BDTP. Implementing Self-service SaaS with Franciska Dethlefsen
34:29 Share

Hello everyone and welcome to Better Done Than Perfect, a podcast for SaaS marketers and product people. Our awesome guest is Franziska Detlefsen, Head of Growth Marketing at Amplitude, and we're going to talk about implementing self-serve SaaS today. This show is brought to you by UserList, an email automation platform for SaaS companies.

It matches the complexity of your customer data, including many-to-many relationships between users and companies. Book your demo call today at userless.com. Hey, Franziska. Hey, good to be here. We're so excited to have you as a guest. It's amazing we're going to talk about self-serve for a company with very complicated setup.

So it's going to be a lot of interesting questions. But before we dive into that, can you tell us more about yourself? What's your background story? Yeah, sure. I'm going to give you the full thing. So I am originally from Denmark, born and bred, was living in Denmark until I was about 20 years old. Growing up, I was a professional badminton player, so I spent a lot of my time inside playing badminton. Then

got the idea that traveling was my passion. So lived and worked or studied in Hong Kong, Malaysia, Singapore, Dubai, and then finally moved to New York to get into venture capital and spent kind of four or five years in venture capital in the States, as well as in Berlin and

And then I really wanted to go back into an operating role, which was what I did before, kind of working in small startups and found a data startup called Snowpile Analytics, where I joined as the growth manager, kind of building out the marketing function from scratch, more or less. That was a really awesome experience where we were trying to build a pipeline for our sellers on top of a really popular open source product.

So learned a lot about open source, learned a lot about driving pipeline for sellers through kind of your classic demand gen tactics, then met two co-founders of a small startup and started advising them on the side, eventually jumped ship and joined them at

their company called Iteratively as their head of growth. Very early days, first non-technical hire, kind of like 10 customers still finding product market fit. Focused there on building out our product-led journey, talking to a lot of customers and learning a lot about event tracking on the fly as well.

We were then acquired by Amplitude kind of eight months, nine months into my tenure there. So spent the last three plus years now at Amplitude, first in the product team as a product manager, then got the opportunity to lead our PLG efforts and eventually standing up our growth marketing team. And as part of kind of a small group of people across marketing product teams,

and engineering. We've launched our self-serve plan back in October of last year, which was really a company-wide effort, but a lot of blood, sweat, and tears from my team and the growth team and the engineers as well. Congrats on launching the self-serve. I'm going to give our listeners an executive version of what's been happening at Amplitude based on our previous conversation.

Essentially, Amplitude is, what is it, 14 years old now, and it's always been a combination of a free plan and a sales motion, completely sales-led company. And over the last two and a half years, you and Amplitude

your teammates built a growth marketing team and launched the self-serve plan. And now people can buy self-serve plans. That's the gist of it, isn't it? Yeah, yeah. That's a good story. So today we're going to go into the details of this amazing journey and how you've been able to do that in a sales-led company and what kind of sales motion and lead gen was happening before you started and what kind of things are happening now, if you can now tell us a little bit more about that.

Yeah, sure. So in the beginning, as we were starting the team and kind of setting up the pods, we were very focused on driving pipeline for our sellers. That was our North Star. It still is today. And we saw a lot of opportunity in our free plan. There were tens of thousands of accounts there using the product, happy with the product.

And it was not really obvious or clear to our sellers how to go in and talk to those accounts, who to talk to, what to talk about, etc.

So we did a lot of things there in the early days to improve our sales pipeline and ended up doubling our product source pipeline year over year because of all those efforts. So a few of the things that we did, one, we work with our ops and strategy teams to build a dashboard for our sellers where they can get an overview of their top accounts on the free plan.

what they were doing, what kind of features they were using, how many users they had, what their usage was, all the things that would be necessary for them to kind of own in on the accounts that were worth reaching out to proactively. Then we also looked at, you know, did some simple regression analysis and looked at the customers that did convert from our free plan. What were they doing before converting? How many users did they have? What were the demographics like? And we use that data to build processes

some basic PQL and PQA scoring, product qualified account scoring to proactively serve our sellers with the accounts that we thought were ready to have a sales conversation. And then we also built more upgrade moments in the product and drove more hand raisers directly from the product, creating visibility around some of the paid features that we have that are very popular that users simply weren't aware of or didn't think about.

