Patrick Campbell on How to Fix Your Pricing
Note: This was my first time doing an interview in person and my standard mic didn’t do a great job, so the audio isn’t the best, but I promise, the content is more than worth it.
Hey everybody, welcome back to episode number three of the early stage founder show, the podcast for B2B SaaS founders looking to accelerate their growth. I’m your host, Andy Baldacci, and every Tuesday I’ll bring you a brand new interview with startup founders, venture capitalist, consultants, anybody who has been there before and can provide actionable lessons to help your startup get to the next level.
Today I’m talking with Patrick Campbell, founder of SaaSFest and Price Intelligently where he and his team help SaaS leaders align the right product to the right customer for the right price in order to boost revenue and learn more about their customer. They also recently released ProfitWell, a free tool that provides financial metrics for your subscription business.
I first came across Patrick at MicroConf where he gave an amazing talk on pricing, and as soon as I launched this podcast I knew I needed to get him on the show.
Patrick has seen inside hundreds of SaaS companies, I’d actually be surprised if there is anybody out there that has seen inside more, and today he shares the lessons he’s learned from that access and how you can apply those lessons to your own company.
We obviously talk a lot about pricing, but Patrick’s insights don’t stop there and he drops some real knowledge bombs that fly in the face of the common startup advice. You don’t want to miss this.
Oh, and if you use per seat pricing, Patrick also shares why you should probably reconsider
So without further adieu, here’s Patrick!
Andy Baldacci: [0:01] Alright. Patrick, thanks for joining me today.
Patrick Campbell: [0:02] Yeah. Absolutely. It's good to be here.
Andy: [0:03] Before we get into it, I want to give people a 30 second, 60 second, a short background of who you are, what you do, and how all that came about.
Patrick: [0:14] I am from Wisconsin originally. My background is in econometrics and math. I worked in the intelligence community and then worked at Google doing what's called value modeling, taking a bunch of input data and then modeling it for some sort of output target.
[0:31] I used some of those skills working for a start up here in Boston called Gemvara. That's a customized jewelry company like Blue Nile a company for gemstones. ACV was four figures, but from there I worked on pricing a little bit and discovered some really interesting trends with, one, that changes in pricing neither have huge increases in revenue and huge craters in revenue, or changes in prices.
[1:02] Then also, that there's truly interesting phenomenon where these pieces that were really standard looking returned. They had a higher return than pieces that were objectively then really ugly. The reason that the really ugly pieces objectively are really ugly pieces are subjective, but they would be not returns, because people put a lot of thought into, "Oh, this is her birthstone. This is our kid's color of her eyes," all that kind of stuff.
[1:31] When they put a lot of emotion into it, it might not have been the most beautiful piece of jewelry. It was something that had a lot of emotion and sentiment to. [inaudible 01:37] the pay was a lot higher, and so…
Andy: [1:40] I see. Even though it was more kind of intentional, like they…
Patrick: [1:43] Yeah. In putting all that together, I think that’s a much nicer story in hindsight. I don’t know if I actually knew all this going forward, [laughs] but that's kind of where our Price Intelligently started. Price Intelligently, we're a team of about 20 here. We have two products. One helps SaaS companies with their pricing.
[2:01] We collect data using our algorithms that basically measure the price of elasticity, as well as preference. We do all the work for you, and then on the other side, we have a product called ProfitWell, which is free financial measures for subscription businesses.
Andy: [2:16] You've seen kind of inside more SaaS companies than just about anybody. Have you seen any kind of common trends about the way they approach, not necessarily pricing, but just their business in general? What is it you learn from that?
Patrick: [2:28] One thing that we talk about, to give you context of why we've seen inside, like pricing is something that is so special to someone's business that everything that they do, from their marketing and sales all the way to their operations, either drives someone to a purchasing decision, or ultimately reinforces, or justifies that price.
[2:50] Because price is essentially that exchange rate on the value that they're creating, what that means is that everyone's involved in the pricing decision, and you need to see all the data to really make sure the pricing goes well. That's why we've gotten this really, for lack of a better word, intimate look on a lot of business, and then through ProfitWell, because it's free, we've scaled it across a lot of different businesses.
[3:14] I think if I had to nail down a couple of trends, there's two big groups of companies that we see, and we have these cutesy names for them, like one is what we call the CAC fiends.
Andy: [3:29] For listeners, it's C-A-C.
Patrick: [3:32] A little bit of an accent really makes it sound a little different. These folks are customer acquisition cost fiends. They aren't necessarily venture funded, but a lot of times they were. They spend a ton of money on acquisition to acquire customers, but when that acquisition comes in, oftentimes they don't really know how to either price those customers, and monetize them properly, or to retain them.
[3:57] Fortunately, a little bit for us, for our business because we help them, it accounts for a lot of companies that we interact with. Most of those companies we see, if they don't really get their houses in order, they do end up folding at some point because it's a machine that requires more capital. If you can't raise capital, or you can't get the capital, and you're basically putting in a dollar to get 70 cents out, obviously eventually you're going to go out of business.
