43. Is usage-based pricing the answer?

Brandon 0:05

Hello, everyone, and welcome to another episode of the operations room a podcast for CEOs. I am Brandon Mensing, and joined by my lovely co host, Bethany airs. How're things going, Bethany?

Bethany Ayers 0:15

I don't know. I cut all my hair off yesterday. I'm not sure if that was the wisest. Yep, my husband likes it.

Brandon 0:24

Okay, that's a positive.

Bethany Ayers 0:27

I think it's a bit puffy. After he cut it, he had to blow dry it to make sure it was even, oh, he comes to my house, my hairdresser. And so unlike a normal salon experience, I don't get to see what's happening until the end, because he's just in my kitchen, and I don't have a big wall of mirrors in my kitchen. And so he's cutting and there's just hair everywhere. And then he blow dries and he's like, Okay, go and see what you think. And I went upstairs. Oh, my God. I just looked like a mushroom cap. I didn't realise I had that much hair. And it was just this big bowl of hair. And I went back downstairs, it's just like, it's puffy. And he said, No, no, it's just because I've low dried, it is absolutely fine. Just tuck it behind your ears. And I feel like anytime the advice is to tuck it behind. It's maybe not really what it should be. So I washed it, I air dried it. And it was just as puffy. I woke up this morning, and it's all of the front was just sticking straight up. And the rest of it was really puffy, I think we can thin it out. So that it naturally maybe get a little thinner, a little shorter, a little thinner, and a little bit less like a mushroom on my head,

Brandon 1:49

you've obviously cut the length of it quite substantially, why it

Bethany Ayers 1:55

this passport photo of me in: 2004

Brandon 2:26

decade have an urging to chop your hair off essentially. Perfect. So we have a great topic today, which is the impact of usage based pricing. We have an amazing guest for this and marks diving, he is the founder of impact pricing. And as a former longtime instructor of pragmatic marketing on the same subject matter, which is pricing. So before we get to that, Bethany, let's dig into usage based pricing, I

Bethany Ayers 2:51

have to say that there are parts of usage based pricing that I think are really clever, or can be. First of all, I think of usage based pricing is not necessarily seat based. I don't like seat base for a couple reasons. One is unless it's technology that really is used by loads of people and so a seat makes sense. And it needs to be personalised like Gmail, Salesforce, HubSpot. HR is you know, all of those, you need to have a proceed because everybody in the organisation uses it and you can't share logins. But something like Marketo, or any of the single or like limited number of seats, people will share logins if your seat price is too high 20,000 pounds for one seat, and I have three people, I don't want to pay 60,000 pounds. But if you structure it like 40,000 pounds for access and X number of records, or whatever usage it is plus 5k or 200 per seat. And these other things, suddenly somebody will pay 60k 100k without realising it. But if you tie it to seats, they'll game the system.

Brandon 4:02

I think what we're seeing fundamentally is a trend line. And the trend line is not to usage based pricing, the trend line is to subscription packages that are tiered that basically take the value metric, whatever the value metric is. And if we go back to my last company, the value metric in our case was transcription. So the number of files that you're transcribing so the more you transcribe, the more value the customer is getting. And that value metric, we would basically bend it for packaging, we'd have a bigger band available for Bambi. And we have an even bigger band available for trc in this case, so we're just like a very clear way to have predictable packages both for the customer and for ourselves. That was a mix of seats combined with a usage based pricing element within those packages. And I think what we're seeing with most SaaS companies is a variant of that world where it's no longer straight up seats and it's no longer a straight up or they used to base pricing is not really the model Ford either. But really this kind of mashup of the the two.

Bethany Ayers 4:57

I would really love to just get rid of C It's and tie it to a usage trace, like if I were working in another company again, that would be my goal is to really figure out what your value metric is, you know, what's the perceived value of it. So for transcription is an obvious one, because it's both usage. And that is the service that they're getting. And so it was like, by definition, the value, and then to have unlimited seats, so that you have absolutely no friction for having people give it a go at the company, but to have the predictable tears. But it's remarkably hard to figure out that one value or usage metric. And a lot of times at least I found that have you had that experience as well, Brandon,

Brandon 5:45

it's always been fairly clear where the value is coming from where the usage is coming from outside of the seat. And that way, it's been fairly clear, I think, historically, in terms of what value metric to use in this case. And I think for most companies, my intuition is that it's not that challenging to find the single value metric that you can actually lean on to, to scale. Can you throw an example at me where it will be? Not clear? So

Bethany Ayers 6:06

I guess compute at some level, or storage in some level, are not really value and it ends up tying to somebody going well, I could just go and store it at AWS for X amount. So why are you charging me Y amount? So then it's like, okay, well, do we use calculations or decisions, or something kind of abstract that the customers may or may not understand? Is like, how do you make it something concrete, but not storage?

