Beyond the seat model: How usage-based pricing is transforming product teams with Brandon Walsh, HubSpot & James Brown, Metronome

[00:00:00] Preview

[00:01:14] James: Hi everyone. Welcome to Unpack Pricing. My name is James Brown, Chief Product Officer here at Metronome. I'm taking over from Scott this week to have a product leader to product leader conversation about usage-based billing, and I'm joined today by Brandon Walsh. Brandon Walsh just recently joined HubSpot as PM Director for their pricing and billing, and I'm super excited to talk to Brandon about all things product and usage-based billing.

Brandon, you want to introduce yourself? 

[00:01:45] Brandon: Yeah, absolutely. Thanks James. As you mentioned, I very recently joined HubSpot, so I won't have too much to say on that topic. But previous to that, I spent the last four years at Intercom leading the monetization teams across pricing and packaging, pricing strategy, billing, as well as quote to cash and all the infrastructure layers that sit there working really closely with all of our go-to-market teams.

And previously to that, I like to think of my career as the intersection of product and money. And before joining Intercom, that intersection was all about banking and FinTech and money where I worked in consumer banking, and then small business lending, and I joined Intercom as the PM for our pricing and packaging.

And all of a sudden that shift around money changed from helping customers manage their money to helping Intercom think about how to best leverage pricing to make the most money, and help customers understand how our pricing helped them manage their money and their bills and products. And this kind of evolution journey has just been one that's been really important to me because pricing and money matters.

It sounds so obvious, but it's so fundamental to everything we do. And I think many product leaders need to spend more time thinking about the money aspect of things and thinking about from the customer perspective of what that money and how they are billed and how their invoice actually matters to how the product is experienced because this is a product experience as much as it is a commercial experience. 

[00:03:13] James: Yeah, I love that. I can tell you one of the most important lessons for my life is money really matters. So I totally, totally agree with that. Well, let's take it, as I said, you know, want to kind of talk about billing, pricing, particularly usage-based and what's happening and its impact from a, you know, from a product perspective or product leadership perspective.

So, just diving right into that, I think there's been... there's been a lot of commentary about the rise of usage-based billing and why that's happening, right? Like, adoption of AI and the push to align price closer to value, power of the CFOs, SaaS proliferation, like all of these reasons around like why organizations are moving toward usage-based billing.

But I'm curious from a product perspective, like how is that like actually changing [00:04:00] our industry, like from a product and product management perspective? 

[00:04:03] Brandon: Yeah, absolutely. Yeah. As a product manager, this probably won't surprise you, but customer experiences at the top of my list. As we think about the experience that customers have around their pricing, it's becoming much more dynamic than it ever has been before.

And this trend is accelerated by AI, as you mentioned, but it's not just about outcome-based pricing, which I think has a lot of the buzz around it. It's just usage-based pricing in general, because so much of the AI workload needs to be usage-based and like this really hit home for me in that I was reading a survey recently from a spin management company called XYLO, and they said that 66.5% of their respondents had said they'd had at least one surprise bill in the past year.

And that they had some difficulty in managing that surprise bill. And so as we see more and more usage-based billing. We're going to see more of these surprises and surprises when it comes to your money is not a good thing. And so, I think that the product managers that are thinking about the in-product experience, need to think about that out-of-product experience, and they need to have a broader perspective of who the user is of their product.

It's not always just the person sitting at the keyboard or managing the AI agent. It is also the finance department. It's the CFO thinking about it. It is the person managing the budget and having to pay for it, and the dynamic between that person and the user at the keyboard as well as the person who owns the vendor relationship.

So you have this triangle of users that you really have to consider in this world because upsetting any aspect of that triangle is going to bend the whole triangle out of shape, which we very much don't want. 

[00:05:41] James: Yeah,. I'm curious, like comparing the kind of overworld of seats and subscription, you know, pricing to more usage-based pricing, how do you see actually product teams need to change the way they operate, right? Like, I can tell you, like, from my own perspective, 'cause I think, in a prior world, which is more, yearly transactions or maybe even multi-year transactions, don't want to be too reductionists here, but like a product team, you're kind of like shipping functionality if it's adopted or not, you know, that it's not as like, directly related to, you know, to the revenue of the company.

But when you get into a usage based world, suddenly every decision you're making is like, you know, far more of a determining factor on your revenue. And so, it does like, change the way I think product teams have to operate. But I'm curious to get your take. How do you see that shift happening from like, prior principles to now?

[00:06:33] Brandon: I think it's a really exciting development, especially for the product managers who care about their craft and what they're trying to accomplish, which is benefit for the user and benefit. For the company, you know, shareholder value. And the reason it's exciting is 'cause you do get this direct connection.

Oftentimes before when you were shipping, you know, you were shipping, you know, do you want to be able to add emojis to a text editor and like maybe you see engagement go up, maybe you don't. But if that emoji is an AI generated emoji that uses a credit to create that emoji and then someone's gonna get a bill or maybe that emoji can be custom to the customer's customer.

And so you're, you're seeing much faster feedback loops between what's happening and how the money gets affected. You don't have to wait till the end of the year for a renewal or even the end of the month for a billing cycle. You can actually watch that happen in real time. And I think that's especially true of more modern billing platforms where you're actually able to stream event level data through.

It's like having amplitude on your revenue stack, not just on your page clickthroughs and on your, you know, user streams. 

[00:07:45] James: I've heard someone say that like when you shift to a usage-based pricing model, suddenly every team becomes a customer success team, right? Like, you have to really deeply think about the success of individual customers because that is your [00:08:00] source of revenue.