[4:24] The other group that's really, really the group that we try to emulate are the LTV beasts, or lifetime value beasts. These folks are really, really efficient with their growth, they know their bio personas. That's a huge thing, like they know exactly who they're targeting. Then ultimately, they're the folks that are truly efficient with their acquisition.
[4:45] Those are probably the two big groups. There's a lot of other trends that we can get into, probably some I can't mention on air, but it's one of those things where it's been a really fascinating learning experience.
Andy: [4:57] What do you think it is that…Because if you look at blogs, if you listen to podcasts, if you just exist on the Internet in the start up space, there's so much focus on acquisition above anything else. Why do you think that is?
Patrick: [5:11] I think it's because a few reasons. First, it's relatively easy. Building a business isn't easy, it's simple to understand though. We acquire customers, we bring customers to the door. If you look at the past, like 10 years of software, we didn't need to work really hard to either differentiate ourselves, or acquire a customer.
[5:35] Here’s why, 10 years ago, Crazy Egg was the only marketing analytics product out there. Now you have Hotjar, Crazy Egg, SumoMe, Clicktale. You have all these different products that essentially do, not necessarily the same thing, but from a customer’s perception they do this thing or across, but maybe an uneducated [inaudible 06:00] , because they're doing kind of the same thing.
[6:02] Competition has gotten much harder, but 10 years ago, in our mindset, in all this muscle memory we've learned how to grow a business. It's, "Oh, we can acquire customers easily because we're likely the only person on the block."
[6:14] In addition to that, things were much cheaper. CPCs have risen over 10 years. 15 years ago you could still get some keywords for pennies on the dollar.
Patrick: [6:28] It was like a goldmine. It was one of those things where not everyone had an eBook. Everyone’s got an eBook, so everyone’s got something where it’s [inaudible 06:37] .
[6:38] They're not all created equal. That's why really, really good content companies or really, really good paid acquisition companies still survive. It's one of those things where we can't get things as cheaply as we could previously and that's where it's easy.
[6:54] The other thing and this is a little more philosophical is that there's this kind of ephemeral dopamine effect of acquiring customers. We've all heard the joke about, "Oh yeah, we'll acquire them and we'll figure it out later. We'll figure out monetization later."
[7:10] The problem is, is that now all of a sudden switching costs between products is so much easier.
Andy: [7:19] You're not as locked in.
Patrick: [7:20] Exactly. We're seeing a ton of data on the fact that the relative value of different features is actually decreasing over time. What I mean by that is like, if you create a podcast app and I create a podcast app, and they basically do the same thing, that means we're going to drive the market down for those particular features.
[7:41] Because if you don't have something that truly differentiates your podcast app versus my podcast app, then they're kind of…
Andy: [7:48] It's a commodity almost.
Patrick: [7:49] Yeah, it's almost a commodity. It's not quite commoditized because I might have a brand…
Patrick: [7:53] …with you. There might be a lot of different things but that's a lot harder than just acquiring someone because you're the only podcast app and you have only these types of features. All of a sudden it's becoming much, much harder to acquire customers and it's becoming much, much harder to keep customers because there's a lot of other options out there.
[8:09] Unless you're working on rocket science type thing it's really, really moat almost. It's one of those things where you have to find another way. We posit, and especially because we see in the data, retaining your customers and monetizing your customers has a much, much higher impact on your growth than simply battering against the acquisition.
Andy: [8:33] I remember you from your MicroConf speech and I'll link that up. You put the slides up online for that, I think, right?
Patrick: [8:40] Yeah, yeah.
Andy: [8:41] I'll link that up. Michelle knows some people. You showed how one percent growth in retention or one percent growth in monetization does have disproportionately larger results than…
Patrick: [8:53] Two to four X.
Andy: [8:55] Yeah, and people aren't focusing enough on that. Right now especially with monetization, especially with pricing, you guys have the Kingsoft, but right now, most fast companies when they're starting out, they'll Google around.
[9:09] They'll check out the competitors. They'll magically come up with a number, assume it's perfect and then never think about it ever again. That's obviously bad, but what is the process that you work with your clients on to be more sophisticated and more deliberate about it?
Patrick: [9:28] Sure. This is going to sound a little arrogant, but what's interesting about it, and I'm a little biased obviously, because I do this every day, but the process isn't rocket science. The process isn't something that's exceptionally difficult to figure out. It's not something that's horribly time consuming. It takes time though, and it does take work. That's one of those things where people flow to the path of least resistance.
[9:57] Like hey, I and I don't really know what I'm doing here. It looks difficult so I'm going to set up another AdWords campaign because we all have learned how to do that. The process that we use, and a lot of people are looking for some magical formula.
[10:10] They're like, "Oh, tell me. I also need to capture in this, and add them, and do all these different analyses."
Andy: [10:16] Right, and just spit out a number.
Patrick: [10:17] But in actuality, it really comes down to, all you have to do is get a thesis on who your buyer personas are. Whether it's Start up Steve, Mid market Mary, Enterprise Eddie, or Sales Steve, Market Mary, or, you know…
Andy: [10:31] Hipster Henrietta?
