Lessons from the SaaS Metrics of 1500 Companies – Patrick Campbell

0:00[Music] [Music] hey everyone so yeah we're gonna talk a little bit about pricing I'm not a one-trick pony though so I have some other things we can talk about as well to kind of get stuff started though if you're someone like me who likes to have the slides and like to take notes and all that kind of fun stuff you can go to price intelligent accomplish micro comp there's a landing page if you don't want to give me your email address just put like face at micro calm calm or something like that we can spam Rob essentially but the I won't actually spam you Rob wherever you

0:41want he's not even listening anyways so the big thing that we're gonna talk about today is kind of alluded to by the title is really this concept of building a software business and I would argue a business in general is something that's deceptively simple there's so many tactics that have been talked about already there's so many tactics that I heard last night but we like to make it super super complicated and we're gonna go through a lot of different information here that we've seen inside a lot of these different companies that we've worked with and then ultimately provide you guys with some good tactics and some good frameworks to not only get

1:13your customer development right but ultimately to get your pricing correct or at least on the right track and to kind of kick things off a little bit of the origin story here and these pictures were chosen very specifically because let's be honest realistic anyways but self-deprecating love it anyways so my background is in econometrics and math I started my career working for the intelligence community and I mentioned that just to alienate you guys even further but after working in DC and at Fort Meade I went worked at Google where I did something very very similar interesting enough and my entire career I do it was called value modeling which

1:51is the whole concept of taking a bunch of data and really getting towards an output so at the NSA hunting terrorists at Google hunting more money and what's really really fascinating about that is a lot of the data that we work with inside SAS companies actually has a lot of applicable information for finding that particular value for your customer or just the value of your company in general it's a little bit more specifically about the companies and it's all one company and we actually do have software we're not just consulting but what we did is we started price intelligently about four years ago and we had a couple of algorithms that

2:23we developed that basically got to customer value so we're able to measure things like price elasticity or relative preference of different features and different products and we started scaling that as an actual software product and then we discovered that selling that for 50 bucks a month ironically a really really bad price wasn't really going to get us where we wanted to go because there are a lot of different issues with the product in and of itself we found that a lot of people who came to us we're asking us about pricing we're basically like hey we'll pay you just to do this for us and not

2:53being you know - he gets egomaniacal around oh we want pure touchless sales and product we said okay and that's how we were able to self fund or customer fund our business in general and then about two years ago what we ended up finding is we're actually sitting in a boardroom of a company that was about to go public in the next six to nine months and that's as specific as I can get for them in particular but we were going through their numbers and in really kind of specific order and what we discovered was is they were calculating something is seemingly simple as m RR incorrectly

3:23and that's because it's not a GAAP metric your CFO of a public company or a large company doesn't really need to have their SAS metrics to be accurate but when they were going on their Roadshow essentially they discovered that that was millions of dollars in market cap but they were missing or at least that's what they told us to make us feel better so about two years ago we started working on profit well which is essentially free financial metrics for subscription businesses and that's kind of taken off and we're doing really really well with that particular product across many of these different SAS companies and I'm not telling you this

3:53to essentially sell you on anything but I'm telling you this just to give a perspective that from at least from a financial perspective we've seen inside more subscription businesses than anyone else out there we have about 1800 companies using profile which have a few hundred companies like Atlassian and some of the other logos that you've seen on the previous slide that we've worked with in terms of pricing and what's really fascinating about a lot of these companies is that as we continue to dig through the data and just as we continue to have these different conversations both in a very high touch capacity on price intelligently side but also in a

4:24lower touch capacity in the profit well side we decided to discover these really clear groups start to emerge now the one group we call the LTV beasts or the lifetime value it's a lifetime value for those of you don't know it's basically a measure of your retention and how well you aren't monetizing your pricing over your or yeah that's the measure of it not the LTV to CAC I'll get to that in a second and then the CAC fiends these are essentially the customer acquisition cost fiends these are folks were really really focused on growth through actual costs of acquisition I was fascinating about this at least

4:58qualitatively what we started to discover that these LTV s-- beasts had very very low amounts of funding relative to their size so someone like Atlassian you know they've raised a hundred million dollars they've gone public but they didn't raise a ton of money to get to that particular point they also have 10x sales teams or no sales teams in general these are the folks that are extremely efficient in actual acquisition and then finally they know their buyer personas and their unit economics like the back of their hand you can walk into the company you can ask the CEO all the way down to the entry level market or who they're

5:30selling to and they'll give you a half-hour discussion about this is the buyer this is what they look like this is the team size this is what they like this is what their willingness to pay looks like but this contrasted pretty starkly with these [ __ ] fiends who were essentially the polar opposites of these individuals these are folks that have raised more money than God doesn't mean they've raised a hundred million dollars but it does mean they've raised a lot of money relative to their size they've 1x sales teams these are the types of companies that you walk into and it's 30 people and 16 of them are

5:57sales people or even scaling up 700 people at a dev ops product and 400 of them are sales people and they don't know their buyers you walk in you talk to a CEO and they say developers what kind of developers they can't really go in depth and I'm setting this up not to you know kind of wax nostalgic about you know some of the customers that we've seen but these are the two groups that we see typically in the market and as we dig deeper and deeper into some data here where you'll start to find is that these lifetime value beasts these companies that we all want to be it's

6:32actually a very very small portion of the market and those folks are the ones who are crushing it for the lack of a better cliche phrase basically because they're so focused on their unit economics and specifically so focused on their buyer personas so based on all this data do we have a unified theory of SAS growth no I don't have magical powers quite yet but what we're going to do is we're gonna talk through these three big trends and then the second half of the presentation we're going to give you those frameworks to hopefully make sure that you're not falling into some of these pitfalls whether you're just