Brandon 6:39

Or compute, a customer can wrap their mind around the usage of the number of contacts you're gonna have in HubSpot or the amount of transcription files they want to be able to process. I think for the customer, it's quite difficult, I think, to understand what their usage actually might look like in that case outside of the vendor, supplying all sorts of calculators and things to help them understand. The

Bethany Ayers 6:59

other question that's a real nightmare for customer success, in particular, I would say is, when customers are not using as much as they've signed up for, and they know, they're not going to use it. And they come at you six months in of the year or two years that they've signed up for that amount. And they say, Can we downgrade? Now, technically, we can say no, because you've signed a contract. But then you have a really unhappy customer. Which way do you go.

Brandon 7:29

So this is all about the range of sales flexibility that you should have to allow the sales reps or the account managers to ensure that come your to the customer is put in the right position where they're again, really excited and incentivized to go down the year to path and also props to this case, ensure that the customer success and the onboarding, and the kind of rollout within the organisation is more successful in year two than it has been in year one and commitments around that. And so that's some ways to that customer, again, to make them feel like they're gonna get good value for whatever they're paying for.

Bethany Ayers 8:00

And so do you think that if you're a new company, and you don't have the predictability, and you've gone for what you think is the right usage metric, or value metric, you should do multi year contracts to start with, or best to do one year and see what actually works as the vendor? Because you know, what, if you end up with a bunch of customers, and you've gotten the metric wrong, and you've gotten your values wrong, and you have a bunch of pissed off customers who have two and three year contracts, with committed revenue, what do you do, then? I

Brandon 8:30

think a year contract is the contract to be signed, because I think both for the buyer and the seller, you're always trying to ensure that your packaging and pricing and your fit is the right fit. Let's finish on commissioning. So I think that's an interesting subject. Because in pure usage based pricing, if it really is that then you're signing contracts at I don't know what zero value to begin with. Potentially you're signing up a zero rated contract, which is you're just waiting for them to actually do something useful to pay you money. Reading

Bethany Ayers 8:59

Frank lumens book, The snowflake book, that's one of the first things he went in and did. So snowflake, everybody was being commissioned on signing contracts, but not actually on the usage. So commission was separated from usage, which didn't drive the right behaviour, overpaid reps, and there was a lot of lack of predictability. So one of the first things he did was come in and re structure the commission scheme based on the usage and the actual money being seen. But it was a massive culture shift, because everybody was high fiving themselves on signing up customers, and nobody was paying attention to whether or not the customers use the technology afterwards. But snowflake grew massively doing that for a very long time because they had a good product and then professionalised very late in the game, could be an argument for that if you have a great product

Brandon 9:49

or you're taking a huge risk. If you're a snowflake, it all worked out because you had a great product. But ultimately, if you're caught in a situation where the product is a good product or not a great product and it takes a bit more Time to get to the usage levels that you need. You could literally bankrupt yourself, a tonne of cash could be going out the door for a much longer period of time putting you in a very difficult position come your next funding round as an example, even if you have a bunch of great contracts that are slowly but surely gaining traction from usage standpoint, but are not coming on board as fast as you'd hoped. But I think this kind of begs the question of that scenario, especially when it's like $0, on the first day of signing the contract. How are you commissioning people? I think just rethinking the funnel fundamentally, would be required in terms of what are we expecting from the sales rep? And what are we expecting in terms of payback periods to ensure that the commission is going to make the most sense? What's

Bethany Ayers: 10:39

the role of a sales rep, I can't decide. So sometimes I feel like no, all you want the sales rep to do is sign as many contracts as possible be a new business Hunter sign sign sign, throw it over the fence. And I think it comes down to deal value size. So if it's a slower value, deal, have them sign as much as possible, throw it over to customer success, or onboarding, or whoever it is to get the usage out of it. And you just have these new business machines, who are grabbing up as much of the market as they possibly can. Or if you're in more enterprise sales with larger ticket values, they might be going in with a very low contract, but their job is to heavily sell and do basically the CS job throughout that year to grow the contract from 10k to 150 200k. And then it's worthwhile having those new business reps be farmers as well.