Patrick: [10:32] Hipster Henrietta. Yeah I know, we talked about MicroConf. Whatever it is, it's like you start off with that buyer persona. Then you collect data, just in a very targeted fashion, like not just any random data, but you collect targeted data on their preferences and on their willingness to pay. Then you consolidate that down, and basically iterate on that process.
[10:53] What that does is all of a sudden, it not only has implications on your pricing but also has implications on the rest of your business, because now all of a sudden, you can sit there and you can say, "alright. Well, we want to build X." X might be an integration. Well, does that integration, which persona does that help? Do they even care about that integration?
[11:11] All of a sudden, you're now starting to talk from a personas standpoint. What that allows you to do is it really allows you to open up your business in a way that you never really had before, because you're not brute forcing things. You're becoming a little more sophisticated in your targeting of your customers.
Andy: [11:26] That first part of it is something that, with the buyer personas, that you've already lost most of the way that…Like most fast founders, they skip that stuff.
Patrick: [11:36] Yeah. It's brutal to skip it.
Andy: [11:38] Right. Because then you're right you don't know who you're marketing to, you don't know who you're selling to, you don't…Like you said, you're brute forcing it. You're just trying to kind of go out there with the same message to everybody and it doesn't usually work that well.
Patrick: [11:49] No. I mean if you think about it too, we’ve all been in situations where we’ve sold to someone who was willing to pay 10X the next person who you sold to. If you ever run into that situation, [inaudible 12:05] of that situation, it’s like a very, very clear sign that the pricing is screwed up. [laughs]
Andy: [12:11] I see.
Patrick: [12:12] Because you shouldn't charge, for instance, the Disney Channel the same price as you charge a Price Intelligently for screening and listening problems. Yeah, so it's pretty interesting, and I think a lot of companies, they get some early traction because they have some inkling of their customer.
[12:27] We don't know how they heard about their customers. We build a customer base for a reason, and I think that what ends up happening is they're like, "alright, well let's just like, naturally, just build off of feedback," and they need to take a step back and be like, "alright, so here's our target, and we're going to be a little more strategic."
Andy: [12:41] Yeah, and it’s tough, because especially for solo founders or even just small start ups, they’re really in a place of doing a bit of everything. They have an idea of their buyer personas, they roughly know who this [inaudible 12:54] and they assume that's good enough, and don't really codify it, don't really kind of go too deep at that. You need to. You need to kind of go that meter, go a little farther with that.
Patrick: [13:06] Yeah. I don't think you need to go super complicated. Some buyer personas breakdowns that we've done obviously because we've been paid to, and also we've dug really deep on it. I've dug really, really deep and very, very quantified.
[13:20] I think if you're a solo founder or you're a small business, you're just starting out whatever it is, like, really, really just open a spreadsheet and put some of those hypotheses that you already have just down on paper.
[13:31] Because, again, you likely know something about your buyer. You're getting a lot of different data, and it's just really important to put it in one place. One, that helps centralize your thinking, and two, as you continue to either add a team member, or the market changes, you have something that kind of is a guided line.
[13:49] Don't even need to pullout any data. Just get in what…
Patrick: [13:53] That's the first step, you know. Then eventually, you'll be like, "Oh, I have a really specific question about something. Maybe we should pullout some data," and then you can start to build over time.
Andy: [14:01] Right. I like that you called it put down your hypotheses, because that's one of the big differences, is that once you have it down on paper, once you have it in a central place, then you're able to test drive hypotheses and see if your assumptions were right.
Patrick: [14:17] Exactly.
Andy: [14:18] Because when everything's living in your head, and you haven't thought too much about it, it's hard to really iterate. I know you talked about surveying, but like, do you literally just ask the customers, "How much did you pay for this?" What goes into that process?
Patrick: [14:31] We always recommend start small. Let's assume for a second that you've put down your three different customer personas in a spreadsheet and you have some rows that you've put some hypotheses down. Start up Steve is willing to pay about 50 bucks a month, because that's what I've gotten with those. Mid market Mary is 200 bucks and Enterprise Eddie's 500 bucks.
[14:53] You've written that down, it's in stone, you've basically in pencil mostly and that's raw data. You've basically set that up, so like that's your starting point. Now, you're iterating, you're iterating. All of a sudden you get to a point where you're like we're making a big pricing change. We're making a big launch of a couple of new features, let's say, so we want to like seriously take a quick look at pricing.
[15:17] Now let's say you just want to look at your price. You know, you don't want to care about adding feature differentiators.
Andy: [15:21] Yeah, OK.
Patrick: [15:23] The first step is really to, one, find a good group of customers, your target customers. Maybe you have 100 customers or users, and then 1,000 people in your lead list, let's say, for prospects that haven't purchased.
[15:36] Even then, if you don't have that many, there's a lot of sources, like what's called Ask Your Target Market, aytm.com. If you want to get super serious, you have more enterprise customers, you might go to uSamp or some of the other big market panel providers, but there's these companies where you can pay to get answers from any type of consumer, whether it's consumer side or B2C or B2B.
[16:00] But if I was looking at pricing, and I had that inkling about pricing, the questions that I would ask aren't how much are you willing to pay for this? The reason I wouldn't ask that is because cognitively, psychologists and economists have found that we don't think about value as a point.