7:06starting your business or you've been growing for many many years there's a lot of applicable knowledge and here let alone some benchmarks for you that are going to be helpful and the three big things that we're going to talk about today is a lot of times these caffeine's these folks who aren't really talking to their customers they typically focus on the wrong benchmarks they're chasing the wrong problems within their business these folks don't take customer development seriously talked a lot about customer development today already and then finally they focus way too much on acquisition and that's where we'll get into pricing monetization and all that kind of fun stuff and thus translate

7:40this and hopefully this is a very unique group in the sense that we're not in this category but some of us probably are whether we like to admit it and frankly there's some stuff in here that just for the nature of time and priorities we also need to get better at as well so hopefully this will be a really really good way that we all can get better within our businesses so the first point here is really focusing on benchmarks so right now if you search for example Google or kora or what's the average gross churn rate and we've all done this or been forward an article

8:09typically what you end up seeing is a VC or someone who's you know very very interested in you using their product to reduce churn we'll talk about how you know when they sold their company to X company it was X percent and then here's you know three to five different you know basically pinpoints in the market to help you basically understand that your gross churn should be at five percent so this is something that we see really really commonly the other thing that you tend to see our articles written about well we want lower gross churn which is definitely the truth but oftentimes when we look at 5% a lot of

8:41companies what ends up happening is they get to 5% and they assume they just have to stop and in some cases that might be true like that's the very nature of a benchmark if you notice that one particular part of your business is going really really well you might put it on the back burner for another type of business but when we found by looking at about 600 different companies in the way that we broke this data down was based on the LTV to CAC ratio so it's basically a measure for those of you don't know if you efficiency of your business a lot of us

9:07in this room really care about it since we're self-funded and we can't just hack our way to success but it was really really fascinating is that 5% benchmark was really in the weakest subset of companies those companies that an LTV to CAC ratios of two or less now those companies that had between two and five those folks who had hit that threshold and the magic number that a lot of people talk about is three those were actually closer to just below 4% and then those folks absolutely just crushing it once again for the lack of a better cliche gross churn was actually closer to 2% now what's interesting

9:42about this data is if you stopped at 5% you might actually not be chasing the right number right numbers within your business now we'll do a full write-up on this because there's a lot of other variables and we'll pass that along to the community here but it's one of those things where when you're focusing on your business as we talked about we need to focus on the right benchmarks now just to give you another one here what percent of total sales is expansion revenue expansion revenue for those we don't know basically upsells basically getting your business to you know make more money off your existing customers rather than

10:14filling your growth with new customers kind of the nomenclature out there if you ask David's [ __ ] or look at the Pacific Crest survey you're looking at 10 to 15% in reality what we found is that those customers who are doing really really well or those companies that are doing really really well they're almost in the 35% range those companies that are doing the worst those are the ones who around 10 to 15% so it's something to keep in mind as you continue to look at where you have those different levers in your business and just to kind of close this point out here let's look at how fast you should

10:44be going or how much you should be growing year-over-year what's really fascinating is mostly everything out there says as much as possible and that's true right you know we want to grow as fast as possible and as much as possible but if some of the other folks are asking questions in the previous talk we're talking about sometimes you want to slow down a little bit you want to make sure that you're building the right product for the right customer at the right price and so because of that here's an interesting benchmark for you guys across not the LTV to cap ratio this time but really around ARR like how

11:11much money that these folks already were bringing in on a recurring Rev new basis and you'll notice that those folks less than 100k a year they're growing substantially right and the variance is really really high and then as the scales down you know it's naturally intuitive that a five million dollar company isn't going to be doubling or tripling every single year here so the point here understand your metrics a lot of these companies don't understand their unit economics and then they don't understand what to focus on something that's really really really important because as we intuitively know if we don't know what to focus on and

11:45we're not focusing on the right things there's gonna end up being some trouble down the road as you continue to try to scale your business in a really really efficient way next up here is this gets to be a little bit more tactical we don't really do customer development seriously and I'm talking in the aggregate here and I have some data to back up that statement but just as a social experiment here and this might backfire just because of the nature of the group here how many of you have buyer personas written out okay of those keep your hands up sorry of those who have buyer personas how many of those

12:21buyer personas are in a central document that anyone on your team can access keep your hands raised how many of those buyer personas do you have broken down by unit economics so you know your kacct or LTV for your different buyers so we lost everyone um great uh-huh and I have about four more questions to willingness to pay best features worst features usage metrics this is the type of buyer persona you want hub spots been talking about buyer personas for a decade and a half and their summary is you know acute avatar and like a pretty picture basically of you know table stakes Tony and some cute name in reality you want

13:02to know your buyers more intimately well that's a really weird way to describe it sorry but you want to know your buyers better than anyone else in the history of your business and the reason for that is because if you know what's valued you know it's least valued you know willingness to pay you know your cap you know your LTV you know your usage metrics what you're able to do is align your entire business around those particular buyers and as we saw in this room and will hopefully help you get some of that work done we I'll have a little bit of work to do to

13:29get this better and to make you feel a little better though we have some good industry stats these are about 1,400 companies all in the SAS and subscription space most folks over half of them they've thought about them and that's kind of where a lot of you are like you don't you know things about your buyers like it's not like you know nothing it's one of those things where it might not be codified and not but it might be quantified only about three out of ten of them have some sort of central document and then less than one out of ten of them have some sort of quantified

13:56buyer persona and that's kind of like what we were showing in the previous slide but not necessarily exactly we don't really do a lot of custom conversations this is another thought experiment or social experiment we could do here most folks are having less than 10 customer development conversations with through or each month through their customer base not a lot of folks are doing more than that less than one out of ten out of the next three categories and so the excuse here is well you'll do surveys well a lot of us don't do a lot of surveys you know most of us are doing no surveys per month or averaging out

14:28there's a small portion of us who's sending one customer development survey a month and then a very very small part of the population is doing more than those and then the next kind of excuse here is we'll do testing well we're not really testing that much either or running experiments just under five out of ten of us or running no experiments or averaging no experiments per month you know three out of ten of us about one to three and then a very very small portion are doing more and more experiments and as some of the growth folks you know if you talk to Lars or you talk to Brian Balfour or any of