Brandon: 11:39

Why don't we take a break? And when we come back, we will speak with Mr. Marks diving.

Bethany Ayers: 11:48

Just in case anybody doesn't know what usage based pricing is, can you bring everybody up to speed?

Mark Stiving: 11:53

One of the really fun things about SAS businesses is once we started going SAS we realise there's a new thing called a pricing metric. And a pricing metric is what do you charge for. In the old days, I would buy a hamburger at McDonalds. And so they would charge me for a hamburger or a bicycle at a bike shop, they would charge me for the bicycle. And I used to buy software, I used to buy Microsoft Word shrink wrapped on the shelf at Fry's Electronics, right. So it was a physical product, I would buy the thing. And that was essentially a perpetual licence, but I'm buying the thing. Once we moved to SAS, the world opened up into what is it that I want to charge for. So Google charges by the click AWS charges by the who really knows what. So there are so many different things that we could choose to charge for. Now, Salesforce came along, they were the kind of the early adopters of this whole SAS world. They said they were only going to do CRM in the cloud. And they started user base pricing. And everybody said, Hey, let's go to user base pricing. So we're going to charge by the number of users, Salesforce is doing so well. And by the way, that's still the most common pricing metric in the world of SAS is user base pricing. But it turns out user base pricing isn't always the best type of pricing, what we really want for our pricing metric is what is it that's highly correlated with the amount of value my customers are going to get. And so when you think about Google charging by the click, what they're really charging for is, hey, somebody out there showed interest in your product, I connected the two of you, I gave you value there. So it makes sense for you to pay me by the click. You may recall early in the internet days, they used to charge by the banner ad, right? How many eyeballs saw your banner ad, that was an okay metric pricing metric. But the click is so much better, because it actually showed Hey, someone showed interest in your product that's way more valuable than someone was forced to look at your ad. So so that's when I think of usage based pricing, I think of what's the pricing metric? And can we make the pricing metric highly correlated with how our customers are using the product, the ideal pricing metric is outcome based. So what if I could charge you a portion of the profit incremental profit I created you as a company, if you think of companies like credit cards, or PayPal, they're taking a percent of the revenue. That's really, really good value based pricing. Because the more money a company makes, the more they're paying PayPal or the more they're paying the credit card company, they're actually part of that transaction. Most of us can't get that far. Most of us have to find some usage metric somewhere in between user and outcome that says as people use more of our product, they get more value from our product, and so therefore they'd be willing to pay us more for our product.

Brandon: 14:48

When you think about a SAS organisation, a b2b company thinking about the value metric as they call it, in terms of what they should choose for that and the kind of criteria and considerations around that. Can you give us maybe just a sense of what are those three or four bullet points of consideration because I went through this at a previous company where we had chosen a value metric as an example, that really was definitely how people perceived value of what we're offering, for sure. But very much was the almost commoditized element of the product in some ways. And we had a big debate as to whether or not that made sense, because that commoditized value metric or pricing metric, we knew that over time, people would perceive dramatically less value in that value metric. Over time, we had a bunch of considerations just around, what are we going to take care of, because if we pick that one, it's going to have a certain trajectory for the company, as opposed to picking something else.

Mark Stiving: 15:37

Let's talk about Dropbox, right? Dropbox is exactly what you just said. They, they charge by the amount of memory, although the price of hard disk, the price of storage goes down, and down and down, it becomes much more of a commodity, but they're still charging by the amount of memory that you use. One of the great things about that is that even though the price of storage goes down, the price of Dropbox doesn't really go down. They're maintaining their margins. In fact, they're growing their margins, just because their cost to serve customers is decreasing, without dramatically decreasing the prices along with it. The question that you asked was, What are two or three things that I would think of as I thought of what's the right pricing metric to use? So number one, highly correlated with value? What I typically do with my clients is we sit down and we say, Okay, how is it that your customers get value from your product, list all of the KPIs for your customer that you're going to move, because of the fact that they're using our product. And now which of our features, which of our capabilities are highly correlated with those KPIs? And those are the things we want to be thinking about in terms of what's our pricing metric? Another thing is the customer willing to accept it? Does it make sense to the customer, sometimes we have to teach the customer that this is an obvious or good KPI. Think of Uber trying to train the world about surge pricing. Right? That was really, really hard for them to do, because they they did something new that the customers didn't like or expect. And the third one, of course, is can we measure it? Right? Is this something that we have the ability to measure and Bill for? So it's an obvious answer, sometimes we might say something like the number of leads you get. But if the number of leads don't flow through our SaaS cloud, then we can't measure it, we're reliant on the customer to say, here's the number of leads. I