And that can vary on a daily basis, weekly basis, monthly basis, or yearly basis. And so I think it forces you, as you said, to, you know, think differently about the things you're shipping, the adoption of what you're shipping, how they're moving the needle on, like this actually like incremental revenue.

And so, I actually think it's for the betterment of product teams, right? Because like you're seeing like immediate value of the work that you're doing, but I think it also raises the bar on the intensity of your focus and how much you have to care about the things that you're shipping into the market for these, you know, for these customers to adopt.

[00:08:35] Brandon: Yeah. I think for product managers in particular, we always like to say we're close to the customer. 

[00:08:40] James: Yeah. Yeah. I'm super customer-focused, customer first, all that stuff. Yeah. 

[00:08:44] Brandon: Yeah. And then when you ask, you know, your average PM one, when was the last time they actually talked to a customer and when was the time, last time that they helped a customer resolve a problem?

You know, the answer's almost certainly too long ago. And they might get a little sheepish and talk about like, 'Oh, you know, they read the reports or whatever, but this model, this every-team's-a-customer-success-team forces that tighter collaboration which I think is very good for our job and for our industry.

And then I think there's like a next order impact of that as well, which is an organizational level impact like org strategy component, which is like if a PM's spending their time directly working with customers as a larger portion of their time, that means they're not spending their time doing something else.

And maybe you need to divide up how your PMs work differently. Like, maybe you have a Palantir-style model where you have a set of PMs that maybe have fewer engineers directly assigned to them working in a forward deployed manner directly with customers. And then you have another set of PMs that are building an infrastructure layer that then the forward deployed people can then build on.

And you have this kind of conversation going back and forth. Instead of your traditional, like you have a pod for your web front-end, you have a pod for your iOS front-end, you have a pod for a platform, you now have these more customer-centric or infrastructure-centric viewpoints. 

[00:10:10] James: Yeah, 'cause there's been this concept, I think, as you know, of outbound product management versus inbound product management. Like, that's a function that has existed for a while. And there is something interesting to think about that like, as organizations shift more toward usage-based and the importance of the product organization, like really deeply understanding the way customers are using the product and where there might be opportunities to maximize usage, which might actually break the boundaries of traditional team structures that more of like an inbound management team might have.

I haven't thought about that. That's really interesting to think about how the outbound role may become actually more important in organizations as a way of breaking down barriers and understanding how to drive adoption to the inbound folks. I think there's also a path where just it becomes the role of the inbound team to better understand how their features are being adopted which is really for the betterment of everyone, particularly as an organization gets larger and larger, how like a an outbound function might play a role there. 

[00:11:10] Brandon: Yeah. And like, it makes me wonder if we're going to see some of the career paths of PMs start to change. You know, like traditionally you see a lot of PMs come in as designers or engineers.

Like, I wonder if we're gonna start seeing more salespeople, because that's going to be a really critical skill is that sales skill, and with AI making some of the engineering parts more streamlined and all, air quote easier here, can you have someone with a less technical background and more of that sales background start to come into the fold of product management?

And I think for a positive way change the discipline to some degree as well. 

[00:11:50] James: What about actually like the product leadership level? How do you see that changing? Because I think in similar way that the product teams are having to now kind of maybe think about revenue in ways that they didn't, you know, need to before have more of like a direct connection into like the lines of revenue of the business.

I think product leadership then suddenly takes more of a role in monetization and the overall monetization strategy that they, you know, may or may not have had before. How do you think about the role of the CPO, of the Head of Product Management in that overall monetization ecosystem?

[00:12:24] Brandon: So, I think it's gonna depend on every business and yeah, every company has their own culture and dynamic and setup and, you know, monetization can live in a lot of different places. But I think your broader point around the importance of product leadership, understanding and being involved in monetization, regardless of whether they previously had a deep role in it or not, it's absolutely critical.

It's something that you can't avoid now. And I'm seeing that, or I've seen that at Intercom in particular with the groups that have more usage-based billing. I would spend a lot more time with those GPMs and directors on the things that they're doing because they understood how their product was directly connected to making money.

And in particular, you know, we had one product leader who owned the phone product at Intercom and. Had to set a whole bunch of different prices based on what Twilio was doing. And so that created a lot of billing challenges. But talking to him was so much more exciting than talking to a PM that managed like a core support help desk experience because with them the conversation was like, well, should we put this feature on this plan or put this feature on that plan?

But when talking to people that are directly connected to the revenue side, they're asking me questions as a monetization leader that I may not have thought of, and they're challenging me in a very good way. And at a leadership level, I think that having a career path that supports those people and accelerates them through is really important.

So I think as a leader, I guess where I'm going with this as a leader, you have to think about the right skill sets. You have to be able to grow the right skill sets and attitudes in your people. I. But then the other point that I think matters here is that cross-functional expertise. You can't just be talking to design and product and engineers all day.

You have to be able to understand the role, especially in like B2B of the sales team. What do they care about? And if you change the monetization, how's that gonna change their incentive structure? How is that going to help them hit their quota or not hit their quota? Because I guarantee you, if you change it and you make it harder for them to hit their quota, you will be hearing about it soon.

Instead of developing those relationships early is gonna help you navigate that in a really important way. And I'd say the same thing is true of the finance team. As things become more predictable or unpredictable, that's what the finance team cares about. They need to be able to see into the future. If you can help them see into the future, you're gonna win friends and influence people and finance as well.

And as you yourself know, you know, as you become more senior, as a product leader, your success depends on the other leaders around you. And so building, I think this core capability will also help product leaders in their own career journey. 

