, designers and PMs [00:21:20] and engineers together. You put a plan and all that. And what he told [00:21:25] me that the end of the sprint zero, we actually have a product. [00:21:30] It's not a working product.

[00:21:32] Chirag Mehta: It's not something that we would actually ship. [00:21:35] But sprint zero is not, not about planning. To your point, there is no double diamond, 

[00:21:39] Trisha Price: right? 

[00:21:39] Chirag Mehta: [00:21:40] The end of sprint zero. You are delivering some, you are [00:21:45] delivering a version of what you're actually gonna ship. Then the build begins, to your [00:21:50] point, people are getting in and now like, okay, how do I bring this thing to [00:21:55] life?

[00:21:55] Chirag Mehta: How do I make sure that this thing scales and this thing has the right compliance? And once again, [00:22:00] this is a bank. 

[00:22:01] Trisha Price: Yeah. 

[00:22:01] Chirag Mehta: Has all these rules. And I think what [00:22:05] we are also seeing that we are hearing this chatter from [00:22:10] non-product leaders as well. And this is coming from CFOs, which is [00:22:15] finally, like, I now have a tool.

[00:22:18] Chirag Mehta: And common set of common set [00:22:20] of set of matrix that I can measure the product teams on. 'cause in the past, as you said, [00:22:25] people would go and say, well, I'm gonna go ship some. They're gonna not gonna find out until the end of the year [00:22:30] whether we increase our ARR or not. So you can't, you can't really hold teams [00:22:35] accountable to that because that's just too complicated.

[00:22:37] Chirag Mehta: So what do we do? So I think [00:22:40] given this outcomes and given this ability to actually get early validation. [00:22:45] We are seeing that more non-product leaders are getting engaged in, in the [00:22:50] lifecycle. Do you see that with your customers? Is it, is it going beyond product now? [00:22:55] 

[00:22:55] Trisha Price: For sure. I mean, I think first is we can certainly hold [00:23:00] ourselves accountable and our peers can hold us accountable as product leaders [00:23:05] to our people using the things that we built.[00:23:10] 

[00:23:10] Trisha Price: And are they delivering value? And I see product [00:23:15] scorecards having to show up in C-Suite. You know, those being rocks that people plan. [00:23:20] And so those are business level conversations. Like it's not an infinite budget, it's not [00:23:25] an infinite set of your cool next idea, but CFOs, CEOs, [00:23:30] and, and really the whole C-suite holding product and engineering accountable.[00:23:35] 

[00:23:35] Trisha Price: To are people using what you shipped? And that's where Pendo is a huge help. [00:23:40] Um, and then in turn, not just did they use it, but are they getting value from it? And [00:23:45] then that turns into to revenue for the company, right? And so I [00:23:50] do see that being a broad C-suite conversation and [00:23:55] certainly one that a great.

[00:23:57] Trisha Price: Finance, CFO leader is gonna care a [00:24:00] lot about, but even the CRO, right? The CROs, you know, [00:24:05] if you've got a leaky bucket, right? 

[00:24:07] Chirag Mehta: Yeah. 

[00:24:07] Trisha Price: This is very applicable in any [00:24:10] software company you're selling a product and if people don't renew. That's [00:24:15] a problem, right? And the CROs spending all this work to acquire new customers.

[00:24:19] Trisha Price: And [00:24:20] if we're losing customers, that's a big problem. And product has [00:24:25] accountability there for sure. And so, again, like that's a great place where you see [00:24:30] CROs logging in and saying, Hey, are people using the product that we sold them? [00:24:35] And then using that data to drive risk conversations of, is [00:24:40] this a healthy customer, red, yellow, or green?

[00:24:42] Trisha Price: Well. I mean, A CRO cares [00:24:45] deeply about that in terms of making their number. 

[00:24:47] Chirag Mehta: Yeah. I was chatting with one [00:24:50] of the, this is a software company, very lost software company. And [00:24:55] they are changing the way the product teams are funded and [00:25:00] the, the retention. There's a very specific bucket [00:25:05] dedicated for just product retention, and they had a version of that.

[00:25:09] Chirag Mehta: Usually it was [00:25:10] funded by go to market teams, but it was very one off in the past saying, "Hey, I have [00:25:15] this renewal with this customer coming up. Some strange thing is going, can you go work on these two [00:25:20] features that's not working anymore." So now what they have done is there is a retention budget. [00:25:25] It is directly tied to whether we can renew someone or not.

[00:25:29] Chirag Mehta: Because [00:25:30] one of the things which I explain to enterprise software or people who are not in enterprise [00:25:35] software, the way it is different than consumer software is an enterprise software. Your buyers are not your [00:25:40] end users. 

[00:25:41] Trisha Price: Yeah. 

[00:25:41] Chirag Mehta: People who are writing checks to you are not the ones who are [00:25:45] using your software and people who are using your software have little to [00:25:50] no influence on, you know, who is buying software.