14:57these other folks that are out there and will tell you that experimenting even if you're just focus on acquisition it's probably the one leading indicator of your actual top of the funnel growth so I'm not here to try to make you feel bad or anything because you know price intelligently we're not running as many tests and experiments as we should be either but I am here to tell you that this should be a little bit scary to you because remember deceptively simple we have all the tactics we all have collective knowledge around building a business but a lot of us aren't doing their homework and that's making it

15:28actually extremely hard and to think about it another way in your business everything leads to a page like this or a conversation that asks someone for a credit card or an invoice or to sign a contract and if you don't know who that Buy is then you don't know who to drive to that particular page to that point of conversion and you don't know how you're gonna actually justify that price so it's something that's that important and frankly it's something that we see that separates those LTV beasts from the CAC fiends most staunchly this next date is very very qualitative just to be very very clear cuz I'm gonna put percentages

16:02up and people then freak out and assume its quantitative sometimes but I'm a hundred and two companies that we've studied over the past year who have failed this past couple years I should say who have failed miserably or add huge down rounds if they were venture backed 98% of them had no sort of quantified buyer persona most of them didn't have a central document either and the flip side of this is once again very qualitative about 80 companies that had exits of a tech so the original investment or more or in the upper quartile of growth for kind of what we were seeing across these different

16:37companies they had some sort of quantified buyer persona so is this something that is going to guarantee success absolutely not but it's one of those things that correlates to success and definitely has some lurking variables that actually help you based on what you're seeing within your business and basically helps you a line so to kind of round out why people don't do this there's two main arguments that we typically see one is the Steve Jobs argument that I love and this I get this one all the time when we talk about some of the stuff in the later part of the presentation here oh you know if I ask

17:09someone what they wanted you know they don't know what they want if I ask them or the Henry Ford quote as well you know if I asked them what they wanted they would have said a faster horse and then Steve Jobs you know whoever said you know the customer's always right was in fact a customer what's interesting about that though is that both organizations both Apple and Ford actually do a ton of customer research neither of them do a lot of testing it's a very very big distinction between testing and research but it's one of those things that if you're Steve Jobs you go be Steve Jobs

17:38for the rest of us you should really really be doing your homework and then the other thing which is the testing you know what we noticed is a lot of people aren't doing the testing as the data says but what's also really really interesting about this is a lot of us in our businesses price intelligently and profit well included we don't have enough track thick to truly test everything oftentimes we need to do a lot of research to get to maybe a very clear a B test and then that's something that we can test just before the fact of being a B to B SAS company even if we're driving

18:08you know fifty five thousand a hundred thousand uniques you don't necessarily have as many successes or as many failures to really account for many many different tests that you could be running so as I said if you're Steve Jobs be Steve Jobs for the rest of us you should talk to your customers and we'll talk about a little ways that you can actually do that after this next point here now the final thing to set this up to actually kind of give you some tools that you can be using is this whole concept that were focused way to much on acquisition so many of the tactics that we talked about or I heard

18:39last night and some of the other ones that we've heard today are really really focused on tofu or the top of the funnel what's interesting about that though is that what we found by looking at a lot of the data and we'll show you that in just a second is that while we're focused exclusively on acquisition it's not really the most effective lever for growth so to kind of prove this out a little bit we did a little anthropological study on blog posts so the reason we looked at blog post was mainly because we write about what we know you know if we've nail a tactic we're gonna write about it and

19:08the other thing was is that we write about what gets us traffic so if someone writes about Facebook channels and we all kind of share it and love it we're more incentive to write more about acquisition and what was interesting is by looking at these 10,000 blog posts between seven and eight out of ten of them they're written about acquisition retention was really around two out of ten of them then less than one out of ten were written about monetization and sadly most of those were written by price intelligently I don't know if that's sad or not but it's sad because we want more pricing friends but to kind

19:39of continue this and kind of talk through why this isn't the most effective channel and kind of show what the data looks like we built out a little bit of model with a about 500 companies 500 SAS companies in particular we wanted to answer the question that if we improve each of those main pillars of growth acquisition monetization and retention by the same amount or the same relative amount what would the respective impact be on the bottom line and what was interesting is if you improve your acquisition by 1% so meaning if you improve the leads that you have or you improve the efficiency of your conversion you're

20:13gonna see about a three percent boost in your bottom line now if you improve your retention by one percent how long those folks stick around essentially you're going to improve your bottom line by just under seven percent and as to not contradict myself of course if you prove your monetization by about one percent raise your price figure out your packaging decrease your value metric in the right way you're gonna see about a thirteen percent boost in your bottom line now what's fascinating about this isn't that necessarily the absolute numbers here but it's the relative impact that you're seeing in the differences of these approaches and really the fact that focusing on your

20:52monetization and your retention especially if you're a self-funded company has two to four X the impact of focusing in on your acquisition as heat and in steli we're talking about it's extremely easy to do what we know right ship another feature ship another blog post do another AdWords test but at the end of the day some of these harder things or deceptively simple things that still require some work those are the ones that might actually give us the biggest impact on our business even though we keep kind of putting them to the wayside because we might not necessarily know or think we don't know how to actually approach them look at

21:28this data another way this is a little bit more applicable to some of us since most of us aren't on kind of a venture path those companies that have no pricing function meaning they've just kind of set the prices up originally and they just kind of didn't look at them for a while or maybe look at them every five years 1.68 LTV - CAC so it's pretty abysmal it means if you put a dollar in you're basically getting 68 cents out yearly pricing review these are folks who are changing their pricing up not necessarily raising the price but changing it right over that three threshold of three point two three and

22:00then those folks who have continual price optimization and what I mean by that is basically they're adjusting their packaging their pricing or even their positioning every single you know six months or maybe every quarter at certain stages their LTV tacacs are averaging out to about eleven so what this tells me particularly around you know dealing with a lot of these different companies is that we really really really want customers but we're not really sure what to do when we get them and what's really fascinating about is we focus on growth so much as a sledgehammer a lot of our businesses you know some of the businesses that I've tried before price