[00:15:24] James: Yeah, I couldn't agree more. What I'm noticing is that when you get to a certain level, like more senior level of product, you then start to get more exposure outside of the world of R & D, right? A lot of times, you have to kinda get to more of a senior level to start getting more direct interaction with sales teams, CS teams, finance teams. And I actually think that, I think the move toward more usage-based models is gonna push that exposure, you know, further down the leadership ladder, if you will, so that now it's not just CPOs who are working closely with the sales team or translating sales team needs into product outcomes, but actually like the GPM levels or the PM management levels, it becomes increasingly important for them to have those relationships, for them to deeply understand how those organizations are operating. Because again, as I said before, every team becomes a CS team. That means you have to learn what it means to be a CS team.

You have to learn how to actually make these customers successful. You have to learn how to sell into them very effectively. You have to just really understand the way that your product decisions are gonna impact these cross-functional departments in a world where the businesses like kind of variable based on the product decisions that are being made. So I think I can say from my own experience, like it was only after a certain point where I started to kind of see outside of the technical world. And I think going forward that's gonna change so that, as you said, you might be bringing in more people from actually non-traditional product backgrounds in go-to-market.

But then certainly, I think people in product are gonna have to more quickly understand the dynamics of go-to-market teams, CS teams of customers themselves in a way that they may not have had to in the previous world. 

[00:17:02] Brandon: Yeah, absolutely. And I think that's only good for the business and for the companies themselves because you can't really understand company level strategy without understanding the go-to market side of it.

And by forcing that learning to come into a really practical way. You're gonna have PMs asking questions that they might not have thought to ask before, which is going to yield better outcomes for the business as they go through their career development and through the product development pathways.

[00:17:32] James: Yeah. Yeah. I think one of those questions that PMs are finding themselves having to ask, or certainly PM leaders are finding themselves having to ask, is about the like, end-to-end customer experience of their, you know, of their products. I want to hear your take on this. I have some perspective on this too, where I think as you go into more of a usage-based world, the.

Kind of holistic customer journey and the way that actually they are absorbing their usage information, their billing information, I think fundamentally changes. And, I'm sure you've tackled this yourself in the past. How have you seen that change?

Like how have you seen the expectations from like an overall customer experience perspective change as companies go into the usage-based experience? 

[00:18:16] Brandon: Yeah, I think that that commercial journey is really important. The overall products experience, as I mentioned before. I think one thing that we're gonna start to see happen more and more here is that there'll be less discussion around a monolithic customer.

Yeah. As PMs, we frequently talk about the customer wants this, the customer wants that, but the company, the customer is really a series of human beings, a series of individuals, and there's no one customer, so to speak. There's no one user. You have to be able to care for these different hats that people wear, if not different people.

And so that commercial journey is experienced by different people that experience the product journey, but they cross over in really interesting ways and they cross over in ways that are especially apparent when you are in a usage-based billing world. 

And so I think this is where you have to think about kind of the spider web of what a friend of mine named Dor Sasson, who's the CEO over at Stigg, another billing company, told me is around a…what he thinks of as the revenue anatomy, which I love. 'Cause it's all those different parts that make up the commercial experience, that anatomy of what you're putting together, which is the purchase journey. It's the ACV, it's the sales cycle. It's how you think about ROI. It's the billing. It's the billing frequency, it's your policies.

And, you also asked like, how have I seen this change? And that thinking about the revenue anatomy is the part that I have gotten a whole lot closer to, especially kind of later in my journey, Intercom as a overall monetization leader, pricing strategist, and particularly with ai, because you start to have to ask a whole lot of questions as you develop any one decision that you make.

So if you, let's say, talk about (inaudible) are doing credits right now, and so do you, when you issue credits, do they expire? When do they expire, how they roll over. And you can think about that as an implication for, say, the finance team or the sales team, because the sales team is looking to make quota at some point.

And so if they sell a bunch of credits and they don't expire at the end of the year. Then, you know, they've cut into quota attainment for the next year or for the finance team, if you have credits that don't expire. Does that actually count as ARR? Is it annual recurring revenue? Is it just revenue?

[00:20:53] James: I love that you're calling that out. I think so many companies are starting to think through this. Like, what is our financial reporting in a usage-based world with committed spend and variable spend, all of these components. I think it's gonna be really interesting to see what the next, like two years are like from a reporting perspective.

[00:21:11] Brandon: And I think that that reporting piece is really critical to understanding this kind of unified commercial journey that, you know, we're talking about here, is you have to be able to instrument every piece of it. And then I. Once you've instrumented it and you can think about the systems involved, and what I mean by systems, not just like the code, the ones and zeros, it's the systems of interaction.

t's the inner, it's that interaction layer or interface layer as well. It's the people and processes. Here, you can instrument that and get a view into it. You'll start to see if you change one piece, what's it going to do down the line. 

[00:21:47] James: I'm curious about your take on the end customer experience too, like actually just like their interaction with the product.

One thing I've found is that I think many product organizations may not actually be fully thinking about the way that suddenly they have to start building experiences in the product around billing that they may not have had before. Right. Like when you're in. Seats in subscription world, a customer says, okay, I, I have a hundred seats and I'm paying, you know, $10 a seat or whatever.

I know what my bill is going to be at the end of every year or whatever the billing period is. But something, when you get into this like usage-based world, maybe you have prepay commits or postpay commits or, you know, just totally variable overall. I think that raises then the need for organizations to provide more visibility, control, even optimization on, an end customer's billing visibility.

I'd love to hear, you know, have you thought about that? Like have you run into that as well and what are some of the ways that you've addressed that from a product perspective? 