And then finally, we also built and are still building lifecycle programs to promote new features or functionality or products that will require sales conversation as well. So lots of different areas that we invested in to drive more pipeline for our sellers from the product.

I'm assuming the pricing for the sales-driven product was rather high. So I'd love to hear more how you approached this question of shaping the pricing for this middle ground between your free plan and your high price.

Enterprise pricing, would it be correct to say that? Yeah, it's very accurate. And it's a big reason why we were able to launch our self-serve plan where users can swipe a credit card and pay us 50 bucks a month. In the early days, we were selling our growth plan, our then growth plan. We've made a lot of changes to it. It's way more feature-rich and a lot more products in it at this point. But back then, we were quoting companies

you know, 50 K 30 K. And it was a huge jump from paying sailor dollars to paying, you know, 30, 50 K for something. So the conversion rate from free to paid was, you know, abysmal 0.01 or two or something like that. Um,

Very, very low, but it wasn't a huge challenge initially because we have so many users on the free plan. So there was always someone that would pay us that amount. But competition and just market dynamics and everything else, we had to continue to lower our price points to keep up.

with what customers were demanding. And so now we had our sellers selling $15,000 a year plants. And as you lower that ASP, it becomes less and less efficient for sellers to operate in that space.

And so part of the business case for our self-serve plan, which is called Plus, was that, hey, let's pick up some of the slack from our sellers. Let our sellers focus on those 20, 30K plus deals and let the product do the heavy lifting on that $0 to 30K.

And in the first iterations of our self-serve plan, we were actually charging the lowest, I think, was 5K, anywhere between 5K and 10K a year, which was still quite high and way too high for a lot of our customers even to make that jump. So throughout our iterations, we ended up really kind of no-brainer. We just got to make it really cheap,

and really accessible to our customers and worth it as well, of course, creating value in that sell-serve plan. And so we dropped the price point all the way down to $50, but all the way up to kind of $40K depending on your usage.

It's pretty fascinating that you use the word self-serve as a way to self-serve purchase the product. But usually, for example, as we have customers going to us, self-serve usually means that they used to have super high touch onboarding only sales before. And now they're making the product itself accessible to the masses without concierge setup. In your case, you have had the freemium approach.

version, the free version for ages. So you had those open doors pretty much open. I'd love to hear about the mechanics of your free plan, what the limits are, what kind of support you have. And you're a data driven company, data heavy company, which means there's definitely some costs associated with supporting free accounts. So how does it all work?

Yeah, it's a great question. So we've had our free plan for many years. I actually don't know exactly when we launched it, but way, way back. And it created a groundswell. Like we got tons of signups from that. Obviously, it really opened the floodgates for broader adoption of Amplitude. And it's still one of the primary channels for us today, even for enterprise pipelines. So it kind of makes it worth the cogs.

with operating this free plan. So in terms of the mechanics and the plan itself, we have some key limits on it. You can only ingest up to 50,000 users a month. So someone who is growing fast and has lots of users will naturally need to upgrade after a while.

and we limit different features and functionality, but you can do basic product analytics on our free plan. And if you're not growing or if you're a small side project or a tiny startup, you can be comfortable in our free plan for years sometimes. But the ideal path for us is that within the first 90 days of using the product, you've seen value, you've gotten set up, you see that this is something that's worth trying

We'd love for you to kind of upgrade to our self-serve plan within that first 90. And at this point, we're getting close to 3% conversion rate from sign up to plus conversion, which is pretty healthy. Another area that we've heavily invested in over the past few years is our sales.