22:34intelligently I would focus on something for six months and not realize that there was a huge assumption that I was making and I was just spinning my wheels for six months before I actually fixed it and that might be some of the stories that you guys have any ore today but if you focus on the right metrics you focus on your customers and ultimately you focus in the right levers that you should be pulling that's where you're going to see that growth and ultimately that's going to that's where you're going to see where you can make a nice big successful business so now that I've made us all feel bad including myself

23:04because there's a lot of stuff in here that we need to work on as well how do we fix this so how can we give you some frameworks some tools and even some more benchmarks around how you can actually make this better well the first step here is quantifying your buyer personas everything starts with the buyer persona and you know if we talk afterwards we talk whenever the first thing I'm going to talk about because if you don't have that a lot of times there's a lot of rippling implications around your business because you're not gonna know what to test at the top of the funnel

23:31you're not gonna know how to convert folks and ultimately you're not going to know how to price or retain those folks because you don't know those buyers deeply enough the second is implementing a pricing process and then finally implementing a multi price mindset or using a value metric which some of these folks talked about already today so let's dig in so buyer personas as we talked about look something like this most of the time it's more of a bastardized spreadsheet it can have many many different rows it doesn't have to just these things here but really what you're trying to get to is understanding your buyer from your buyers perspective

24:04and collecting data that allows you to know what they value what their willingness to pay looks like and then the unit economics so you know if you're targeting the right buyer or sometimes you need to get rid of certain buyers as Claire was talking about this morning now what's interesting and just to hit the nail on the head here a little bit more if you don't know who you're driving to that pricing page just very very flatly you're not going to know what to put on that pricing page and pricing because of all the different levers in terms of packaging position and the actual price especially if

24:32you're a multi tiered product you can't test your way out of that whole you have to talk to customers and make some very very serious decisions so let's walk through an example because I think it helps you know top pricing to a lot of folks and I think you know it makes it a little bit concrete this is also my favorite parks I get to talk about my mother and I like to title this my mom's quilting business has better unit economics than your real business so for those of you don't know quilting it's actually an interesting industry but quilting is a hobby where people basically make blankets if you've ever

25:04had a child or you've ever you know had a child in your life it's more than likely they've been given a quilt of some sort a baby quilt and this is my mom and she is the master at quilting it's basically her Hobby she's a trade show marketer by day she quilts at night as her hobby and what's interesting is I basically wanted to help my mom make a business that she could retire essentially and have enough you know income to basically survive and maybe even grow an interesting business and before you laughs the quilting industry is actually a four billion dollar industry the median household income is

25:33about a hundred and twenty six thousand dollars and the average amount of money the average quilter spends per years between five and six thousand dollars so it's better than some sales teams and sales products actually which is interesting so what we did is we came up with this concept we're gonna do a box of the Month Club in the quilting world basically whether its supplies patterns whatever it was gonna be we're gonna put that in a box and sell it and we figured if it's a four billion dollar industry we can figure out how to sell a few boxes at 100 bucks per month and that

26:01was just you know assumptions and that's kind of where you guys are at right like you have a lot of assumptions about your buyers so the first step was I sat down with my mom over lunch and I basically said all right mom what are the different groups of quilters out there this is kind of where you guys are at right you have different groups of your customers you kind of know that bather it's based on size or role or whatever it is and we came up with five but just for the sake of time we're gonna talk about two was hipster Henrietta so this is like a Brooklyn like hipster like as

26:28the name implies who thinks quilting is cool because it's old it is making a comeback like the median age for quilters is lowering over time I'm just realizing I know more about quilters than I think I know about a lot of things and then the second persona here middle aged Mary which is basically my mom she's a little older than middle age but don't tell her and then I had my mom basically fill this out in terms of different hypotheses so like what do you think the most valued features given this product are what do you think the least valued features what do you think the willingness to pay looks like I had

26:59to explain to her what pack and LTV were and and then she said I was like how much do you think it's gonna acquire and we filled this out just very very tactically over lunch and we had the start right we had a central document that we could start iterating on and every single data point that started coming in we could validate or invalidate everything that we put into this particular grid and understand if we're improving on our buyers but also understand who we should sell to so to actually fill this out though for a lot of you guys or to test this you go to

27:29your customer and like for the love of God just talk to your customer everyone's talked about it today in some form it's really that important and if you're not doing enough talking to your customer just stop and start doing that that future can wait that marketing campaign can wait because you're gonna have such compounding effects when you get the right data from your customer base and this is really important in the context of pricing because cost plus pricing and competitive base pricing or just pure [ __ ] like just gonna say it like that flatly cost plus pricing in SAS or even information products this makes zero sense your cost relative to

28:04an enterprise type you know product or something in the hardware space really really doesn't matter especially in the sense that your marginal cost per user might be one to three dollars whereas your revenue or ARPU on the monthly basis for that user might be in the triple digits to put it another way like your customers don't care about your costs they care about their costs and that's something that's really really important to understand particularly because a lot of people look at their costs and then they think oh well let's look at our competitors as well and competitors aren't that great either because first of all like you're you're

28:37sensing and you're seeding your value to that competitor so for instance you're also assuming that your competitors done their homework on their buyers and judging by this room like it's pretty clear like the savviest of us just haven't done our homework in terms of our buyers but the other thing is you're probably not selling to the same type of buyer so for instance we used to use Salesforce we were you know we're a five-person sales team and what was really fascinating we didn't use 85% of the product we were going after the wrong product and we turned off of it very very quickly and we moved over to closed I Oh

29:10mainly because that allowed us to have a little bit more of a product persona fit and a lot of times you guys are even in really competitive markets such as CRM the CRM space you're focusing potentially on the wrong types of customers and so what's cool about this is you do keep those particular inputs in mind but you want a price based on value or use that as the main focal to your business and that's just kind of a very very ambiguous way to say you want to talk to your customers collect the right type of data and then cross-reference it with your costs and your competitors to make sure that well