Bethany Ayers: 17:37

talked about pricing with customers quite often. And with companies and everybody has been built for seat based pricing, even though they aren't necessarily the best way. And particularly since their AI companies, if they sell by seat, often part of their business case is that they're going to be fewer seats in the future, because the computers are going to be doing more of it, or the machines are going to be doing more of it. And then we work on and identify a pricing metric. But then we can't do anything with it. Because it's not that the leads are flowing from somewhere else. It's just that the platform hasn't been built to metre, or throttle or toggle anything. What do we do that?

Mark Stiving: 18:19

My strong advice is metre every feature, take every feature and make it so that you can toggle it on and off, do it from the beginning of your company. Here's why. Over time, as you watch usage of different features, you're gonna see some customers use some features more than other customers, that's actually a strong indicator that you're dealing with different market segments. They're dealing with different problems. And so maybe we eventually create different products for different market segments. That's going to help us understand how we would craft a good, better best product portfolio that's going to help us so that we know what are the features that are highly, highly correlated with value. So now we know which ones we can charge for. So the idea that people should be metering and putting flags on every single feature, it's so crucial in the world of pricing.

Bethany Ayers: 19:12

Brendon was mentioning something that solves one of the operational issues of usage based pricing, which is the lack of forecast ability if it's truly usage based. So if you're charging by compute, let's say and a customer uses more compute, and then other times uses less customers love it, because they're actually paying for the value and they're not tied into something. Sometimes customers love it. Sometimes they hate it if they suddenly are using five times more than forecasted. And as an operator, it's not only you're trying to predict your sales, but you're also then trying to predict your revenue every month. So this conundrum of true usage based pricing versus tiered usage based pricing. Do I

Mark Stiving: 19:53

like tiered or direct usage based pricing? And the answer to that is it depends upon On, do our customers look at this as a cost? Or is it a direct piece of the value? So Pay Pal, as we talked about earlier, PayPal, when they charge a percentage of revenue, boy, I can't wait to pay pay pal a million dollars, right? Because that means I made $50 million, and I'm a happy camper. But if they charged me for some other metric, then what I might be doing is trying to figure out how do I use less of that metric? As it is now I want to know, how do I pay them more, I want to keep paying them more and more money. So there's no games being played. There's nothing hard about charging based on that outcome, because I'm right in the value chain. On the other hand, if I'm doing something like Dropbox, I might be saying, well, how do we as a company use less storage, or if I'm doing a Perseids, I've got one one client of mine that was selling per seat to a law firms, and what they were doing as each lawyer and small and mid sized law firms would use a copy of their software. But when they got the really big law firms, they would hire someone just to use the software. And so there were fewer seats and larger firms, and there weren't smaller firms, right? How do we work around the pricing mechanisms that our clients or vendors are giving us? And so that's one of the things we're thinking about as we do this pricing metric? And do we put it in the tiers? Do we put it in the direct revenue, if it's in the direct revenue, so if I charge per seat, then our clients are trying to figure out how can I use fewer seats? If it's directly in the value chain, then I can charge per item per usage? Because clients want to use more of it, because they're making more money. Let's separate the world into platforms and solutions. So a platform most AI chat CPT today is a platform. Zoom is a platform. LinkedIn is a platform. And so by platform, what I mean is it solves a lot of different problems. In platforms, it's hard to know, who gets the most value, how can I charge someone more value? So platforms, although they do weigh more, we end up paying less for them. And then there are things called solutions. And a solution is a solution to a very specific problem. We can quantify what's the value of solving that problem, it's easier to put prices, we tend to charge more money for solutions than we do for platforms. And solutions are more predictable, because we know what those problems are how often they show up how people are using our product. And now, many, many companies I deal with start out in the world of platform, say we built this great platform, then what we usually advise them to do is hey, go find a single market segment, go find a single solution that you can go build and sell to them. And you're going to charge a higher price. It's going to be more predictable. Quick example. Let's talk about LinkedIn. Right. LinkedIn is a platform. LinkedIn has Sales Navigator, which solves a specific problem for sales organisations. You could think about zoom zoom has telemedicine options. And so they solve specific problems for specific market segments.

