Hi, welcome to the fifth episode of the Growth Machines podcast. I'm your host, Vincent Jon, and today I'll talk with Leah Tharin about the implications of adding PLG when you have a sales team in place. The interview is part of my book about Product-Led Sales. If you're wondering how to combine PLG with your sales operation, then this book is for you. For more info, visit plgandsales.com. Now let me introduce Leah.

In her career in tech, she's held a number of roles in product and growth, for example, as product lead at Small PDF and as interim head of product and growth at JU. More recently, she operates as an adviser and has worked with dozens of senior leaders and companies on their growth strategies. She also takes interim roles like interim Chief Product and Growth Officer at Got Photo. Leah publishes her thoughts online as well as has produced some of the most read guides on PLG on her blog. Hi, Leah, great to have you today!

Thank you for having me.

So, you've worked with many different companies on their product-led strategy, right? To get us started, can you share what you're seeing in the market, what's working well, and what are some of the common mistakes that are being made on that front?

I think an interesting movement that we see is that traditionally, we thought about product-led growth and sales-led growth as kind of separate functions. Then, we started to have some combination in regards to Product-Led Sales, which was also a little bit fuzzy, you know, like, what are we doing with it, what are we not doing with it? The way that I'm looking into this is, we also see this with the positioning of some of the newer players that are on the market trying to serve this need. This is really about data-driven sales. On a conceptual level, I always talk with companies from two perspectives. The first one is: what we do really well in PLG is understanding the behavior of someone. We try to measure what they're doing in the product, but we don't really know who it is. So, we don't have a lot of knowledge about the firmographics. Whereas, if we talk about what sales does really well, it's the complete inverse. They don't know what people are doing in the product, but they know exactly who it is. If you think about a company account in five or six figures, it matters a lot who you're talking to. Is it the buyer? Is it a bigger company account? Are we talking to a team, or are we talking to a division?

What we see right now that seems to work really well is if you combine the best of these two worlds into each other as much as you can, and that's usually the starting point. So, this works really well. Rather than just looking at your pipeline in terms of we have self-serve and we have a sales-led pipeline, we try to combine it as good as it gets. That is much more complicated than it sounds right now, but I think that's a very good starting point. We see some movements in the market as well that this is going to, let's say, fortify itself. This is becoming much more important. We see this, for instance, with HubSpot, that was doing the Clearbit acquisition. This is about data enrichment of your leads, so you're trying to give a little bit more meat to who you're actually talking to. All of this is going into getting data for sales before they call someone up in the first place. So, we try to arm sales with data as much as we can—enrichment and also the behavior side—to help them close better. That is a different way of making sales more efficient. It's not just about getting new leads; it’s also about identifying which leads are worth reaching out to, and how can we drive our CAC and CAC payback more importantly down.

You know, to really get ahead of this commoditization that is happening right now in the market, where more and more players are starting to compete for the same amount of revenue per account.

When you're saying, like, we're looking at the behavior, there’s still a lot that can mean. What are the most typical types of behavior you're looking for? I mean, every product is different, but still, there are some commonalities.

So, the first decision that you have to make for yourself is: am I looking at the behavior of an account, and what is an account for me? For some products, this is a team, but it can also be that some companies have enterprise-level customers who are single users—like traders or anyone giving you a lot of money if they make a lot of money, such as asset managers. So, the first decision that you have to make is: on the scope, what am I analyzing? Am I analyzing the individual user? Am I analyzing the team (this is usually where we start from), or am I analyzing the account itself on the behavior?

Now, when you talk about a company like Slack, for example, which is also doing Product-Led Sales in some regard, then you talk about the account. Their typical customers are a collection of a lot of teams. So, they are still measuring on the team level, but they are defining some of their metrics on the account level. A very good example of this is: if you think about what Slack defines as a behavior that tells you this is a valuable account worth reaching out to, then you come across things like, they’ve sent 2,000 messages in the last 30 days across the account. For usage-based products like this, this is very valuable because they know that, as a commonality, it doesn't matter how many teams are in there, but if they start to send that many messages, the chance that the entire company is starting to onboard onto the tool is very, very high, and that makes the job much easier.

So, what we usually try to analyze as a very first step in every Sales-Led company is to get alignment on what is a common 'aha' moment that gives us a signal that is measurable, that the team or the account has now received value. They understand what the product is. They have moved from being promised something, or having an idea about something, to actually experiencing it. At this point, they should be informed enough to know what we're talking about. That's usually the conceptual side of this question.