29:42one you can have a business and two you're not ten X or ten percent of your particular competitor because then something might be wrong with your data initially so to do this it's actually pretty straightforward this is not rocket science a lot of people anticipate really complicated formulas and like a magic secret but really it comes down to that customer development so you're going to do is you're going to start with that buyer persona that you've kind of figured out you have it down on a page somewhere similar to what we did with my mom you have some knowledge about your customers make sure it's down on paper you're

30:16going to set up an experimental design which is a fancy phrase for just what type of surveys and what type of data are we going to collect from them you're going to go out you're going to collect that data segment it down which is probably one of the most important parts for discovering the right ways and right places to go for your customers and then finally consolidate and analyze it down make a decision and do it all over again what we typically recommend is having between ten and twenty different qualitative conversations and then following that up with a quantitative actual survey where you're collecting data at scale and the reason for that is

30:47because a lot of times if you have that qualitative conversation set you can actually pare down a lot of the different features or a lot of the different assumptions and then collect it at scale to kind of prove out if those assumptions are still true or if your hypotheses need a little bit of a different looking at and the reason this is also important is because when you look at a pricing page or even any manifestation of a pricing page you have two different axes here you have the features on the x-axis and you have the price on the y-axis and if you start to unpack your pricing page in that manner

31:16you can start to collect data along those two different axes to get to your particular pricing personas and make sure that you're aligning that particular data correctly for the best monetization strategy out there and so in terms of experimental design what we typically recommend is three different layers of data so demographic information and if you're a consumer product you might actually have to collect like some traditional demographics you know age household income things like that in the b2b space this might be things like you know company size company revenue etc there's future data that you're gonna collect an we're gonna show you how to do that or

31:53how we recommend doing that and then pricing information which is ironically the most straightforward stuff to collect and what's really kind of fascinating about this particularly in the demographic information what we recommend is never ask a question that you can reasonably get the answer to with a little bit of work so what's really really frustrating and we see a lot of surveys so our software our algorithms are contingent on getting survey data so we've sent about 15 million at this point and what's really fascinating about it is we'll see a survey that asks for the first question like for the email address even though the company had sent the email address

32:27too like that's how they sent the survey to someone and so just really make sure that your battle testing what you're asking with these in these particular surveys because you don't want to piss people off like we've all had 45 questions surveys that we just either didn't answer or we really just gave really bad answers to so typically you want to get a survey down to about four minutes or less and ideally it's between thirty and sixty seconds and that also allows you to train your customer base or your user base to basically send surveys between every three and four weeks and you can do this in app there's

32:58a lot of different ways you can get this data if you don't have customers or you don't have prospects you can use things like ask your target market a YTM com or different market research firms there's really no excuse for getting in front of someone who has this data now the tools that we're going to use to get the x-axis and the y-axis data are as one it's called the relative preference analysis I mean there's a price sensitivity analysis and we'll go through these in depth if you have any questions we do have blog posts that have been written on this so you can do all yourself and don't fear don't fear

33:31the data that's all I have to say here in terms of you know digging into this so what do people value so we're gonna start with that x-axis in terms of what features should go where or even validating things like value metric like which value metrics should we sell on pricing preferences there's a whole host of things we can use so what we're not going to do is ever ask a question like this ever again in the history of your company these are the worst questions and it's intuitive why right because you get data that looks like this where you can't really tell if one is really truly

34:01better than the other oftentimes what ends up happening is if especially if you have sales marketing folks you'll everyone's a nine or a ten right everything is important well if you're gonna give me the option and not force me to make a decision I want everything if you send these types of surveys to dev or more technically focused folks typically what you'll you'll get some true answers but it's still not really a good statistical framework to use instead what we recommend is using something called max diff so max diff is a statistical model where you force someone to make a decision so you take the same features that you're just about

34:34to do that nine to ten ranking on and you basically ask them out of these four things what is the most important and what is the least important and the reason that this is so powerful is because on the back end in the math is really really simple I'd trust me like it really truly is is you start to get not only that rank order but you also start to get magnitude so what I can tell is that this first feature is actually three to four times more valuable at least from this customer perspective than that second feature and when I start to break this down by

35:04demographics you start to notice really interesting things here such as oh well these wealthier folks really really value this feature poorer folks don't value that feature and vice-versa for that second feature there and so you can start to see how this can impact your pricing and impact your modeling whether it's from marketing or sales now all of a sudden I know like well we kind of need these things here because these are most valuable and if we want a point of differentiation we can really separate these out and ultimately we're just not going to build that if this is the first things that we're building basically

35:36because we've broken this down on a value by value basis and if you know the demographics of your particular personas all of a sudden you can start to break this down on a persona by persona basis now this starts to become really really powerful when you do it not only for kind of top level features but we start to break this out and kind of a you know primary category and subcategory manner so we mean by that is we might ask this main category question where we'll ask people to compare for my mom's sake basic quilting supplies unique quilting supplies patterns other hobbies etc that's one question we'll get a lot of

36:12information on a very very categorical level but in addition to that we'll also ask a question same most least around basic quilting supplies they want thread fabric needles etc what that allows you to do is basically figure out what's the most important category and then within that category what's the most important item or feature in this case what's cool about this this is like taking this to a sass basis or even a software basis is you might ask you know your main you know main type of feature compared against support compared to enhance analytics compared against you know another type of you know main feature integrations that you have you might

36:51discover that support isn't the most important but maybe it's like one of the least important but then there's a really strong contingent of people who answered the support question with I really really want to dedicate at a count manager and all of a sudden that's the opening for a potential add-on that you might have intuitively known but all of a sudden you have actual data to back this up and when you break this down on a demographic basis all of a sudden you have this rich tableau of data that allows you to figure out well maybe we still need to build this particular feature but only if it's an