activation and onboarding, like you said, you know, require services and help to get started with Amplitude. And so we've invested heavily, especially our product growth and engineering teams in making onboarding as easy as possible. One line of code, low code or no code, uh,

ingestion sources so folks like you and me can get started with Amplitude by ourselves, just making Amplitude a lot more easy to use and accessible. And then we're investing heavily in continuous education as well. We have our academy, we have community where we get other Amplitude users to showcase their use cases and just try to

create an environment where if someone is eager to learn more about Amplitude and how to use it, there are different paths for them to take in product in our academy or in our community. So since your product is based on the data sources, the data integration, your strategy is to allow for any lightweight usage of that. And any data source is better than no data source. So let them just...

have something coming in from Zapier and be good kind of angle. Not trying to help everybody adopt a proper data integration. While we have almost doubled our activation rate, it's still kind of below, I would say, industry benchmarks because it is hard to set up

And we don't want you to compromise completely on the data that you get in, because if the data you get in is really poor quality or not very usable, you're not going to have a good experience in Amplitude. So we really try to prescribe a way to get data in that is reliable.

reliable and trustworthy. So one line of code gets you kind of default event tracking that gives you a fundamental set of accurate and rich data that you can get started with. And then we have things like our starter templates and dashboards where once you have those events instrumented,

you will get out of the box insights. So we try to kind of reduce the time to value there by making it super easy to get started one line of code and then populate some basic, not always basic, but some interesting insights for you to get started and then kind of learn from there. Once you have your insights, you will ask more questions of that data and you will kind of advance in your learning of Amplitude as well. Now you've seen a chart, how...

Next step is to create your own, apply filters, whatever it is. I'm assuming that at 50 bucks a month, you can't really do white glove anything for people, but you're using some scalable ways of human help.

one-to-many, you call them, to help support user groups. How do you do that? Tell us more. Yeah, we have a few different tracks. So obviously Amplitude Academy is available to everyone who wants to up-level their skills. We have our community, which is doing a lot of kind of live webinars and office hours where we find Amplitude Power users that help other users out

ramp up on their amplitude knowledge and share interesting use cases, like how they do retention or how to do acquisition and amplitude or whatever it is. And then part of my team is solely focused on creating lifecycle programs and in-product nudges to drive the right behavior. So encouraging users to add new use cases or try a new chart

or kind of advance in their onboarding and adoption of Amplitude continuously across the journey, which helps users in a one-to-many fashion, right? Like we're just sending emails to everyone based on kind of the current usage in the product, what we think they should do next, and all based on kind of, yeah, the behavior that we are tracking ourselves in Amplitude.

The community aspect is huge, but getting that off the ground for a small product, that's virtually impossible, honestly, very heavy task. Does this go back to very early years when you had the free plan and tons of users and you made it work somehow, and then you just kind of have this resource? Or did you specifically grow it over the last few years?

Yeah, so I think it's about a year and a half ago that we started thinking about community at Amplitude. Our then VP of revenue marketing had seen, you know, community being successful at her previous companies and was eager to make the investment herself. And so Esther, who was on my team, kind of raised her hand and she's a go-getter, a

just rock star extrovert yeah well she's she would say tell you she's not but she has to be a little bit right she's like let me build a community that that seems like a challenge and she more or less single-handedly build a thriving community now that we have across slack we have a newsletter and we have these live events both online and in person that really are just focused on

helping end users get better at their job and get better at using Amplitude. We tested different things in the beginning. Could community be an acquisition vehicle for us perhaps in the future? But over time, as we were speaking to users and Esther talked to hundreds and hundreds of users, it was really clear that what everyone was craving was more Amplitude.

Like, help me. I don't know how to do this. I don't know how to do that. I want to see how others are doing things. And to be honest, like I am learning every time I participate in a community session. There are so many power users and new users of Amplitude that are doing things

cool things that we never even thought were possible with the features that we've launched. So really the learning from others and kind of spreading that learning has been really powerful for not just our self-serve customer base, but also our enterprise customers. Congrats on nailing that for sure. Did you do anything specific to seed the initial efforts for getting that off the ground?