[00:25:52] Chirag Mehta: It happens, especially in large companies. Versus [00:25:55] consumer software is you're paying, you know, with your money, paying with your wallet. And [00:26:00] so you, you have a say in what you actually use, what you don't use. 

[00:26:03] Trisha Price: Yeah. 

[00:26:03] Chirag Mehta: And I think it's good to [00:26:05] see that this, this rapid acceleration AI compressing [00:26:10] lifecycle is actually making it is [00:26:15] actually favoring end users because now that retention end users are logging in, whether they're [00:26:20] doing something or not, it actually matters.

[00:26:22] Trisha Price: Yeah. 

[00:26:23] Chirag Mehta: So let's, let's [00:26:25] switch to AI a little bit, right? For a few minutes. [00:26:30] Now there's AI obviously that is [00:26:35] impacting the lifecycle and the way you do things differently because you are using AI [00:26:40] to do X, Y, and Z. But also many companies now, they're [00:26:45] just building AI products and that software looks very different.

[00:26:49] Chirag Mehta: That software does [00:26:50] not look anything like that we have seen before. It probably uses some of [00:26:55] the components that we know. It still runs in the cloud and it still has maybe containers and [00:27:00] it uses APIs and it tests certain SDKs. But the nature of [00:27:05] sort of how you build software, as in you train models to, you build agents to, you deploy this [00:27:10] agents to you deploy this models you, the way you work with your customer [00:27:15] data to the way you work with your own data.

[00:27:17] Chirag Mehta: This is, this is very complicated. 

[00:27:19] Trisha Price: Yeah. ?

[00:27:19] Chirag Mehta: So [00:27:20] how are you helping. People build what we call sort of AI native [00:27:25] products? 

[00:27:26] Trisha Price: Well, first, you know, if you go back to Pendo [00:27:30] roots of analytics, we helped you understand if features got [00:27:35] used, what people's workflow or click paths were to [00:27:40] getting a job done and helped you really understand that and optimize that.

[00:27:44] Trisha Price: Well, [00:27:45] the same thing needs to happen in Ag Agentic experiences, except there's not really [00:27:50] clicks, right? We're typing in questions, we're having conversations, [00:27:55] and there's a lot of similarities, yet there's a lot of [00:28:00] differences. Similarities mean I still need to get value. I came in to ask the [00:28:05] software question.

[00:28:06] Trisha Price: Did I click and go run a report or look at a dashboard, [00:28:10] or did I ask an agent a question? Either way, I need the correct [00:28:15] response. Now, here's what's wildly different for engineering and product teams. [00:28:20] If you go back to our Qlik traditional world of [00:28:25] SaaS, things being correct was pretty [00:28:30] binary, right? We built unit test cases.

[00:28:33] Trisha Price: Even if it was an API [00:28:35] call it came back, yes, no, it worked. We clicked it. You [00:28:40] know, the, the answer came back. But when you're talking about having a [00:28:45] conversation. It's not binary. There's not one correct response, right? [00:28:50] This is an agent on the other side that's been trained, and multiple answers could be the right [00:28:55] answer.

[00:28:55] Trisha Price: And so understanding quality and getting evals right [00:29:00] is critical for all of us building ag agentic experiences [00:29:05] and also as product managers being on to able to understand the [00:29:10] conversations your agents that you're building are having and the workflows that they're completing [00:29:15] and what is your user base trying to accomplish, right?

[00:29:18] Trisha Price: You kind of knew what they were [00:29:20] trying to accomplish when they went in and clicked X, Y, and Z. Now you've got to inspect [00:29:25] conversations. To have that same level of knowledge. And for us at Pendo, that's [00:29:30] been a big focus for us over the last year, is building what we call agent analytics [00:29:35] to help product managers really understand those conversations and what their users [00:29:40] are trying to accomplish.

[00:29:42] Chirag Mehta: Interesting. Now one of the things [00:29:45] which actually none of us and none of the customers, you [00:29:50] know, we talked to have a clear answer on so far at least, [00:29:55] is. How do we structure teams and how do [00:30:00] we, there's a golden ratio in the past between product managers and designers [00:30:05] and engineers and Microsoft had one way of doing things and large [00:30:10] company Google did a different way.

[00:30:12] Chirag Mehta: And the industry kinda adopted a [00:30:15] sort of version of that, and we were fine. I think we understood what that good ratio kind of looks like for [00:30:20] your company. Whatever you are doing. That's falling apart. I don't think that [00:30:25] ratio is true anymore. We are also hearing and this is sort of [00:30:30] unfortunate, where, hey, I can replace a junior engineer, you know, with AI.

[00:30:34] Chirag Mehta: I don't need to [00:30:35] hire anyone, you know, from, from college anymore. It's obviously not true. I [00:30:40] think it's, you know, he colleagues, one of the most bizarre ideas that he has ever heard, [00:30:45] that you can replace entry-level jobs with ai. There, there's no such thing. We, we have known this thing for a [00:30:50] while.

[00:30:50] Chirag Mehta: But the thinking you know, people are still thinking about it. It's kind [00:30:55] of, you know, odd that way. How do you see that? Do you, do you think [00:31:00] ratio is changing? Do you think it's the same thing? Do you think team composition is [00:31:05] changing as you build AI products? 