[00:22:51] Brandon: Yeah, absolutely. Like, so I think kind of what I call the billing experience here has four key characteristics for a good billing experience.

It has to be transparent, predictable, auditable. And fair. And so transparent means you can see what's happening both before you're going to do it and as it's happening. Predictable means that if you do something, you know what the outcome is going to be. And in AI, that's actually really interesting because you could consume a certain amount of credits or not for something.

So predictability is one that when I developed this framework, it was a pre AI world. And so predictability is a little more challenging, but I think it's something that we should still aspire to. And then auditable is you have to be able to prove to the customer why you build them, what you build them.

And I think that's where a lot of companies have fallen down, is they'll issue an invoice and it's not clear why. Whatever usage was, how it added up to the charge. And then you get on the phone, you have to talk to your sales person, you have to talk to like a billing operations person or a CS person.

It goes back and forth. You have to pull up a custom spreadsheet.

[00:24:00] James: that increases, like aggravation, potentially like churn risk. All of that, you know, or leads customer satisfaction is impacted. 

[00:24:07] Brandon: And then, this brings me to the final one, which is fair, which is kind of an amorphous thing, but it's what would the customer think is fair in this given situation?

It's kind of like a common sense item, but I think it wraps the other ones up nicely when you're thinking about what does this billing experience look like? But those are like nice aspirational things. I'll give you a real world example of how hard it can be to build this sort of thing in.

So at Intercom. One of the capabilities Intercom gives like many SaaS platforms is the ability to send outbound emails. And so if you're gonna send an outbound email, we've billed for that in multiple ways. In the past, the most recent iteration is per email sent. It seemed very straightforward and clear.

And so we enable certain users within the platform to be able to set a email campaign live, and then that will start a billing event. And sometimes it'll be trigger based. So, if a user does X, then send an email. So, it's hard to actually predict how many are gonna be sent at any given time because you don't know how many users are gonna do X like in an onboarding campaign.

And there have been many instances where a customer has come back and said, Hey, I had a huge bill spike. You know, I saw 50,000 emails sent this week. I didn't realize this campaign was gonna hit that many people. Maybe they launched a new product, a whole bunch of people came to their website, signed up, something like that.

They would get really upset. And so the obvious answer in this instance is let people know ahead of time, provide some sort of a learning, learning or warning that you might reach this many people. But then that, then you have to ask, who do you let know and when do you let know? When do you let them know?

So if every time you're gonna set a email campaign live, do you pop up an in-app message that says, ‘By the way, this is going to cost you X dollars to send this email campaign’. 

[00:26:07] James: Yeah. This will cost you a hundred dollars. Does a person sending an email campaign, do they care?

[00:26:12] Brandon: Right. And they're like, I don't know what the budget for this thing is. And maybe then they have to go ask their boss. 

Like, do they have permission to do it? Then you have to set up like who has permission to send or do you send an email to like the billing owner, or do you give like really good in-product dashboarding?

And there's no real clean, easy answer to this because every instance is somewhat unique and there's an obvious answer in that instance. But like in so many things of the product, you have to take that instance and then scale it across dozens, hundreds, thousands, millions of interactions. And this is where like the product experience matters a lot to the billing experience.

And so , you have to ask your que like, ask yourself the question like, who are the users that are going to care? When are they going to care? And oftentimes, how undoable is this thing? And so like a great, like a grace policy. So this is a non-product thing, but having a grace policy can help alleviate a lot of these things.

You know, do it once, it'll get forgiven. Do it again, you've kind of learned your lesson. And then you have to enable the people to use that policy. And so this brings us back to the broader thing that we've talked about, which is product managers need to think holistically and you may end up writing CS forgiveness policies.

‍

[00:27:37] James: Yeah. And wherever your launch, yeah.. Like, for 60 days we'll let people do X, Y, Z, but then, you know, yeah. 

[00:27:44] Brandon: Yeah, absolutely. 

[00:27:46] James: I think, what kind of play off of something you just said, which is that you mentioned the words like product experience and billing experience.

I think for me what's interesting is that suddenly billing becomes a part of your product experience. Like, I think that's essentially what you're saying. And I think what's interesting is that as you're calling out, there's many different aspects to that.

There's the guardrails that you put in place within the product, the alerting, the notifications, are you sure you want to do this aspect? So that means that you have to have awareness of your monetization policies and how certain behaviors in the product might drive those.

Which I think to your point, like a lot of product leaders, product managers may not be thinking about as deeply as they will need to be. But then I think there's also this like whole world, this whole product surface area around like the way customers are absorbing their billing information.

That just is not really a core part of a lot of products today. The way I think about it is you think about all of the rich capabilities that like AWS, GCP, Azure, you know, like the CSPs that have been really usage-based for a long time have built up over the last 10 years, right? So, it's not just visibility, it's not just being able to see what your bill is or understand your consumption in real time. It's also your ability to set budgets within the product and then even optimize your spend. I think that's a really interesting element that organizations I think are gonna have to think about increasingly is that common sense would tell you like, 'Oh, just let a customer spend as much as they wanna spend or their spend can be as inefficient as they want it to be'.

I actually think that opens up like a massive churn risk in like usage-based businesses because as soon as a company realizes that, 'Oh, a bunch of usage is actually unnecessary'. it can open up a wider question around like, well, 'Hey, why did you let me spend all that money? Why did you let me, you know, blow out my budget in that way?' And so I think actually like adding like usage optimization aspects in the product is going to be more and more a retention driver, like an actual, like net dollar retention. I should be positive from like a net dollar retention perspective.

And so all of this stuff is just this whole product surface area, I think, are just things that product teams are just not having to deal with right now. Like, it's all fairly new territory. How do you suddenly become an expert in the end customer experience for billing?