And do you have any examples of what doesn’t work well? If you try this approach, what should people not try to measure? Vanity metrics or anything?

So, I think a very common mistake that people make is that they mistake engagement for a buying signal. For instance, when we're doing Product-Led Sales, what we're trying to make sure is that we really only look for signals that indicate buying pressure. That depends on your pricing, of course. It depends on your product. Does it make sense that if someone is using something a lot, they also want to buy it? That's not always the case. So, we're dealing now with a couple of new problems. Let's say you have a self-serve motion that gives you a lot of new leads, and you have, I don't know, a thousand users per month from your sales funnel. Just because you look at the ones starting to interact with the product a lot does not mean that they are becoming good sales targets.

There's a complicated question behind this, where you try to figure out as a company, what actually is a good buying signal? Rather than telling you a very specific signal that is always wrong—which I cannot tell you—I can tell you that processes that do not consider this and are iterative about it usually fail. It is very rare that the signals you're starting with stay in place the way they are after six months. So, the very first approach we have is usually just a starting point. We see what works. In Product-Led Sales, it's very commonly not just one signal, but multiple signals.

For instance, a good one would be if you have a very specific page that you'd only look at if you're actually interested in buying the product—like a very detailed pricing page or some kind of SLA agreements. Even though those by themselves are also not buying signals, they are important if you want to figure out whether they are more likely to be interested. Then, those together are usually generating a score, and if the score goes above a specific threshold, we would consider them to be interested parties that want to buy from us. That's the principle.

Figuring out this threshold and what kind of signals go into this is quite a piece of work and also a reason why I make money with it as an adviser, giving companies context on where to start and how to get there.

Right, right. You mentioned something called buying pressure. Was that included with what you just said, or what did you mean by that?

One of the particularities in the market today is that it's possible for companies and people to love your product but have absolutely no buying pressure. One of the conceptual ideas we usually work with is the question: why should we give something for free on a perpetual basis? That is another way of saying "freemium." For a lot of companies, you would ask yourself, "Why is Slack giving a product for free if the people that are using it are not buying it?" You can use Slack indefinitely for free, unless you want access to specific functions, like having access to all the messages that are older than 30 days or some other things.

The question is always: why should we do this? We should do this because buying pressure is not always equal to people loving your product. Why does that matter? Because the portion of free users you have, even up to the enterprise segment, is always higher than the amount of customers you have—those that are actually closing. This huge funnel serves an important function for you in many areas. One of these functions, the free users, is that they make it easier for others to recommend your product to someone who has buying pressure. It can be time; it can be just a wrong moment; it can be no budget; or it can be that they don’t have the use case. But they can still love the product, and this is something we have to separate in our minds: both of these things can be true at the same time.

I love that.

When we're talking about this combination of Product-Led Growth and Sales, in the end, what we're trying to do is make things possible—one is more efficient. But in some cases, we're making a go-to-market possible that just wouldn’t have happened otherwise, right? That just wouldn't be possible if you did it by yourself. Do you have any examples of that?

I think all companies right now have to deal with this. And one of the reasons is, if you're doing this well, you're doing it cheaper than the competition. A very good example of this is, I usually have calls with companies that ask me, "Okay, should we have a freemium or should we not have a freemium?" That's the wrong framing of the question.

So, I mean, in some ways it's correct, but the problem is not whether you should have one or whether you should not have one. The question is, if it's possible to deliver value for free to your customers or your pipeline in some kind of way and you're not doing it, and your competition does it, then your competition will be cheaper.

Why does this matter? Because at the moment, capital is not that cheap, and everybody wants to have efficiency, right? Like, we need to also work with the money that we're getting in. Most of the time, VC money is just harder to come by, so you cannot afford to let go of that efficiency.

That's the first reason. The second reason is that we see a lot of pressure on the very classical sales metrics, you know, like CAC, CAC payback, and what is a good payback period for whoever we acquire, and so forth.

This goes a little bit into the same topic in that there is something happening in the market. People are paying less for the same solution, expecting bigger quality and an easier onboarding. This is due to two factors:

AI – Everything becomes easier to build, everything becomes easier to compete with, and therefore, the prices are starting to come down. Interest Rates – Companies themselves do not have that much money anymore because of the interest rates we've had in the last two years. Therefore, if they have less money, they can spend less money with you. These two effects are kind of like making this a topic for everyone unless you are in a vertical that really cannot be commoditized in some way. Very few verticals are not affected by AI, so I just want to make this clear – we are, in some ways, always affected right now.