[16:16] This microphone, like I don't inherently know that that microphone is like $200. I actually don't know how much it's worth, so it's perfect. But it's one of those things where I can know that that's probably not worth as much as your computer, and that's probably worth more than the pen that you have in front of you.
[16:34] We can take advantage of that, and if I'm doing a survey that's more scalable, what I would do is I would ask at what point is this microphone way too expensive, that you'd never consider purchasing it? For me, it'd be 500 bucks. I would never purchase that, and assuming I'm running a podcast and need a microphone, I'd never purchase that for $500.
[16:54] Then I would ask at what point is this getting expensive, but you'd still consider purchasing it? For me, it'd be probably about 250 bucks. At what point is it a really, really good deal? I'd buy it right away? I'd say about 100 bucks. And then, at what point is it too cheap that I'd question the quality of it? Probably around 75.
[17:12] All of a sudden, if I build that across let's say even 30 people who respond, all of a sudden, I have this value of data of, you know, a 120 data points that I didn't have that can at least give me a range. I'd want to do some basic calculations. I think, you know, price elasticity curve.
Andy: [17:30] You would send this kind of survey to, it could be to the market research company if you went to any of those websites. How much time did you spend on a survey to explain your value propositions so that they know what they're actually buying?
Patrick: [17:44] Yeah, exactly. If you're doing it with someone who has never heard of your product, typically we see people use video. They use a couple of slides on like, "Hey, this is what the product is. This is what it does."
[17:55] Then what's kind of important there, which is this huge side note here, is you actually do check for understanding questions. I would ask those gotcha SAT type questions, where it's like, "Which of these is not a feature of this microphone?" Or, "This microphone does X?" Yes or no, true or false. That helps you qualify these clips.
[18:16] Some people are like, well, they can't touch and feel the microphone. I don't touch and feel a microphone on Amazon before I buy it.
Andy: [18:22] True.
Patrick: [18:22] I don't touch and feel software.
Andy: [18:24] Right.
Patrick: [18:25] I look at the value prop that's presented to me in a certain way, and then I make a purchasing decision, or at the very least I can make a value decision on how much this is worth.
[18:35] If you're talking to your current customers or prospects, they already have some confidence about your product. That's why you want multiple layers of data, because it's really interesting to see what your prospects think it's worth versus who your current customers think it's worth.
Andy: [18:49] In my head, I'm thinking pushback, they were thinking he just wants to charge more. Why is there incentive to be truthful in answering?
Patrick: [18:57] Totally.
Andy: [18:57] Do you get pushback from that? How do they structure in a way to avoid any of those issues?
Patrick: [19:03] Yes. It's a good question. The first thing I'll say is, people know things cost money.
Andy: [19:05] Right.
Patrick: [19:06] Yes, we all want cheaper things, but we also realize that, for me to get some sort of value, I need to purchase something. Then again, we've all met customers who it does not matter how much value that we're giving them, they want cheaper, cheaper, cheaper. To answer your question a little more directly, though, the ranged nature of the questions accounts for a lot of that bias.
[19:30] Keep in mind, we’re not trying to get a picture perfect supply demand curve across the [inaudible 19:37] , because software supply is infinite. I'm trying to find where that customer's range is. What you'll do is, obviously you're going to have some jokers who are going to put zero, zero, zero, zero, or one, one, one, to all of those questions. You kick that data out, because you have that data and you can see who turned what in.
[19:54] That's probably a big thing, too, as a side note, is you want to make sure to see who answers what, so that you can tie that back to usage and demographic data, so that all of a sudden you can find, people who find, people who signed up in January are willing to pay more than people who signed up in March, or something like that.
[20:08] Long story short, it accounts for it not only in the questions, but you also aren't looking for something that's necessarily literally a bulls eye, you're just looking to find the center of the board, if that makes sense.
Andy: [20:19] Again with the current customers, not to get super, super into the weeds, but when you reach out to them, are you saying, "Hey, we're doing some research, trying to get a better understanding of how everyone is using it?" How do you pitch it or position it?
Patrick: [20:35] Totally. You typically find that if you send a pricing survey, people are more than happy to take it.
Andy: [20:43] Really?
Patrick: [20:44] It's related to pricing, because they do get a little spooked. They're like, "Wait a minute, what's happening in pricing? I need to take the survey." The second piece is you want to make sure you have a really good relationship with your customers.
[20:55] The way to do that is to copy what you see that’s [inaudible 20:59] well is something along the lines of, 'Hey, listen Mike, value is a two way street. We want to make sure we're providing you the value that we think we are, and we want to make sure that as we continue to improve the product we're tracking that value appropriately. Here, take 30 seconds, please answer these questions. We really appreciate it," something along that tone.
[21:19] People more than likely…We've sent, gosh, the number is up to like 15 million at this point, and we've quite literally never gotten a bad response.
Andy: [21:31] Really?
Patrick: [21:32] It's never been, "I can't believe you're asking me this," blah, blah, blah.
Andy: [21:35] What do you get?
Patrick: [21:37] You typically get, you get some people who are like, "Oh, you're sending me a survey, and you haven't answered my support ticket." We get stuff like that, but more often than not we get, "Hey, the survey, I took it, here's some extra context. Here's why I think it's not quite worth this. Here's why it's worth more than this." You get a lot of specific detail.