How do you suddenly build all the capabilities that like AWS has provided for an end customer? And I think what's interesting about this conversation is that like we've kind of called out a number of different things in that same vein, like going into a usage-based world means that suddenly you have to be thinking about much more from a product surface perspective than you have before.

So, I'd be curious of just kind carrying on that theme. I think this is the things that we're talking about , is what is leading a lot of organizations to think of billing differently. Like, think of billing as actually a core part of their infrastructure, and then start to think about whether building that infrastructure themselves and getting into all of those the use cases that we're talking about or, working with a vendor to provide that infrastructure.

Like, they're doing that build versus buy analysis. I know, you know, you've gone through that yourself. And I'd just be curious, from a product leader perspective, like what kinda led you toward more of a buy versus a build decision? 

[00:31:13] Brandon: Yeah, I think this actually connects really well to, to the point that you were just making about how it's hard to have to think about all this surface area all at the same time.

And it reminds me of what a mentor of mine used to say, which is, it's hard to think from first principles on day four. And what he meant by that is establish what matters to you early and then be true to that and help and use that to make decisions or use as a decision making framework as you move along.

And so, like to your prior point, like how do you think about all these different components set up something around like, what do you want your billing experience to be? You want it to be fair, transparent, maybe you want it, maybe you have a principle of revenue extraction. Maybe that's like. Actually a valid tenant for you and your company at this time.

And so if that's the case, then you should think about things very differently. I think revenue extraction is a bad idea. You should think more long term. But like, if that's the principle, then you'll start thinking about it differently in that term. And I think this also applies to your billing infrastructure and how do you think about infrastructure as a company first and foremost.

And so Intercom has a great principle called Run Less software. And I think that it might have actually been, it might have come up with my new boss at HubSpot, a guy named Rich Archbold. And the idea there is that build the stuff that matters for your business. It's the same as Amazon saying only focus on things that make your beer taste better.

And I think that building infrastructure is very much one of those things. It is all about outsourcing the thing that is not accretive to the direct value that you're adding. And instead go to a best in breed provider who can enable you to do your product work better and create tooling and foundations that you can then build on.

And that was very much the journey at Intercom is that, you know, early on early days we were PLG-only company and it meant that, you know, billing was pretty straightforward. It was really a Stripe integration and set it monthly and you're off to the races and then you add on a hybrid billing model, you add a sales-led motion, you add multiple new product streams, you add-ons, you have different contracts with different terms, and all of a sudden the complexity explodes and you need something that can think through all that, and it requires a lot of people thinking deeply to be able to set that up correctly. 

And at Intercom with this principle of run less software, it was let's outsource that to a best in breed provider and let's be able to build on top of what they're doing so that we can focus on what we're doing, which is trying at, you know, at Intercom's case is that's to enable customer service teams to be able to deliver the best and highest quality service through the use of AI tooling.

[00:34:18] James: Yeah. I think... obviously, I'm biased here, but I think there's a few things that I think about, like you, you just mentioned the management of the complexity of models, right? Like, that takes a lot of cycles, experimentation of models, right? If you want to be able to try certain things out and cohort certain customers or do some tests on a new pricing model, all the end-customer experience stuff that we were just talking about, that whole product surface area.

Something we haven't talked about is you mentioned like finance teams, but the reporting and all of the stuff that they need on top of the billing system.

All these things are not really like a core competency of many organizations, right? To your point, it's not really where a lot of organizations should be spending their dollars. It should be on building the solutions that are actually going to lead to the charges.

Like, to the revenue coming in, the infrastructure, the ecosystem billing is becoming increasingly complicated and I think that's opening the making people kind of rethink as they go from a subscription business or seats business into usage-based, starting to realize how complex it is and how much flexibility is needed, and ability to move quickly, it's opening that, you know, that conversation. And something I think about a lot is ultimately, you know, particularly in a usage-based model, you really are only as fast as like your billing system will allow you to be, right?

I think that's like a really important point is that when you're just kind of taking features and you're putting them into your existing packaging, like you can move fairly quickly there, but suddenly when you get into line item revenue, things like that, like your billing system has to be able to keep up with your shipping velocity.

Otherwise your like overall velocity will deal, will decrease. And I think that's very dangerous for an organization and is leading the companies saying like, 'Well hey, wait a minute. Do I want to manage this whole thing? Do I wanna to have to be like, you know, do I wanna have the burden of building and maintaining this ecosystem and potentially slowing down my teams or do I wanna think differently about this? Go with an infrastructure provider that can do all these things out of the box and then just focus on making money. 

[00:36:34] Brandon: Yeah. And I think that managing the ongoing piece is a huge one and it's something that people don't think enough about. Like, it's easy to think about the billing is that kind of last mile.

You do all the product work and then like, you have to be able to charge for it before you can go live. And if you are not able to do that charge piece, you get delayed going live. And, you know, that certainly happens. I talk to other leaders in the space and it happens and it's the easy thing to bash on.

But the much harder part, I think, is the legacy components that you create. So what happens if you create three new products with unique billing models and you now have cohorts of customers on these unique billing models? And then you decide that, you know what, you did your experimentation, those aren't working. How do you then move them to a new model? Or how do you grandfather them? 

And then that creates a lot of complexity because then your data and systems need to understand different customers with different billing profiles. The financing needs to be able to predict against that. And, you know, maybe you have a perspective.

I don't know, like how does Metronome think about enabling customers to manage that backend complexity? Because that's something that, you know, at Intercom and other companies I've talked to certainly becomes a big headache is all these different components as you are trying to iterate, you now have iteration debt that you have to pay.