37:22infrastructure feature or only if we think all the data for our personas is wrong or instead maybe let's focus on these features that people actually care about and one objection to this oftentimes is well can't we just look at usage data which is interesting right because what we found is that for about half of companies usage does oftentimes correlate with value it's not perfect but oftentimes we also see that it doesn't correlate at all and to give you an example think of an accounting piece of software a lot of times invoices or invoicing will be the most commonly used for from a usage perspective I can just tell

37:58you with the six companies in the accounting space we've worked with invoicing was never in the top top half of features it's just something that's very very fascinating and that's why you need to go to your customer to kind of figure out like what they care about and you'll also find that with a lot of this information this helps in their sales conversations because if you're talking to a technical person you might lead with the most important feature for that particular persona and then mention some of the other features that are still potentially important to them but you at least have an aggregation of that particular customer profile so in that

38:29sense we've now filled in for my mom in this particular case what we can see in terms of most valued features and least valued features now willingness to pay it's actually super straightforward as I mentioned on this side what we're looking at is that y-axis so we've started to figure out those x axes and now we got to go after what the willingness to pay looks like we actually recommend asking these range questions now this model it's something that was developed in the 70s and 80s it's called the Venn westendorf model and the reason that this works out so well notice how we're asking range questions is because human beings don't

39:04think about value as a single point in time we think about value as a spectrum so I know that this water bottle is less expensive than this laptop it's a very very intuitive thing that I just kind of know naturally similarly I know both of these things are less expensive than this hotel very very dramatic example but you can take advantage of that psychological effect that we have to basically ask you know what monthly price point is this particular product way too expensive to consider purchasing it all the way down at at what point is it too cheap that you question the quality of it and this one's actually

39:36particularly important in terms of people who are really insecure in their products is oftentimes what we'll notice is oh your product is priced way too low that people just don't think you can fulfill the promise that you're trying to make and when you translate this data to kind of a visual format and do some little bit of math that's fairly simple will share the model afterwards if you actually see that each of those particular questions aligns with one of these lines so you see that at what point is way too expensive at what point is it getting expensive at what point is it a good deal and what point do you

40:08question the quality you get this really really nice diamond here that at least tells you for this particular persona the general place that you should be priced and if you go the extra mile here you get this really really nice elasticity curve and this is basically comparing percent of sales lost versus price point and the reason that's so important is because a lot of you folks your products you might not want to priced at the lowest entry point because the ROI might be there where you might make 2 to 3 X the actual revenue on a particular customer even though they're not necessarily an adoptive price but even further you might

40:46discover that those customers just suck for a different reason maybe they're really really bad at you know being retained by your product or even further they just might not be those customers that are going to give you that growth over time that you really really need so we filled in the second part the last couple of pieces there's a lot of articles that have been written about basically filling in things like CAC and LTV or estimating them a lot of you probably have some sort of estimate right now or some basic intro data which I recommend kind of filling that in for my mom we basically looked at will be

41:18anticipated average number of clicks would look like for a couple of different channels and then for the channels which we're going to test anyways we basically looked at alright where are these people and we did some really really basic google research and all of a sudden my mom here in this particular case knows well do I have a business let's look at the willingness to pay data versus what we anticipate CAC would look like measure that out we know kind of the initial things that we would need to build or the initial things that we would need to market and we also know kind of the channels that

41:46we're gonna start testing at and what's fascinating about this is when you translate it into a pricing page this is where a lot of the fun decisions start to happen because you're not going to make everything perfect your buyer persona to pricing page conversion will just never be perfect and frankly your pricing will never be perfect it's just like your products never perfect everything's always iterating and you should treat your pricing similar to how you treat your product similar how do you treat your marketing development you should always be testing something and doesn't mean just the price on the particular page but in this case we

42:15decided all right we're going to get some hopefully some network effects out of a free plan and we decided not to actually launch this until you know we got a ton of data from these buyer folks and then we didn't even launch the community yet so really we started with this basic quilting supplies box and a unique quilting supplies box and what's really really fascinating about this is this whole process took eight hours now I've done this before but what's interesting about it is wasn't eight hours like all in one swoop but it was eight hours over a few weeks and it only costs us twelve

42:45hundred and sixty four dollars and that was because we paid about $2 per response and asked your target market where we could get quilters you can get anything from a soccer mom or dad in Kansas all the way to a fortune 500 cio one costs a little bit differently than the other but what's interesting is depending on you know what your business and what stage you're in a lot of this stuff could save you a ton of time and money we did all this before we even launched anything and what's kind of cool about my mom in this case is we still have not launched a website we

43:15launched the product she's at about 5km our are in the past I think she launched about four months ago and everything is through phone and everything is through pamphlets at quilt shops sometimes I joked that my mom's kind of a drug dealer because it's like hey you know let's get that was a joke come on guys no she's not a drug dealer you saw or it was like very it would be a very weird drug dealer but a joke because she gets these calls which is like oh I saw the you know quilt box at the quilt store you know I'd really like to sign up for

43:42this plan boom booked you know she contacts them back and basically gets things through the door it's amazing what you can do when you actually think about who those buyers are and what you can do with that data so in terms of some of the other points here implementing a pricing process this seems intuitive but hopefully you guys have seen that this really isn't something that's super difficult it just truly is a process similar to your hopefully what you are doing some custom is and also your product development your road mapping your marketing development but we recommend is you need to evaluate your pricing every three

44:14months you don't need to do a huge research project but you really should be meeting whether it's just you or your founding team or maybe your exact team depending on your size and you should be making changes every six months and the reason that that's so important is because your product is probably improving at a much faster rate and if your product is improving every six months your price should be improved be improving every six months and that doesn't necessarily mean you're raising the price you might be lowering a value metric you might be adding a feature you might be moving the feature you might be

44:43structuring your page a little bit differently there's so many different things that you can evaluate with your pricing in this process we typically recommend is this is no one's full-time job so depending on your size you might assign this to a VP of Marketing or just your marketing hire or an intern even in certain cases meet in kind of the first week of the quarter basically decide like well how's pricing going do we notice you know some particular data point that came out last quarter or is there something that we want to think about testing set up an experimental design and do some of this market