Andy: [21:53] It's super valuable feedback in general too.
Patrick: [21:55] Exactly, exactly. It's one of those things where I think a lot of people are just scared to ask people about pricing, because it feels a little, it's like religion, politics and sex. It's a little awkward, you don't know the person and you don't want to say something. I think at the end of the day it's one of those things where oftentimes people are more than happy to answer the questions, but you have to reinforce the value.
[22:20] You have to understand that if it comes back that you want it to be 100 bucks a month and it's actually 50 bucks a month, there's something wrong there. If it comes back that it's 300 bucks a month and you want 100 bucks a month, you're still going to have to do some work to close that gap.
Andy: [22:33] Right, I see.
Patrick: [22:34] It's one of those things where we've seen plenty on both sides. You handle each side a little bit differently.
Andy: [22:40] It was funny, so as a side story, one of the big, famous poker players is Daniel Negreanu. He was an old school live poker pro, back in casinos, not playing much online. When online started blowing up he didn't own a computer.
Patrick: [22:56] Wow.
Andy: [22:56] When he went to buy it, he brought cash with him. This was in mid 2000s, so not that long ago. He thought it would be $30,000 for the monitor, because he saw how much money people were making playing online poker, and this was a tool that let them do that.
Patrick: [23:13] That's a fascinating story.
Andy: [23:14] He had no clue, and the value to him, he was literally, cash in pocket, prepared to pay that.
Patrick: [23:20] That's because you know I'm going to make $100,000, so yeah.
Andy: [23:23] Right. That's one of the commoditized markets where they don't tend to capture the value that they could. Once you get this feedback, these ranges of data, is it used for directional feedback? We have more work to do to close the value gap to be where we want to be. Or do you start making changes based on that? What's the step after that?
Patrick: [23:44] It depends on the data, but if we're doing a pricing study, all things being equal in terms of our features and things, typically you can start to make direct changes right there, meaning, alright, we're too high, we're too low, or we're wherever. They, we need to move the upper tier higher, or move the lower tier lower, that kind of thing.
[24:05] Typically this will give you, for lack of a better phrase, a finger in the wind. Now you know this is where we are in terms of pricing. I think before I would make changes overall to your pricing, I would then ask a second couple of questions, and these are around relative preference. What I mean by that is, we know price now. I know the general price, where these different personas lie, but I also want to know what features are important to them.
[24:33] I also want to know, what should my value metric be? Value metric is what you pay for. It could be per user, per visits, per poker hand, a number of different things. I think that this is where a lot of companies get things wrong, because they don't have the right value metric, so they end up charging all of these different companies that are all different sizes the same price.
[24:55] I would ask them questions around features and value metric first. The way those questions are arranged…As a huge side jump, I think the big note is that that we as a collective tech ecosystem tend to be really bad at surveys. Typically, the pricing surveys, maybe not, but I've definitely seen pricing surveys where people are like, "How much are you willing to pay?"
Andy: [25:18] Right, right.
Patrick: [25:19] Single point, with no range data. I don't know how they're actually using that data. Or we've taken surveys that are 45 questions long with no incentive.
Andy: [25:30] They have a progress bar where it's like page 1 of 100.
Patrick: [25:34] As a side note, typically your survey should absolutely be less than four minutes. Unless you're paying for it through a market company then it can be 15 or 20 minutes and you can get really good responses, because of the people you're dealing with.
[25:46] The second thing around surveys is if you're going after your current customers typically I recommend going even lower, like 30 seconds. That means at most you're asking four or five questions. The reason that's so important is because the longer the data and the longer your survey is, the worst response rate and the worst quality of data you get.
[26:08] All that said, if it's a 10 minute survey and you're not incenting anyone, out of 100 people you might get 1 person to actually complete it. Out of 500 people you might get 5 people who complete it, 3 of them just kind of gave up.
Andy: [26:18] Right, and just started clicking through at the end.
Patrick: [26:21] Overall, I think it's one of those things where surveys are a really powerful tool that we just don't use well. I think it's because a lot of us don't come from a consumer background where surveys are all range used all the time.
[26:35] Back to the features, what we recommend doing is, I wouldn't ask you of this microphone the condenser quality, the color, and I wouldn't ask you, "Please rate each of those on a scale of 1 to 10." What ends up happening a lot of times is all of those ends up being eights or nines in the aggregate, unless the standard's an issue, unless the standard's really bad.
[26:58] What's fascinating about it is, especially if you're talking to sales to consumers in the marketplace, all of the sudden I don't really know which one of these is the most important, so I don't know which one to advertise chiefly.
[27:11] What I would do is then roll it into one question, "Out of the sound, the color and the stand, what's the most important piece there, and what's the least important piece?" For me it would be that the sound quality is the most important, and then the color is the least important. Those would be my votes.
[27:28] When you’ve collected that data on a technical side or a software side or a technical side, all of the sudden you start to figure out enterprise IT cares about these three things, [inaudible 27:38] doesn't care about these. You start to figure out not only how to target these folks, but also what to build, also what you could differentiate.