[00:37:54] James: Yeah. Like I love the call he made. In so many systems, you're almost managing customers on like a per-customer basis, right? Like you have a customer with a contract, and maybe that contract is like a copy paste of something else, but you have a thousand customers, you have a thousand different contracts, you have to migrate your customers over.

This is something that we've actually invested , this speaks to the infrastructure component of the platform. We've invested a lot of attention here in the management and scalability of just like our data platform with customers. I think we've talked about this in the past, but you just kind of recently released what we referred to as like Metrodome 2.0, which is built on this concept of a rate card, which is a kind of singular pricing construct that you can align many, many customers to. And so when you have to make a change, like using your example, like you have that cohort that's on a certain specific rate card, you're just changing that rate card. And that's kind of like immediately propagating, throughout those customers versus what we see a lot of customers having to do in their home build systems or, you know, maybe with other vendors, which is more of like a programmatic change that has to proliferate across their customer base, which can ultimately take a lot of time and be a pretty big headache.

But I think that's like a great example of like, you know, we're doing that because we're thinking about this from a long-term perspective. We're thinking about those problems where I think a lot of organizations that are, you know, certainly going down the the build route, may not be thinking like those 10 steps ahead, right? May not be thinking about like, well, what happens when I have to suddenly change the pricing for a very specific set of customers, but that's a thousand customers or something. Like, how does that, how do we do that? How do we make that work fluidly and effectively?

[00:39:39] Brandon: Yeah, and I think a lot of it comes down to contract management as well, and customer management here, which made this maybe the steps outside of like metronomes area. But like I've found that, you know, even if it is easy to change a rate card, it's telling those thousand customers, like let's say you're raising the price, you have to go tell them like, 'Hey, this is gonna cost you more money'.

And then what if those thousand customers, you know, some have a contract signed in January, some in April, and some in June, your rate card has to incur, like the new price will only come into effect when they renew and managing that component and like contacting and working back. With the CRM and the contract management system becomes the real technical challenge.

It's not actually changing it just in billing or like just on a contract. It is managing a customer level. The real world impact for them. 

[00:40:34] James: The relationship and yeah, the navigating of that. Yeah. Yeah. Yeah. I know we're coming up on time here, but I would be remiss, like we cannot end in today's world, we can't end a podcast without talking about AI and talking about agents. And I have a two-part question for you on AI. I think the first one is just how do you see AI actually helping optimize the pricing world? Like the world of pricing?

Like, think about your daily job. How do you see the rise of AI in agents actually influencing the way you manage monetization, manage pricing strategy? 

[00:41:06] Brandon: You know, the part that I'm really excited for is. Like conversational reporting and data analytics, which I think is universally applicable, but certainly applicable monetization.

Like I would love to be able just to ask a data set, you know, how many customers are on annual billing that we signed up last month in North America and not have to write a SQL query or not have to go find a saved SQL query and then be able to rapidly use that and then combine it with another question.

And so I think that AI is going to make the data burden of doing this job a lot lower and easier. So that's kind of like the selfish inside baseball piece is like I'm looking for other providers to help us do our job. But then I think the other part that's gonna be really impactful is how AI helps run the like, especially like the billing ops function.

So many times when a customer comes in and says, I have a question about my bill. If you're in a, if you're, it's like a B2B company with complicated offering, it can be really difficult to answer that question based on maybe there's proration, maybe there's true ups or even true downs. Maybe there's credits involved.

Even if it's something as basic as seats, like Windows seats might expire. If there's a monthly component or quarterly component, it gets very complicated. And so the first thing is, I think the first layer is enabling a conversational interface for the customer to be asked that question to an AI to get an answer, to release some of the burden from like a billing ops team that'll take some portion of it, but there's complexities that aren't going to go anytime soon.

The next piece is a lot of times. What billing ops folks have to do is they have to log into six different systems. Then they have to pull down the context from those systems. Then they have to put that into a spreadsheet, and then there's a judgment call that has to be made. And I think AI can help with each part of that.

So one is the log into the systems, get the information. So, can we as an industry, get information to our operational partners faster and easier to make their job like the collating the data faster? Next step is go ahead and build that spreadsheet for them, knowing what they're looking for, learning from what they're doing.

And then the final thing is judgment. Like, that's gonna be the hard part I think for people to give up is where does the judgment call lie? And how can you enable AI to clearly understand processes and be able to work directly with a customer to make that happen? 

[00:43:54] James: it's gonna be really interesting to see what the next, I mean, even year, holds.

When you think about like the billing ecosystem you have, like just to make usage-based billing work, right? You need usage, you need your revenue data or your usage to revenue. You need your customer data, right? Like customer information.

And then also some amount of cost information, right? Like the cost of actually like delivering your services. And I think as you think about like that, like data lake forming, there's this very rich opportunity for AI, like agent-based workflows, to help to your point, like optimize a lot of the decisions that are being made today, maybe inefficiently or manually or however they're done.

You kind of hinted at this, but I think there's a very interesting opportunity for product teams and product leaders to. Actually leverage their billing systems and their usage information to make just better product decisions, right? Like, where aspects of the product that are being like really heavily used versus others, how does that usage lead to monetization, identifying areas where you might wanna actually change your monetization strategy to align around usage.

These are all things that like can be automated. Then you get into things like even broader ramifications around sales compensation planning, or a lot of like the finance workflows that are being done largely by hand. It's gonna be interesting to see one how AI can disrupt those.

But then two actually, is that something that like a billing vendor billing platform should be delivering because it's just a part of that overall billing ecosystem. So, it's gonna be really interesting to see where that goes. 