This is an interesting move because it makes companies much smaller for the revenue that they create afterwards.

When we’re looking at a sales team that's already in place, people often get concerned when the idea of a premium or giving something of our product away for free comes up. They start thinking: What if we introduce a lower price tier? Does that cannibalize our revenue? Are we giving something away that we could have sold?

How do you usually deal with this when it comes up?

Why are sales getting scared when this conversation starts? Because this is the one function in the company where you’re telling them their salaries are in danger. Very commonly, the way we incentivize sales is through an OTE compensation plan, where a bigger part of their money is actually coming from how much they close, and it’s usually when the signature is happening, depending on how mature your organization is, and so forth.

So, of course, we are threatening people's livelihood there.

How do we get this kind of story together that we were talking about with product-led growth? In the heads of people, product-led growth is seen as a premium, but that’s not what it is. Even with people who are very prominent in the area, they still misunderstand this.

It’s not about premiums. It’s about either self-serving value where it is possible. That can be a premium, but it can also be a trial, an interactive demo, or in some ways.

Conceptually, I always think about it in these three levels. For almost every company – even API businesses without an actual interface – you can think of self-serving value in kind of three stages.

One of them is a freemium, where we give something for free for an indefinite amount of time, providing some kind of experience with the product. It's usually limited by usage, or you have a trial where you can experience the full thing for a limited amount of time. Alternatively, you might have an interactive demo, which emulates what a product could feel like from different perspectives. Sendes is a really good company that does this quite well.

So, why does this matter now? Because it's very rare that I tell a company with a sales-led approach that they should jump straight into freemium if they have never done a trial or an interactive demo in the first place. To them, interactive demos are usually like, "Oh, we go to a marketing agency, they create a nice video for us, and this is somehow the interactive demo." We have a lot of new companies right now that are trying to take care of sales demos. For example, we have interactive demos from Ntic, Storylane, and tools like Gong, which is becoming much more sophisticated with their sales tools. There's a lot of market out there to make sales processes easier, and some of them are already dealing with this.

Now, when you jump to a freemium right away, you run into problems. Your existing pipeline will ask themselves, "Why do I pay $5,000 per user per year when you're giving it away for $30?" That's what's happening in salespeople's heads, and we need to address this slowly and carefully, of course. This is the reason why they are scared.

So, to your question, how can we avoid this? First of all, by making a layered plan. A very good approach on this usually comes from two ways. The first one is making sure that sales still earn the same money they did before, as individual people. The other one is that with an existing sales pipeline, you should make sure your pricing is more aggressive rather than very lenient. The reason for this is that we want to make sure that we can open it up. Making things cheaper is always easier than making them more expensive. But, yeah, that's usually how you want to approach this—be more aggressive with pricing, and with aggressive, I mean more preventative, so more expensive rather than cheap. The other thing is tackling them with compensation plans.

Yeah, I understand that. So, you'd rather have less conversion but get them in at a higher price? Correct. As long as you don't know what you're doing, you have to kind of figure this out slowly. So, we really try to figure out, "Hey, what does this do now?" The first time you have a trial going is probably also the first time for a sales-led company where they have to do something with this data. UX researchers suddenly have trial users—they've never had that before. They probably just had customers. Designers now have to deal with the growth onboarding flow—when do we limit someone? When does this show in the product? And marketing is also involved—how do we reconcile this?

There are a lot of structural changes that need to happen, and yeah, those are challenging for a lot of companies. So, do it step by step.

You mentioned an example where you go from $5,000 a year to a much lower amount. But what if a company is selling six-figure deals—$100K, $200K, $300K—and they're thinking, "Is PLG even for us?" I'd love to hear some examples of where you've seen that work well, if you know any.

Very good examples that combine both approaches require a little bit of an asterisk first. If you look at your company value—what you're going to acquire, like the ACVs we talk about, or how much money they pay with you the first time—then this view is a little bit dangerous. What we're really looking for in PLG or product-led sales is usually expansion revenue.

Expansion revenue is when we consider an account to be matured. Now, this is extremely important because it is possible that you are offering some kind of trial version that allows a team to onboard instead of an entire account for a very low price point.