Bethany Ayers: 23:08

So you're still talking about predominantly doing some sort of packaging? Like a good better best, most of the time? Or do you think that there is more of a move towards straight usage? Like a snowflake or an AWS? Even

Mark Stiving: 23:29

in straight usage, there are packages, there are good, better best packages. So you could think of usage of the standard capability. But what SLA is do you want? Well, how fast do you want your service to be? How fast you want us to pick up the phone call? How many people do you want us working on your projects. So there's almost always a good, better best opportunity. I think of it this way. Some people get a tonne of value for our product, and are willing to pay us a lot we should find a way to win those customers are really high prices. And some people don't get much value at all or struggle to get value. And we still want to win them. So we have to win them to low prices. Our job is to figure out which customers are which. And then craft a product portfolio. So the people who are willing to pay us a tonne pay us a tonne. One recommendation I have for every company in the world is at least consider good, better best packaging. It doesn't always work. But boy it would 90% of the time that works. And it is so valuable in both subscription businesses and non subscription businesses.

Bethany Ayers: 24:36

So basically the impact of usage based pricing on sales because you're almost de facto moving into a land and expand motion. Do the same people hold it throughout to get to the full commission. Are you paying every new wind is felt very valuable but without building the structure in place? You're not guaranteed that expansion who gets paid Do suddenly new business get paid much less than account management, like moving from a more predictable seat model to usage based pricing creates all kinds of issues in the go to market? And how have you seen companies address this? And what are some of the ways we can think about it,

Mark Stiving: 25:18

even if it wasn't usage based pricing, and it was just land and expand, and I want to get you go from a good to my bad or my better to my best, or I want you to buy new options, you have the exact same problem, which is you've got one salesperson who's trying to win a new account when a new logo, and then you have a whole team of people who are trying to grow that account somehow. Now the first comment I would love to make is that I see way too often, companies say the phrase land and expand, and they spend tonnes of resources on land, and almost ignore expand. And so we have to truly think about how do we expand our customers? This is a real strategy that we need to go implement. And part of that implementation is who's gonna get compensated? Are we watching usage? are we monitoring usage? Are we driving usage, because if people aren't using our product, they're not going to use more of our product, they're not going to value our product more, they're not going to upgrade. In fact, they're likely to churn. And so we need to be watching usage just for that reason. Now who gets compensated for it? My answer is always it depends on do we need salespeople to do the expansion or not. If I'm sending a salesperson in to make the sales call to make the expansion happen, then I'm compensating salespeople for it. But if I'm not sending salespeople in, if I'm using my customer success department, or they're just upgrading because they want to or they're just using more because they need to, then I'm probably not compensating sales specifically for usage. But what that also means is think about the value of a new logo to us. I run into way too many SaaS companies who say we try to sign up a customer with a three year contract. One of the advantages to SAS is I can get a customer into my product portfolio for a month. And if they don't like it, they can quit. And so they're going to try it. And our job is to deliver value and show them that it's really worth it. And they should be using our product. But if we force him to sign a three year contract, we just made the sale really hard. And we took away one of the advantages of having a SaaS product. So if we can bring that back to a one month contract, then a great compensation plan for salesperson as you bring me a new logo, you get this, they stay for a year, you get this. And then from that point on, it's ours, right, we're going to deal with it, whatever it looks like. But the Commission or the compensation you're giving your salesperson is going to be way higher than whatever that original commission rate used to be. Because you're compensating them, you should be thinking about what's the lifetime value of a new logo, if we do our job, right as a company, and

Bethany Ayers: 27:54

then commissioning on that total lifetime. I mean, that requires having quite a bit of cash, you're already investing a lot for future value in SAS, and now you're suggesting investing even more.