We did a ton of stuff. I mean, we started our Slack space. Esther started building kind of an events calendar. We spend a lot of time figuring out what did people care about and what would get someone into our sphere. So we did a lot of live sessions with experts on topics that were highly important to a lot of our users. And then from there, it kind of

to screw from there, right? We got people in through our events. We encouraged them to join our Slack. And then in there, we encourage conversations and build kind of programs that were part Slack, part live event. We started dabbling in in-person events to kind of get users together in real life, which has the extra kind of power of

of meeting people and connecting and then really focused on just, you know, amplitude and getting better at using the product or better at your job, like how to define your activation metric or just really up-leveling people's skills and doing so in a really transparent, open and easygoing way. Like you're here to show a failure. That's great. We can all learn from that as well.

Speaking of the community, you mentioned before that community is the type of support you provide for your free plan. So your free users cannot really pick the brain of your support agents, can they? No, they can. We definitely help if it's something really critical, like a bug or similar, but basically

Not really. No. On the free plan, we do not offer email support. We offer community support, which means we have a community space where you can write your question. And then we have power users of Amplitude who jump in and help you or sometimes Amplitude ourselves jump in and help.

But ideally, we're trying to encourage as much as possible other users helping other users. And that's all we can kind of realistically afford on a free plan with tens of thousands of customers. But one of the benefits, obviously, of upgrading to our self-serve plan is that you would get email support. So we kind of also try to differentiate it that way. It's not ideal, but it's hard to justify kind of the human intervention at that scale.

What's your philosophy around different onboarding tactics and mechanics when that self-serve user is dropped into your product? There's obviously a gazillion ways you can approach this from tool tips and tours to blank states to video splash screens. At the moment, what's your best guess at how to help the user adopt a technically challenging product?

Yeah, right now we're really focused on making it as simple as possible. So I already mentioned kind of the single line of code. So we want to make it a no brainer. If it's someone technical signing up, they can drop that single line of code in their product quickly and get on their way.

If it's a non-technical user getting started, we have built out integrations with GA, Segment. You can import a CSV just to see your data visualized. We've opened up the ways for non-technical users to get started with Amplitude. So really focused on using the product.

and driving kind of the key steps that we need you to do in order to successfully onboard into Amplitude and build that into kind of your classic onboarding checklist, which the product growth team and an engineering team has been working on

tirelessly on for the past year. And we've seen a lot of progress. Like I said, we've almost doubled our activation rate during that time. And then kind of from there, my team kicks in as well. We send various lifecycle sequences with the right instructions at the right time. Kind of now you have data and here's what you do next.

or you don't have data in here, alternative ways to get started. So we kind of try to nudge the user and create those actions that we need them to take in order to be successful. And then we also do things like one-to-many office hours that we host for any and all users that are trying to get started. They can jump on a call with a solutions request

consultant and we'll take them through the process and kind of give them the basics of how to get started and give them the opportunity to ask questions as well. And then they can always jump into the community and ask questions there as well. And then once they're activated and happy, hopefully happy, uh,

We kind of, you know, kick into kind of how do we make sure they adopt the features that we think they should adopt? How do we encourage them to evolve their use cases with us? And how do we make sure that they're aware of all the things that Amplitude can do for them? And then

Ideally, you know, within that 90 days, they will upgrade to our plus plan, swipe that credit card and start paying us. And then from there, more lifecycle kicks in to make sure that they're leveraging the features on the new plan that they're on and that we create that stickiness for them. So they will continue to use the features, see value in them and remain customers of ours. How did you arrive at this 90 day mark?