45:13research and then depending on the severity of the change if it's just something really really minor that you don't need to announce the customers just push it and make sure you're collecting data and testing on it if it's something super super important you might do an impact analysis and you might discover like oh we were only gonna raise the price five dollars but we're estimating Chern is going to be out of control based on some of that elasticity data that we looked at and if you find that you might not make the actual change and then if it's super big we recommend is going back to a customer

45:42advisory panel and basically showing them and saying hey we're gonna do this are we crazy don't ask it that way but like what do you think of what we're about to launch normally that'll give you you know did we miss anything that'll give you some data on that particular access and then if your team is big enough what we typically recommend is having some representation from the core aspects of of your business product and marketing finance and sales and the reason for that is because pricing is so central to your business everyone has an opinion and it doesn't mean those opinions are wrong or right it just means that you want to

46:14make sure that everyone's kind of going in the same direction with any pricing change that you might be baking but the most important aspect of this slide is making sure there's some sort of decision-maker a lot of times you know and Heaton talked about this in terms of calendaring a lot of times we'll have people who collect a bunch of data they'll look at the data for like a year and they'll just never make a decision not because like they didn't want to it's just because they didn't get enough consensus and they didn't put a decision point down on the actual page to make sure that they were coming out on top

46:42and then the last point here is utilizing what's called a multi price mindset so if you remember from one of the benchmarks that we showed expansion revenues something that's pretty important especially in the contacts depending on what your gross turn looks like that's how you can kind of overcome a lot of problems but a lot of times companies that we see and this is a graphic here that's showing growth from new users and growth from existing users this is the relationship we've seen a lot of companies that are those kak fiends in reality you want both of these going up into the right and if you

47:11have both of these going up to the right that means your net retention is over a hundred percent you have negative turn and what's great about that is you can bake that into your model just your pricing model in a pretty simple way one is through features this isn't the best way but a lot of you probably are already doing this making sure that there's some sort of upsell opportunity where if someone turns out at the $49 plan and someone upgrades to the 149 dollar plan you've offset that turn and gain some but the best way to do this is by using a value metric so this is a

47:42company called Wistia there are also fairly self funded they have taken some funding but it's one of those kind of businesses where they do video hosting and analytics very slick product very slick marketing but what they've done is with their new pricing in particular they're pricing based on number of videos and there's also kind of a bandwidth fair use limit to kind of protect their back end or someone who has like very very few videos but huge amounts of usage and what's cool about this is they have a ton of people obviously on these particular plans and they might be churning they might be upgrading but they also have these

48:14really really large companies so they're people paying them thousands and thousands of dollars a month based on their usage because they scaled out this particular value metric what that allows them to do is offset a ton of this short and turn for some of those customers that might not necessarily be the best customers out there now to set up a value metric what we recommend is it's kind of basically aligned to your customers needs so per user pricing interesting enough oftentimes if you really look at it does not align to your customers need if you're a sales product if you were a helpdesk if you're something where someone logs in and sees

48:46something different than another person logging in then you have a really really good Pursey type model that you could use most companies are using per seat model incorrectly and and really should be using something like visits you know if you're a devops product there's just a world of things you could be using or you're tied directly to revenue you might actually use a revenue saved or revenue earned type model it's got to grow with your customers you should never be charging or Wistia should never be charging you know the disney channel the same price as price intelligently and further it's got to be easy to understand and this is the hardest one

49:19and this is why this like little triumvirate is really hard to achieve because oftentimes what ends up happening is you might have the perfect value metric well we save them a hundred million dollars last year but it's so hard to like get to where your value really is in that particular situation and so we recommend if you have a sales model you can typically have a little bit more of a complicated value metric but if you want something to be touchless sometimes you're just gonna have to grit your teeth in your value metric might not be the greatest but you might have something that kind of

49:47collates with your true truly best value metric so in all here monetization matters we went through some of that data happy to talk more about pricing I am a little bit of a one-trick pony because everything led to pricing here but it's really really important just to kind of recap here so you don't become one of those CAC fiends you have to understand that you have to focus on the right metrics and we're in a world now where there's no excuse for not knowing your metrics there's literally no excuse there are dozens of profit well competitors some of them you know some are great some are not so great

50:20but what's awesome about it is I I mean I don't really care why I do care who you use but frankly I care more if you're actually understanding your metrics and focusing on the right parts of your business take customer development seriously it's literally one of the biggest things that we see between companies that are succeeding and not succeeding and finally focus on your monetization more than you are currently there's all gains that we can make about pricing it's not a black box it's not hard just have to make sure it's a process and finally just understand we're all in the trenches we all have a lot of knowledge this isn't

50:55something that's super super difficult just takes a lot of work and don't do work that is unnecessary and don't do work that takes up too much time here's my email address if you ever want to chat that's where the slides are but appreciate the time

51:14when you're trying to figure out the willingness to pay should you focus more on asking those questions that people have never used your product so they're not biased or should you ask your current paying customers yeah great question so ideally you want three layers of data or three points of data you want your customers you want prospects so people have heard of you but aren't using you and then you want people have never heard of you and are your target customer so typically though we recommend to start somewhere like if you're very early maybe you just go and get people who've never heard of you because those are the only people

51:48that are out there um if you're a little bit later stage like the path of least resistance might be fire up a survey and send it out to your current customer base but it's really fascinating about some of this stuff is when you triangulate that data all of a sudden you'll start to see like okay well someone who signed up in the past 30 days how different they are than someone who you know signed up six months later or who's been on the product for six months or you can also like it's always worrisome when you see that people have never heard of you are willing to pay

52:15more but it helps you kind of level set what you're doing right and what you're doing wrong and there's a world where that data and we try to use it this way a little bit is actually better than things like NPS because you get that elasticity data around that segmentation they're really kind of helps you figure out like what we're doing right and who's willing to pay the most what they look like and who's not willing to pay and what they look like good question that was that was really good I love when people like you go up and kind of make more of an abstract thing and say