[00:31:07] Trisha Price: I think it does depend [00:31:10] on the kind of product that you're building and the team you're in.

[00:31:12] Trisha Price: I mean, even AI first [00:31:15] products typically have some sort of. [00:31:20] Backend special sauce, right? Where the work is happening, where [00:31:25] the data's stored, where the analytics might be processed, and those teams might [00:31:30] look like they looked before. But when you talk about in [00:31:35] general. AI is giving engineering scale [00:31:40] and velocity.

[00:31:41] Trisha Price: Can it make great architectural decisions? I don't think so [00:31:45] yet. So can it build software on a shaky [00:31:50] foundation and create a shaky experience? It can. Can it augment great [00:31:55] engineers who have built a great architecture? And a great foundation and give [00:32:00] it scale and velocity. I absolutely have seen that happening and I think that's only going to [00:32:05] continue to expand.

[00:32:06] Trisha Price: So if you multiply that by the [00:32:10] efficiency you get by product managers using prototyping tools. And [00:32:15] not wasting engineering time, sort of in the experimentation [00:32:20] phase and in the discovery phase, and being more confident in the [00:32:25] outcomes that they're gonna deliver by the time they get to engineering. I am definitely [00:32:30] seeing areas where the ratio of product to engineering is [00:32:35] changing and where you might have needed a product manager, you know, in [00:32:40] seven, eight engineers to build a great product.

[00:32:43] Trisha Price: You might get away with a [00:32:45] product manager in three or four engineers, sometimes two, [00:32:50] to build the same great product. But that comes back to the product manager [00:32:55] knowing. The outcomes, having a very strong product sense going through [00:33:00] and curating experiences through their research lab. And then it depends [00:33:05] on engineering being top-notch [00:33:10] architects, you know, distinguished engineers in terms of their foundation building and then [00:33:15] utilizing AI to scale them.

[00:33:16] Trisha Price: But I'm definitely seeing that happen. 

[00:33:18] Chirag Mehta: It's interesting [00:33:20] you say this and I. I've described this, that [00:33:25] people have misunderstood the the biggest hidden part [00:33:30] of ai. The AI initially was seen as sort of the efficiency, [00:33:35] engineering efficiency that you can drive, you can write code, you can ship [00:33:40] code, you can write test cases is great, but I don't think it's the biggest accelerant.

[00:33:44] Chirag Mehta: I think it's a [00:33:45] low hanging fruit, which everyone should use it. If you [00:33:50] can, if AI can write code, you should ask AI to write code. You do. You don't have to do that, but I don't [00:33:55] believe that's the biggest unlock. The biggest unlock is [00:34:00] if you make a right decision upfront. [00:34:05] It'll give you the scale and the velocity that you're referring to.

[00:34:08] Chirag Mehta: So in the past, [00:34:10] even if people made bad decisions or not so good decisions it didn't really mean [00:34:15] much because you wasted some effort anyways. And it took year, year and a half, and by the time you [00:34:20] have no idea what happened. So a lot of good product managers that, that I have [00:34:25] worked with and that I advise they were very frustrated by this [00:34:30] fact that their good decisions were not being rewarded.

[00:34:33] Chirag Mehta: Even if you had great [00:34:35] product sense you became a chief backlog officer because like, what am I gonna do [00:34:40] now? It's a great decision. I know for sure that we need this. I have [00:34:45] customers demanding this. I know we can get revenue. I know we [00:34:50] can get to the outcome, but you could not actually go deliver that.

[00:34:54] Chirag Mehta: So how do [00:34:55] you drive, how do you use ai? To make informed decisions and judgments [00:35:00] upfront so that you can actually use the scale and velocity at the tail end to actually shift much [00:35:05] faster. So that, that's fascinating. This is great. [00:35:10] I would like to ask you one last question. If [00:35:15] you want product leaders to have one [00:35:20] key takeaway.

[00:35:20] Chirag Mehta: If they remember one thing out of this conversation, what would that [00:35:25] be? 

[00:35:26] Trisha Price: I think the number one would [00:35:30] have to be, have a really strong product scorecard [00:35:35] and understand the outcomes you're trying to drive. And I'm not just talking about [00:35:40] revenue at that level, but the specific product metrics that are [00:35:45] leading indicators that are going to drive those outcomes and [00:35:50] then create that research lab and experiment relentlessly to achieve your [00:35:55] outcome.

[00:35:56] Chirag Mehta: Awesome. Thank you so much, Trisha. This was a [00:36:00] fantastic conversation. I really enjoyed it. Thank you. 

[00:36:03] Trisha Price: Yeah, thanks Shaad. Thanks for having [00:36:05] me here. 

[00:36:06] Trisha Price: Thank you for listening to Hard Calls, the product podcasts, [00:36:10] where we share best practices and all the things you need to succeed. If you enjoyed the show [00:36:15] today, share with your friends and come back for [00:36:20] more.