[00:45:36] Brandon: And I think , it willow, it will allow pricing strategists to make more multidimensional decisions easier and faster.

And what I mean by that is, let's say you implement a new pricing strategy that's usage based. Let's say when you do that, what's going to happen is you're going to get some amount of money from the usage increase there. But is the sales team discounting something to compensate for that? And so is your total a RR picture increasing by as much as you thought it would?

Is your discount policy or your discount discipline getting eroded? So now you're thinking about the sales component and then maybe this is really easy to onboard to, and so you're seeing billing spikes. And then on the backend what's gonna happen is you're gonna have a whole bunch of refund requests.

And so your CS team is now going to be issuing refunds. So if you made. A million dollars, but then it's discounted by 10% and then another 10% of that is given back in refunds. You really only made $800,000. And then you have the COGS component of AI eroding that. And what we see today is it's easy to see the million.

If you're doing a discount analysis, you can look at discounting, but that's a horizontal view across everything. And then you, or maybe you're looking at your refunds policy and the amount of refunds it's connecting and this happens across the revenue anatomy everywhere. But in just in this example, you have a price point, you have a discount, and you have a refund.

And it's so hard to see how those things interact, especially in anything approaching real time. 

[00:47:19] James: Yeah. Yeah, totally. So, second question on AI is there's a lot of, I'm sure you're following it, there's a lot of discussion around agent pricing, right? Like, as these agent workflows come out, how do you think about pricing?

Things like outcome-based pricing has been increasingly a topic of conversation Intercom was actually like, on the cutting edge, right? I think a lot of people point to Intercom as an early adopter of outcome-based pricing or outcome like based pricing over the last couple years.

I'm curious to get your take, like, how do you see the world of agent pricing evolving? What are your thoughts on like the current state of outcome-based pricing and some of the like, pros and cons of that approach? 

[00:48:02] Brandon: Sure. I'd say that again, it goes back to first principles, like when thinking about outcomes based pricing, I.

Anybody should be asking themselves, does this align with the principle of what we're trying to accomplish or how we want to build product or how we want to make money? And sometimes it aligns until sometimes it doesn't. Like jumping into outcome based pricing for something that is like inherently not well designed for, it is not going to be good for the company or for the end user.

It needs to be very specific. And Intercom was in a very fortunate position in that aligned very much with the customer first positioning of how we build product or how Intercom build product. And it aligned really well with what customers are looking for. So for those of us, for those of the audience that aren't familiar, Intercom has a AI bot called Fin, and it's priced based on resolutions.

And a resolution is if a customer asks a customer support question and FIN can answer that question without. Having it passed to a human, that counts as a resolution. It's charged at 99 cents. That makes a whole lot of sense in a customer service world because Fin can gather the context, understand what the user's trying to ask, and then be able to provide a direct answer.

And so when we were designing that pricing, we weren't designing to be outcome or success based or anything else we were designing for what do our customers want? Our customers want a great answer. They want it fast, and they want it at a price point that is better than what a human can provide for them.

And that led us to outcome oriented pricing. And this was, gosh, about two years ago now. So before anyone was really talking about it... 

[00:49:57] James: ...before it was cool. Yeah. Before it was cool.

[00:50:00] Brandon: Yeah, exactly. Exactly. And so, it has its challenges here and there. But I think companies really just have to start with what's good for their business and what's good for the customer, and then ask that right series of questions, it'll lead you to probably where you should be with it.

[00:50:20] James: Yeah. It's interesting. I do think it's becoming such a popular topic and I do wonder about the how quickly organizations can get to that 'cause I think exactly to your point, there's a risk of doing something that's inorganic that doesn't actually align with what the customers are looking for or just, you know, how they want to think about their own pricing.

At least in the near term, I know like for as many people as there are talking about outcome-based pricing in a positive way, there's also people talking about the predictability of spend on outcome-based pricing and how that might be kind of a scary thought for a lot of enterprise companies who budget and have line items for their vendors like, more predictable spend with a current seats and subscription model. And then when you go into outcome-based world, suddenly it's like, you may be able to predict how many support tickets there are on a monthly basis. Like that may be relatively stable, but the number of resolutions that an agent can actually have on those support tickets can be pretty variable.

And, was it entirely that agent that resolved it or was it some combination of agent and human and what does that mean? So, I think it's really, really interesting where this is going to head. I do love, though, the idea of being able to, as agent technology progresses, it really is labor, right?

It's labor in the form of the machine. And so if you think about an agent that is able to do coding at the level of senior engineer is that, do you price that just by the headcount cost of the engineer? Is it priced based on the value of the output that the engineer creates, right? Which could be multiple million dollars a year. Is it some percentage of that? What does that look like? And, I think that's where a lot of organizations will head. I just have a feeling that for the near term, for most companies, what we're kind of more like practically seeing is more of a modification of a seats-and-subscription business to include more usage and then, you know, maybe gradually over time, head toward like pure usage or even like outcome-based, outcome-based pricing. 

[00:52:30] Brandon: I think the ag agentic comparison to labor is a really apt one, and it helps you think through this in different ways because if you think about how labor is priced, it's not outcome based most of the time.

Now there is some types of outcome labor that's outcome-based and, and that's hybrid. Like think about a salesperson, you know, you're paying them on deals closed, switching their quota 

[00:52:54] James: variable for the deals they close. Yeah. Yep. 

[00:52:56] Brandon: They get some sort of base pay, and maybe that's their seat cost. But then you have people that are paid an annual salary, you know, you pay them however much work they do, you're just making a bet on that person that they're gonna do a job.