52:41these are the numbers this is what you do and the one thing I had a question about was you talked about the the price point a similar to the last question are you literally in the survey asking them at what monthly price does this become too expensive that you'd never consider purchasing is that is that how you're doing a question yeah if you're talking to someone like one-on-one like you're doing one of those ten to twenty first conversations well we ticket typically recommend is like don't ask all four it's a little awkward in a conversation we typically ask what point is way too expensive and what point is to a really

53:10good deal and you can qualify that to like because sometimes an enterprise product it might be oh this is way too expensive we'll just hire someone and so you can say like well what point like would you hire someone and what point would you just like not talk to us unless there's an issue and like that helps you kind of triangulate what you can do is and the reason it's important to ask open-ended even in surveys is because you probably are gonna anchor yourself in a weird way like we know companies that have tripled their price had no churn and just didn't even grandfather people and it's because

53:42they never did any of this research and like they've discovered oh wait we're like we're not priced not only we're not priced correctly we're priced like so low that people just think we're not a good product so if you do this more and more you might be able to start adding you know different attributes meaning maybe you say 100 50 75 you know some some actual inputs because your survey data like response rate will go up but I mean we've seen great response rates even on open-ended yeah sorry rambled they're a little bit into bubble points but that's yeah that's what we recommend oh we have a question over here you

54:17mentioned that most companies are using per user pricing incorrectly yeah can you give some thoughts on what correct usage might look like yeah exactly so there's a couple ways one like I kind of said what litmus is is if you log in and your experience is different meaning like there's actual like people aren't gonna share logins and get the same same experience that's a good way to actually price the other way the other kind of litmus is if you look at per user pricing in particular and you notice that the cap on the number of people in an organization to use your product even if they're a really large company is

54:55relatively low like think about a marketing team like a marketing team at like a venture back startup in a fortune 500 company a department not necessarily the whole come the whole marketing team relatively the same size right so the value doesn't necessarily correlate to more usage those are a couple of things that I would look at sales help desks you know slack those types of products are really really useful because people have different experiences if you have a product that people will share a login or people won't upgrade because they don't have more users even though they're using the product sometimes what you can do is limit users on the lowest

55:31tier it's a HubSpot does this really well where they limit the users on the lowest tier and then you know once you get up to the next year it's unlimited mainly because they're trying to protect small teams with big traffic or big contact rates essentially so we wrote a pretty big article on this too I can share it afterwards or if you download this stuff I can share with the group as well good question all right we have time for one more hi my name is Joe and I have to caution you I'm kind of bushwhacking in the swings I knew you were coming and I had this

56:05question in my head so it's just a it's a it's a tactic I've been toying with I've done some basic experimentation but we don't have enough data yet to make sure it'll work I kind of want to gamify my pricing I want to be able time right so I'm not sure if I love it yet I just like to get your opinion and maybe you won't have one but it's just tonight yeah how about if I give 5% off to every customer who brings me a referral and that new referrals got to be in the system so he's going to be looking at a dashboard where he sees you know I

56:37brought in five guys and one of them fell off so now I'm only getting a 20% discount instead of the 25 I got last month maybe take it down to some minimum but it happens that we're targeting a business that is itself very referral friendly relates to real estate industry it's real estate call it that no mortgage brokers in particular so these guys are custom to referrals they they you know that's a big part of their own business that's familiar to them and the math kind of works it may and when you take it down to a certain level you end up getting you know a reasonable

57:09percentage and increased rates yeah ever heard or something like that and the math works like on a spreadsheet that you may eight oh yeah you know just how much am i giving away those new customers it's it's a tolerable amount so yeah so what is so the basic idea of gamifying and getting customers involved in pricing as a referral model sure so so you're you're essentially incenting okay so there's a company called Paribas have you heard of them you connect your Amazon account and then they just crawl for any price decreases I mean like if they see a decrease they'll pass it on to you and they'll take a cut they do

57:46this like it works really well consumer based businesses I think the problem and b2b is that like one like referrals it's that naturally kind of happens at the highest rate already if that makes sense like no one's truly figured out like b2b referral systems unless there was some like Network effect in the product like slack like I add you to a group therefore I start using slack then I might open up my own channel my other issue with that and I'm not saying it can't work like it's something that interesting in the attest but I would I would question like why don't you just you like traditional referrals like you

58:19get a hundred dollars off or something like that and as a credit and the reason for it is because when you get into all those percentages you're basically giving a discount to that new customer for doing what they should already kind of do based on like how good your product or experience is I know that's really easy to say like I know I'm not standing here and I'm like yeah I do that but like I think oftentimes you never want discounting and I someone asked it but like discounts they should never they should always have an expiration date like people who give like lifetime discounts it's like well

58:46you should just lower the price to that point if that's the only way you can acquire those users but one thing you might look at it's actually free some sort of free product so like free just to keep in mind like we always get asked about free with pricing but like freezing acquisition model it's not a revenue model but what's really fascinating that's happening across all these companies is that the the basically the the willingness to pay per feature is steadily going down and so there's always this debate of like all software's going to zero right because people are actually paying for value now more and you can see that with more

59:16complicated revenue models so now things like AWS like you don't pay a set monthly price you pay a monthly price based on a bunch of different inputs like those value metrics and so what I would say is like give a feature-rich product that has some sort of you know value metric tied to it so a number of leads are a number of something and then like that can act as or either as like an interesting first trial or free trial basically like in disguise work and act in a way that like you can actually like get those upgrades and at least nurture that lead over time so I would just I

59:48would just think about that a little bit because like referral tactics and b2b like it's really hard to get them to work affiliate stuff in b2b works really well but it's it's one of those things where like you know if I'm using a tool I'm already gonna tell someone like you don't really need to even sent me too much make sense yeah thanks again Patrick [Applause] you