[27:44] If you do that on a value metric, all of the sudden you might find out they're willing to pay about 500 bucks. The value metric they really most value is recordings, number of recordings. I would never have a microphone be priced based on number of recordings.
Andy: [laughs] [27:58] Right.
Patrick: [28:00] For a software product it's definitely a unit of…
Andy: [28:01] Right, because you don't just ask, obviously, which would you prefer to pay, number of seats, flat. It's a do, or do you?
Patrick: [28:09] You can actually ask most preferred, least preferred on pricing based on user, pricing based on whatever. You can also ask, when you think about this product, where do you see the most value, number of users, number of X, number of Y? There's two different ways. It depends on exactly the question you're trying to answer and the hypothesis you're trying to validate.
[28:30] It's one of those things where, again, these are just tools you can maybe use to answer these questions. Oftentimes if you're just starting off where you're a smaller company, or even a larger company, because a lot of large companies aren't doing this either, all of the sudden what you can end up doing is, we have our personas now let's validate price. Next quarter let's validate feature preference. Next quarter we'll do this.
[28:52] Based on all that data you can start making changes not only to your pricing, but also to your overall marketing strategy, sales strategy, retention strategy, etc.
Andy: [29:01] Not to try to simplify it too much, because this is a process that all of the steps of value should be done, but at the same time have you found any rules of thumb for thinking about value metrics like, when does it make more sense generally to use per seat pricing, versus usage, versus the other alternatives?
Patrick: [29:23] Exactly. More often than not, per seat pricing is the wrong choice. [laughs] That's a blanket statement, but it's funny because in the history of SaaS the reason per user pricing is so prevalent is because back in the day before software was on the cloud and it was perpetual, you would sell licenses.
[29:45] That was the easiest way to understand how much value you get, because it was going to be in your workstation, it was going to be in my workstation, and so we would use two licenses. All of a sudden when the cloud came, people were like, "Let's just do that." The problem is, one, we have much better ways of tracking value for charging them value. The other problem is you can share logins, whereas we couldn't share a workstation if we were both working 9:00 to 5:00.
[30:22] You can share a login really easy. The wrong thing to do in that case is to IP block. It's one of those things where it's like, wait a minute, if you're building technology to IP block people from sharing logins, maybe you're not priced on the right value metric.
Andy: [30:34] I see.
Patrick: [30:35] A really good litmus on, are you pricing correctly, or should you be using per user pricing, is, can people share logins and get the same experience. If you're a sales product or a help desk or something that's very individualized to the person who logs in, probably the answer is per seat pricing makes a lot of sense.
[30:53] Leads are different. Tickets are different. If you're an analytics product, per seat pricing is the worst. You can still use it, maybe to prevent someone from backsliding down, someone who has really, really high value data, not a lot of it but should he be using more than a couple seats?
[31:12] You see this with HubSpot, for instance, where their lowest level planning…Let's imagine a really high ACD product like, a lead was worth $1,000. That's why I only need 1,000 leads, which is their value metric.
[31:27] They make folks want more users basically go to their higher level planning…
Patrick: [31:33] …and users. You can use it as like a backstop. But ultimately it's probably not the right one. Other rules of thumb, I think that a lot of times, the whole nines versus zeros and fives debate. We have never seen conclusive evidence in the software space.
Andy: [31:51] Really?
Patrick: [31:53] Yeah, ever.
Andy: [31:54] And this is for [inaudible 31:54] numbers.
Patrick: [31:56] The old school thought in retail was if you end your prices at nines, it feels like a discount. They're feeling like they're getting a fraction off, therefore you're going to get higher velocity. But if you want to be a luxury product, you'd be fives or zeros because that feels like a whole.
[32:13] FreshKon or Tiffany’s the jewelry store, all the prices are in zeros or fives. There’s no nines. You go to a discount shop, everything’s nines or eights, sevens in that case. Go to [inaudible 32:23] , you'll see the before prices are typically zeros and fives, the after prices are all sevens and nines. It's kind of weird.
[32:30] I don't know if they did that intentionally because they're obviously discounting. I'm sure they did it intentionally but in software we've never seen any conclusive proof that one works better. We've also never seen conclusive proof that putting your highest tier on the left is better than putting your highest tier on the right.
[32:46] Kind of like growth hacks or acquisition tactics. Typically if there's a blog post written about it, it's one of those things where it's like, you still have to test it. There might be proof in the sense of that works for that product, but it doesn't work for that product or it's not something like everyone should end up in this but everyone should put their pricing like that.
Andy: [33:06] With ProfitWell, it's basically a freemium model, right?
Patrick: [33:10] Yeah.
Andy: [33:11] There's so many blogs on why freemium is horrible.
Patrick: [33:13] Yeah. We have written some.
Andy: [33:16] So what made freemium work in this situation?
Patrick: [33:20] It was doing our homework. This is a lesson on, you can launch something and really screw it up if you haven't done your homework. What we found was we I mean we do this on the other side of business with our other software, and I don't really watch this there's value here. We've charted a lot of value for it.