What does that mean? So, to your example, let's say your typical product costs $100,000, and now I'm going to introduce a trial that allows you to use some parts of the product in a limited way for $5,000. Now, you might say from a sales perspective, "Yeah, but maybe we could sell them on the bigger plan," or whatever. But if you have a good strategy in this regard, then this might be the plan all along.

What we're trying to do is show people the standard freemium—"Hey, here's the free value, blah blah blah." If you have this, that's fine. We are starting to see a trend where companies—and I'm a big advocate of this, I think I'm actually the first one talking about this on a broader spectrum—try to use these trials to get a foot in the door with the bigger accounts.

What does that mean? If you're a product manager, for instance, using a tool that can also be deployed on an enterprise level, take a CRM like HubSpot. HubSpot works in a team, right? If you have an autonomous team, it also works in a team, but it can also be rolled out to the bigger company if you have a discretionary budget where you can say, "Hey, I want to buy this software for $3,000, and I can do it without approval.

Then what's happening is two things. The first one is, if you buy the software and we are now the product that is being sold into this company, then we are open as a vendor. That means we are in their system, we start to have champions in that company, and we're starting to spread. So, we're starting to grab some value from them, even though it's not the full one yet.

Now, if you have a good product like sales motion, you will know, with huge teams, for like let's say you're starting to sell into Microsoft. If we start to have, from I don't know, from Redmond for instance, five teams that are using our tool individually, then I can start to group this and I can start to reach out to whoever it is, right? So, like to the director of the entire department. I can start to move up, and even though the ACV will still be around 100,000, we demand much more money. It is incredibly useful at that point to say that, "Hey, we've been using these guys already for like one year." The due diligence process becomes much, much easier. The entire sales pipeline, like the duration to close, is going from 10 months to 3 to 4 months.

This is important to understand because you might think that you have four teams in your pipeline when you actually are starting already working on the big account. And that's how we think about this. So, you have to be really careful in thinking that, "Yeah, you know, sure, we take the lower value accounts. Okay, it's kind of fine, and we cannot convert everyone to the bigger accounts," but the goal is still to go after Enterprise. This is a very, very clear moment that we see also, like in the market, everybody wants to go upmarket, and the ultimate goal is to get Enterprise Value before it becomes Enterprise Value. And HubSpot does this also really well. But I'm going to pause here for a second.

Yeah, no, I think that's really interesting, and it's definitely, you know, that momentum you can build up much faster with a team and you can also close that deal faster, right? So, you cannot just compare the initial contract value of the two things with each other because that's just missing the point.

In the end, you said something interesting as well there. You said, "You need to look at the companies once they're matured." So, how do you measure that? How do you know this is all I can get out of this company?

So, there are two ways in this, right? So, like, let's say you get into a big company through a smaller team. That's what I just described, right? So, like, we have a team that is onboarding, and then eventually it might expand or whatever. But there's also another way. So, like, let's say HubSpot is doing this very well, and you know that they're doing this if you just look at their onboarding. So, what do they do? HubSpot could not exist as a company with a crazy valuation that they have right now if they did not bank on the fact that smaller companies, so you have already onboarded the entire company, but there are smaller companies that, some of them, will expand later on into bigger companies. Not just in your system, but also because they are growing. So, when you're working in tech and you have startups and scale-ups that are with you in this regard, then there is a chance that a percentage of them are becoming bigger as we go, right? Like, so also their revenue is growing.

Now, we sometimes forget that we are in SaaS and tech, very, very unique in this regard. Because if you just like investing in a mom-and-pop shop around the corner that is not in tech, you do not expect that to happen at all. So, the percentage of businesses that will grow with time is starting to get smaller.

So, how do we know that they're doing this? If you look at the startup program from HubSpot, they give you a heavy price discount in the first one or two years. If you do not manage to get out of this kind of growth zone at some point, then you're sitting on a relatively expensive HubSpot thing. Because if it doesn't happen for you in the first one or two years, then it's not going to happen for you at all. And why are they doing this? They want to make sure that if you're still sitting on the product and you're still liking it and you're paying enough, that's kind of what it is. But it should also not be that you have like these ultra-small accounts that are now in costs for you on customer success, you know, on the support and everything. Because if you were a paid member from HubSpot, you get access to all of this.

So, this is a very predictable function of revenue because you know that from about 100 accounts that are having, I don't know, an ACV of 2,000, some of them will expand later on. And just because by way of growing. So, you also should pay attention to and measure that. Was your question? Who, like, are we now at the limit of a company or not? And this is a moving goal, right? So, like some of them will grow. Like, if you see funding announcements, for instance, H-funding rounds, so like, hey, this company has invested, and so forth, then you know also from sales, hey, these people are going to grow. So, we can start to also predict our pipeline in this regard.