Mark Stiving: 28:08

So I'm totally okay with saying, Hey, we're gonna pay it out over time as the client says. But one of the things I love about the SAS businesses and traditional pricing the way people have price acid when they first came out, or they switched from true to transactional to SAS, or perpetual licence to SAS, in three years, you've made back the same revenue. And so if you imagine that you didn't change a salespersons commission plan, in three years, they made the same amount of commission. But then they're making more in year four, and year five, and year six. And a good salesperson builds their portfolio up to where they don't even have to work anymore. They just get paid a commission rate. And just like the company is making tonnes of money on existing clients, so is that salesperson, that's not what I want to pay a salesperson to do. I want to pay a salesperson to go win deals. So

Brandon: 28:53

pricing and packaging is, you know, immensely important to the company's success. And you know, doing usage based pricing and packaging tiers and figuring out value and correlated value and so on. Can we just talk through, where should it sit? Who should be responsible for it? And if you want to look at pricing and packaging on an ongoing basis, as opposed to a one off, which most companies typically do if they do it at all, then on an ongoing basis? How should that evolve over the evolution of a company going from nothing to some kind of committee that's looking at pricing and packaging on a consistent basis?

Mark Stiving: 29:26

Even in enterprise level companies, they don't do it well, right? It's really hard to go back and visit pricing and packaging for existing products in order for this to happen. Typically, it's whoever has p&l responsibility. So for most startups, the CEO has to be involved. Now what I find is that most of the time when I'm working with cross departmental organisation, so I've got some product management and product marketing, some executives some sales, and we start discussing what does value really mean to the customer? Then every department then starts to wake up. And it's like, oh, yeah, this is what we should be doing, or this is how we should be thinking about it. But without understanding value to the customer, what's really happening is most of our departments are focused on their specific department. But once we start focusing on value to the customer, it really does bring all of these departments together, so they start thinking about it more.

Brandon: 30:22

One of our previous companies, we had toyed with the idea of bringing in a consulting company to deal with our pricing and packaging, they're called profit. Well, they're very successful in the SAS space, the role to be bought by paddle, and their programme for, I think, 12 months was $250,000. us. And we made the decision to do it ourselves. And we started exploring different approaches to make it happen.

Mark Stiving: 30:47

There are many, many consultants out there some that are less than 250k. Although I gotta say profit, well is a fabulous company. Here's the thing, if you want to hire someone to come in and run the numbers and do analytics, and come back and say, here's the price I think you should be using. There are lots of consultants that can go do this. And they're going to be in that ballpark of 100 to 500k, something like that, because they're doing all the data collection, the analytics, the research, they're going to bring that in a different approach, which I like a lot, is have an advisor who just sits there and helps you figure out what you need to do ask you the really hard questions that you don't know the answers to. And it's time for you to go find the answers.

Brandon: 31:29

A lot of companies when they do usage based pricing, they talk about overages. And this question of phrasing of overages seems like a very punitive way of talking about usage of your product, we

Mark Stiving: 31:42

always have to be thinking about what's going on in our buyers mind. And so what's going on in a buyers mind is when we start talking about things like overages, then they're afraid that they're going to get stuck with a huge bill that they have no control over, they're afraid that you're going to try to take advantage of them. And what we need to be able to do is give them the right programme, so that they're comfortable with what it is they're doing and know that we're not trying to cheat them. What my recommendation typically to clients is something to the effect of whatever plan they've set. So let's say I'm going to have a million uses whatever uses. And I know that today you're using 800,000 uses. But if you go over a million, it's not a big deal. Don't worry about it at the next contract cycle. We'll talk about it. And so if the next contract cycle, we go in and say you use 1.2 million last year, you're in the million plan, it's time to move you to the 1.5 million plan. And so that feels really fair to me. And it feels really nurturing to my clients from that respect. Now you might put a an overage on that says, hey, look, you know, the first 10% overage, we just ignore it. And I kind of like that as well, just to tell the clients we're going to do that.

Bethany Ayers: 32:57

If our listeners could only remember one thing from today, what's the one takeaway?

Mark Stiving: 33:03

Buyers trade money for value? Do they understand the value of your product? Perfect.

Brandon: 33:10

So thank you, Mark for joining us on the operations room. If you like what you hear, please subscribe or leave us a comment and we will see you next week.