[33:39] What we notice, and this is through public information and then some information we have privately. It's one of those things where analytics products have really bad retention. If you look at our retention curve from a Mixpanel, they've been public about this but it's one of those things where you see a retention curve where there's a lot of one, two, three months drop off. Then they even out in a really low number. That's because for that kind of product, you really have to commit.
Andy: [34:09] Right. You don't get as much value just by looking. It has to be…
Patrick: [34:14] If you sign up for it and you're like, "Whatever, it's a little too expensive to just use." With that fact, we're like, "Well, if the analytics products is really hard…" And also, analytics product is pretty difficult in terms of convincing someone to use it every day.
[34:30] It's not the most active usage product because it's not like your email, where you have to do something every day. Sometimes you just don't look at your numbers because all of a sudden you're like, "Oh crap. There's just so much stuff going on."
[34:42] Then, “Oh, it’s the board meeting. I have to get my numbers right.” It’s one of those things where all of a sudden, you have the situation where analytics products are really hard. [laughs]
[34:53] It's one of those things, maybe that's a reason to give it away for a lot of money. What we did is, we ended up lowering our price. We did a simple price test and we said, "Listen, here's how good the product is." We had some alpha and betas at this point. We said, "Listen, here's what it is," and we got some fast other people in the market, non parametric curtain over…
Patrick: [35:14] …those folks. Then we said, "alright, for this type of a product, what's your willingness to pay?" What we found was that the ARPU was not incredibly high.
Andy: [35:23] Which is average revenue per user.
Patrick: [35:24] Yeah, sorry, yeah. It wasn't incredibly high but the more important data point there was the difference in ARPU between someone like HubSpot and a one person band starting off their own software company was not that different.
[35:38] What I mean is, to give you an example, a $5,000 lifestyle business app, they're willing to pay about 50 bucks. HubSpot might have been 250 bucks. The delta, 5X but not like 10X.
Andy: [35:53] Right, not an order of magnitude.
Patrick: [35:55] All of a sudden, we're sitting there and we're like, "alright. Well we have this back end data for the market. Features are going to zero in terms of relative value." It's really, really hard to retain and prolong those products. Insurance is really, really high.
[36:10] And the ARPU doesn't really differentiate them. But we know there's a market here and we know that there's a better way to monetize this post. That went into we should give this away for free and we should give away a better product for free.
[36:28] One of the things, we're behind them a couple of places but we really, really focus on accuracy. We're the number one in terms of accuracy in the live build and things like that. There's customers who're our users, who go to some of the other product and they turn them away because they can't handle the data. We're all able to handle that data.
[36:47] What that has done is, all of a sudden we've built this really nice server base of people. As we close, maybe, feature gaps and as we close different design gaps, now all of a sudden it's like we're at parity and distribute.
[37:03] You're more of a beacon in the SaaS community and then we can sell folks things that are more closely tied to revenue. During these tests, we also found that we lost retain, which helps with delinquent insurance and helps with overall retention.
[37:21] It not only helps you get your delinquent folks but we also help you upgrade folks and we get insights, things like that. All of a sudden, what we found was that $50 customer was actually willing to pay 3 X if you prove to them we just recovered this much money for you.
[37:38] All of a sudden it's like, our ARPU can be much higher for these companies that weren't willing to pay much for analytics. We can acquire them and now we have, I think it's a fifth of our base is on retainer.
Andy: [37:51] It almost acts like a lead magnet. It gets people in and it also, literally, gives them the proof. You're going to look at the metrics…
Patrick: [38:02] You have a problem here. Pay for this solution.
Andy: [38:04] Right. Then you'll be able to demonstrably show that it improved.
Patrick: [38:08] Totally. It's not all proven yet. There's data that's supporting that this is the right decision, but as far as things where, this also comes with a whole slew of problems.
Andy: [laughs] [38:19]
Patrick: [38:20] The impossible product substantiate. We have multiple value propositions. It's not crystal clear and clean. There's also a lot around juggling metrics and being 100 percent accurate is really hard. People think it's like, "Just do it. You just show me a number. Can't you just give me that number?"
[38:42] It's like, "No, there's 800 different edge cases that we just…"
Andy: [38:47] I see.
Patrick: [38:48] You have to know that when the time zone shift happens, through that revenue where it goes. Oh wait, they upgraded on the 16th, they upgraded on the 19th, we have to account for all those certain things.
[38:59] That's why I empathize with our competitors. Simply because it's like, "I know this isn't easy."
Andy: [39:03] It's not an easy problem and there is a value in it. That makes sense. What does the future look like for ProfitWell, for Price Intelligently? What does the next month look like for you guys? Then what is the long term beyond that?
Patrick: [39:17] The month. It's funny, the long term's an easier answer, the month though.
Patrick: [39:22] The month is where we’re finally going to redesign [inaudible 39:26] . This is something where we basically spent the first year just heads down on accuracy. Any time a bug came in and still any time a bug comes in in accuracy, we drop everything and fix it.
[39:37] We're now at a point where the drop anythings are happening less and less. Cool. What's happening with us is we are diving too deep into…We built the product in a very nimble fashion, particularly when it comes to data.
[39:50] Now it's all about the user experience and digitalizations. We're launching this redesign that's pretty dramatic from where it is today and that should really push things forward and usability as well as depth for our users.