So, this is a very clever way of kind of figuring out, like, is this going to expand at some point? Should we now reach out to them? Should we make sure that they're not going to another competitor, like Salesforce or something like this? And yeah, this is something how we can engineer this in a clever way.

Good. Send me one more question before we wrap up, which is about like how to facilitate all this, right? You mentioned in the beginning we're tracking behavior, we're looking at firmographics, and we're starting to do quite some complex things. So, what's your preferred tooling to have in place?

So, either you hire Leia, someone like me, very expensive, but let's say you don't have the budget for this, which is for most companies not the case. I think for many, many sales companies—and speaking now here from the perspective of sales, and then we can also talk about PLG companies that want to move more upmarket—for most sales companies or sales-led companies that are trying to sell into the Enterprise market still and have been doing for all their time, it's very difficult to kind of make this data step.

And what I always say is that before you change anything on your go-to-market strategy, first thing that you really have to do is you have to sit down and make sure that you're talking about the same customers and their problems. And what I mean with this is that there's usually a very heavy misalignment between sales, product, and customer success in marketing. Because sales has probably traditionally, in these companies, been using product as an acquisition machine. And what I mean with this is that they are sitting in calls, then they hear, "Oh, if we just had this particular feature, then we could close this contract." That's very dangerous because all this does with the product is they start to build stuff that is good for acquisition—so closing contracts—but it's not necessarily good for retention.

So, this is the other side to this, right? So, product-led growth really makes sure that you're paying more attention to the revenue side. Sorry, to the retention side. Like, net revenue retention, does this account expand afterward? And this is why this is so important to start there and really understand, okay, what is happening in our product. Only if we measure this data—and that's usually already quite a journey—only if we measure this data, we can then use the data. And only once we have that, we can also incentivize sales on it. So, you cannot start with saying, "Hey, sales, start to analyze the behavior" if they don't have the tooling, which is usually not the case. Then you cannot do that.

So, analyze data first, have some dry exercises on how this can be used by sales, and then start from there. So, this is quite a long process. Conversely, for PLG companies who have already started, like, you know, we have self-serve, now we try to figure out how it is for product-led sales, it is a bit easier because they usually start to have inbound where that says, "Hey, do you also offer volume license deals? Like, do you have 50 licenses, 100 licenses?" and so forth. So, for these companies, it's much easier, but for both companies, one important principle is true: do not hire more salespeople than you have leads. Because whatever you do in the qualification of your pipeline, if we're using usage data, if we're using optimized processes and so forth, this is only valuable if some of the leads are not handled. That's what we're doing, right? Qualification of pipeline means that we identify the good ones. So, if you have too many salespeople, then you're starting to handle the ones. And then what's happening is that you have a pipeline that seems to be inefficient, if it actually could be efficient.

So, that's something where we have to deal with it, and that means that for the same pipeline, we need fewer sales. That is a reality, and if you're doing it well, that's also what's going to happen.

Yeah, of course. And you know, traditionally sales doesn't want to do it. So, like, just really quickly, an interesting number in this regard is that Battery Venture said that because of AI and all this new tooling, we will be able, at the end of, in the next couple of years, whatever, to handle the same pipeline with 75 people that took 125. So, we need to kind of find this efficiency somewhere that is driving this 33% increase in pipeline handling.

Yeah, and PLG can definitely help there. Yeah, if you didn't do it, then that's it.

Yeah, that's right. Great. Well, Leia, thanks so much for sharing your experiences today. I want to give you the floor as well, maybe for a brief moment before we finish up. Like, is there anything you want to share with the audience?

So, I usually say that you can visit my website, but my name is so complicated that you can just go to Google, you can type in PLG and then Leia, and that's it, and then you see all my stuff. I have written one of the most read guides around PLG. It's all completely free in true PLG fashion, and yeah, that's where you can start from. And yeah, my DMs are always easy as do.

All right, great. Thanks for joining.

Thank you, Vincent.

All right, that's it for today. I love the views on specific customer states, like customer maturity, and if a user has buying pressure. And the examples of how to gradually dip your toes into the PLG waters provide great guidance if you're considering bringing more automation into your existing sales process. For more insights like this, check out my book about product-led sales on PLG and sales.com. Thanks for listening, and see you next time.