And if they don't do that job for whatever that annual salary was, well. They get let go and you pull somebody else in, but it's too hard to try and break it down. And so maybe that's like an annual unlimited commit version as like an annual salary. And then you have some people that are paid on hourly.

So you know, someone in retail fast food or something. They're getting paid some amount of dollars or euros per hour to do a job. And some agents, depending on the task, should probably be paid based on pure...

[00:53:39] James: Just runtime. Yeah, exactly. How long are they running or how, yeah, 

[00:53:43] Brandon: Yeah, exactly.

And so like, I think this labor analogy actually enables you. You know, like you and I have talked about in the past, like nothing is new in this world. And so if you use the same analogies in the past, it'll probably help us think about where we're going to be in the future. And which brings me in something else you and I had talked about in the past, which is like, it's a bridge to the future.

Like, we're in a world now where I think hybrid pricing is going to become. Well, it's probably already the dominant pricing model, but it's going to continue its dominance. We'll see seats start to slide, but you have to cross that bridge carefully. Otherwise you're just gonna, you know, burn one side down while you're trying to still get the bridge to the other side.

[00:54:26] James: Yeah, I totally agree. And well... we'll end on this, which is, I'd love to get a hot take from you. I'd love to know, like, where's Brandon Walsh's hottest of the hot takes for this space? Mine would actually be that, you know, what you just said, which is I think if you look at what OpenAI is doing, right?

Like, OpenAI is selling subscriptions or selling, you know, seats. But it really is like usage with a wrapper of something that we are familiar with, which is just like a monthly charge, but you can only do certain number of things per package. And so it really is like a subscription or like usage that's pretending to be a subscription. And I think to your point, I think that will actually be the dominant model for many organizations in the near future. I think it's easier for the organizations themselves to understand just internally. And then, I think also for customers to absorb, right, 'Okay, I'm gonna pay $25 a month, and then if I go over a certain number of whatever, you know, certain number of AI agent invocations or actions, that's gonna be an additional $10 per hundred actions or something like that.

I think that is like the training wheels version of usage-based pricing, which will then put them on, as you said, that bridge maybe heading toward pure usage, right? Maybe at some point heading toward actual peer usage or pure outcome. But for the, you know, near future, they're probably in the middle of that bridge, which is more of like a hybrid between seats and usage.

So that's my hot take. It's a little cheating 'cause you kinda just said that. But what's yours? Like, gimme a hot take on the space. 

[00:55:58] Brandon: Unfortunately, I've never been great at being hot or super controversial, but I think that this part of the world and the era that we're in lends itself to that a little bit more in that the... like, the thing I used to say is don't innovate on your price.

Like, do what the market dictates and be really good at the product. So, if all of your competitors have a good-better-best pricing model, you should probably have a good-better-best pricing model based on seats. And if everyone has a mid-tier that is somewhere between, you know, $50 and $80 per seat. Your price should probably be between $40 and $90.

[00:56:36] Yeah, 

[00:56:36] James: yeah. Somewhere in that range, 

[00:56:37] Brandon: Depending on where you wanna position everything else. 

[00:56:39] James: $49.99... maybe. 

[00:56:41] Brandon: You know? Yeah. It doesn't pay to confuse people with pricing innovation, because buyers are busy and their lives are complicated, and every decision point that they have to make on the path to buying your product is another decision that brings them probably a step further from buying your product, because people don't like to do extra work if they can avoid it.

So take that piece of work away. That said, that was the advice that I would have previously given. I think we're now in a space where buyers are more tolerant of pricing innovation, and they're ex, quite frankly, they're expecting it. And so my hot take here is be bold with your pricing and realize that it is not a one-way door.

Pricing has always been considered a very difficult thing to change. And it is difficult to change, but it's not impossible to change. And there's so much value sitting out there to be grabbed that if you were just following the same well worn ruts, I think you're gonna leave too much money on the table and get out competed.

And so make a bet that may not work. And what's gonna happen is we're gonna see lots of companies making bets. Many of them won't work, and there's probably going to be some spectacular failures coming, and that will be exciting. It'll be all in the press. But what will happen is we'll reach a new point of stabilization at some point relatively soon in the future, and people will align on what's the right thing to do.

And then we're gonna go in this innovation cycle again where we'll be, it's obvious that everyone prices this way. I don't know what that obvious thing is yet. You don't know. Nobody knows, but we're gonna get there soon. We'll stabilize, and then the next thing's gonna come along. And so lean into a period of disruption while you have permission to do so, and make a big, bold bet 

[00:58:35] James: And be willing to change. 

[00:58:36] Brandon: Exactly! Iterate on it and be fast, and don't hesitate to iterate and iterate quickly as soon as you're seeing that it's not playing out or double down if it is playing out. 

[00:58:47] James: Yeah, that's great. And it goes back to what we said before, which is also means you have to have an infrastructure that allows you to do that, which is, I think, an important through point.

But Brandon, this is awesome. Thank you so much for joining. You know, really appreciate it. And for everybody listening in, thanks for joining in the conversation. Hopefully I'll be doing many more of these, like talking to product leaders like Brandon in the space about, trends and different things that are happening.

Thank you so much for joining and we'll see you soon.

[00:59:14] Brandon: Likewise. Thanks James.

[00:59:15] James: Thanks man. Bye.

[00:59:17] OUTRO: Thanks for tuning into this episode of Unpacked Pricing. If you enjoyed it, we really appreciate you sharing it with a friend. We'd also love to hear from you. Feel free to email me@scottatmetronome.com with feedback and suggestions for who you'd like to see on the Future Podcast.

‍