[40:05] Then we're continuing to march on retain. We have some really, really cool stuff launching. It becomes a game of 99 get to market share. You have a product that has really good MPS, really good active usage and then one that's working on the retain side. Now it's like watching different integrations, things like that.
[40:27] The next month is the reason I'm answering your question directly. Then, we have some other fun stuff. We just hired our VP product marketing, which is pretty exciting. It's like an adult hire.
Patrick: [40:38] We're an adult team now. Then the future, we're going to continue. This might get a little too out there but there's a fundamental problem with business intelligence we're trying to solve. If you look at Adomo or Tableau even, or you look at Looker, some of these other products. The onus is on you to calculate the metrics.
[41:03] The onus is on you to put all your data in one place and do your calculations. What we've found is it's really hard. Also, it needs to be accurate. ARPU analytics products can be a little bit off because you just want the trend.
[41:15] [inaudible 42:16] if I tell you I have $1,000 in the bank account, $1,000 better be there. For us, what we're trying to do is we're trying to fundamentally change the way businesses and intelligence is run in the stats. That means not only connecting the financial data, which is probably the hardest data to get. That's why we started with it but then connecting that data not only to the up funnel but also the post funnel.
[41:40] We start to connect all of that data, all of a sudden, I can, with turnkey precision, give you all of your attribution numbers. I can tell you who your most active users are and I can give you that entire look at your business in a much more focused way rather than being like, "Hey Andy, here's 60 graphs, good luck."
Andy: [laughs] [41:59]
Patrick: [42:00] I can actually give you…
Andy: [42:03] Right.
Patrick: [42:04] That’s what we’re attacking. That’s the big, scary, big ass hole we’re going after. I think the focus is basically on solving problems with business intelligence. It’s not very [inaudible 42:28] . Adomo tries to go after every single type of customer out there.
[42:21] It's not turnkey. It's, "Hey, we give you the tools to use your data."
Andy: [42:25] Then get to work.
Patrick: [42:27] …stuff like that. That was actually the inception of ProfitWell. We were sitting with a company that was about to go public. They calculated MRR incorrectly, it's not a cap venture but it definitely helps.
Andy: [42:39] Yeah, right.
Patrick: [42:42] This was a CFO taking two of their companies public. It wasn't due to a fault of our own. It's just, this was not something that a lot of people know.
Andy: [42:50] It's also, from your angle, they don't necessarily need to. With the right software with the right tools, they shouldn't have to know how to do that.
Patrick: [43:01] Yeah and the other thing is, you think about vi software, software in general, why wouldn’t [inaudible 43:10] making it work? It should just do it. That's so easy to say. It's so hard to do. Why would I need to give you 60 graphs? Or why do I need to make you calculate it?
[43:22] We look at it every day. We should calculate it for you. There's nuance there, obviously. Things internal look differently and stuff like that but we want to get to a place where that keyboard exists.
Andy: [43:34] The last couple of questions I want to ask. They're just quick ones. They're basically, what do you think right now you spend too much time on in your business?
Patrick: [43:42] I think right now I'm spending a little…There's things that I should evolve into not doing, if that makes sense, because we need to build a team, stuff like that, around it. I think right now, I do all of our support on ProfitWell.
[43:57] I think it's one of those things where I'm spending…And I'm learning a lot from there but it's also a huge time scale. We need to automate some of it because I think it's super personalized. You sign up, you're actually getting an email from me.
[44:13] PR responses and stuff like that but personal.
Andy: [44:14] Yeah, exactly, right.
Patrick: [44:15] It's not like it's an automated…
Andy: [44:18] I see.
Patrick: [44:18] We did that for a couple of reasons that we can get into another time but basically, it's one of those things where I was spending a ton of time there and then, I think in addition to that, there's a ton of time being spent on my side in the weeds in a couple of different places.
[44:35] I get sucked in just doing my role as a CEO where I get sent in a couple of different places and I need to make sure that I hire someone, a general, let's say. Someone to take care of these things, so I should just not worry about it.
[44:48] I want to make sure that…It’s hard, especially if you get bootstrapper’s mentality. It’s hard not to jump in and just let someone do your job. I feel like I just kind of punted on that. [laughs]
Andy: [45:00] We'll go from that though into, where do you think you need to spend more time? Where could your resources be…are needed? Where are the biggest leverage points for you?
Patrick: [45:10] I think our marketing. Fun fact, I told you this, we don't have anyone on marketing. There's no full time person. I think we do OK.
Patrick: [45:19] …person. That's because we're really educational about it. I really believe and the team really, really believes in studying the education free. Getting all the education out there. You shouldn't have to pay for a pricing course.
[45:36] If you have a really good one, top of the line, you shouldn't have to pay for some of this stuff. I want to spend a lot more time doing a lot more around educational content that people can use. Just so that they can continue to integrate with businesses.
Andy: [45:52] I think you gave us a ton to comprehend and distill down so I was glad about that. This is one of the interviews I've been really looking forward to. Patrick, I just want to say, thanks for coming in. I really appreciate it.
Patrick: [46:05] Absolutely. This was fun.