Well, I mean, it could be 30 days as well. Right now, our North Star is the 90-day window. It takes time, honestly, to onboard into Amplitude and see the value. So 90 days seems like a realistic timeframe for you to get set up,

get the data in, get your first insights, try a few of the features, and then run into some of the limits of the free plan that will encourage you to upgrade to the next tier. So I think 30 days is ideal, but I think 90 days right now is kind of the

the sweet spot for us giving users a good chunk of time to get going with us and see the value in the tool. But in the future, I could see us kind of shortening that timeframe and see how we kind of get users to upgrade faster and see value faster. Do you have a single specific ideal customer profile that you're tailoring everything to? Or like what are key personas? And also,

adjacent to that. It could be a set and forget kind of tool, but you don't want it to be. But there can be different usage patterns based on the role. You can be logging in once a month. You can be logging once a day. How do you even think about this? Yeah. So a healthy amplitude customer is someone who has two or more users in the product active two or more times a week.

So that's a pretty strict engagement rhetoric, but that is our engagement North Star. We really want Amplitude to be a weekly usage pattern. You have to use it twice a week in this case. And like you said, we don't want you to set it and forget it. So we continuously invest in things like Slack

or email to kind of get you, nudge you back into the tool and, hey, there's a new insight or here we're spotting an anomaly. Jump back in. So our engagement North Star is two or more users active two or more times a week. On ICP for self-serve, we have a few. One is obviously the startup.

someone or small team that's growing. We have a startup program for small teams as well, where they get a free year of our entire platform. So startups and SMBs clear ICP for our self-serve plan. We also have, like I mentioned earlier, the side projects and someone having an app on the side or really small business or whatever it is that are perfectly fine on our free plan. And then we also have

uh, larger companies, uh, and enterprises, uh, testing out amplitude this way. So they sign up for a free plan and they get some data in, they, they upgrade to our self-serve plan, uh, and they test that for three to six months and, and, and get a flavor for, for what amplitude is. Maybe they don't want to talk to sales yet. Maybe they're just piloting some different tools or maybe they are talking to sales and doing kind of a POC with them involved. Um,

But we're actually seeing a good amount of larger companies going this route too, to understand what Amplitude is and really get a good feel for it before going down the full sales journey. One of the challenges you mentioned on the mental slash team side is that you used to be a completely sales-led organization, right?

And obviously your sellers, as you call them, could be offended or somehow hurt by this new direction. How did you make peace with this internally? Yeah.

Yeah, our idea seller, it was definitely a long journey. And luckily, we had already built a lot of trust with our sales teams in the early days through our product-led sales work and driving that sales pipeline for them through the product. So there was a clear understanding that the product was a powerful lever for sales and a

So that was, you know, fundamentally, we were all aligned behind that. We also all aligned that our sellers shouldn't be focused too far down market. Like it's not affordable or efficient for our sellers to sell 5K, 10K plans. And we knew we had to make an offering for those types of customers that were willing to pay a smaller amount for a less feature-rich plan.

So we kind of aligned around the fundamentals there. Where it got more tricky is kind of the possible cannibalization that our self-serve plan would have on our sellers. So one of the strategic decisions we made as a company during that time as well was to focus a lot of our sellers more upmarket, to actually give them more time and space and focus, rather

on these larger accounts and then kind of little by little show that the product could do the heavy lifting in that sub 10K, then sub 20K, now sub 30K deal sizes, that the product would take that. And so the sellers could focus further upmarket. And then finally, as we kind of launched and grew the Seltzer business, we've seen a lot of up

And so now the sellers are seeing that there's a lot of customers who are already paying us, who are already seeing value. And it's a lot faster to get them to buy from us.

faster sales cycles and a lot easier sales conversations to talk to someone who are using our product, is happy with it and want more of it than someone like you're demoing and building a business case and really kind of selling them on something. So the product is doing a lot of the selling for our sellers, which is why we were able to kind of bring us all together to get this out in the world.

What's the typical mechanic of a plus user converting to an enterprise growth plan? Is it based on volume or certain features that are only available in the bigger plan? Yeah, both. So once a customer gets to a certain size, sales is able to offer volume discounting that we're not able to offer on our self-serve plan. So at some point around the 30, 40K, it becomes more affordable for you to talk to sales and get a...

a quote and a custom pricing based on kind of your needs and how fast you're growing. So volume is definitely the main one for us, but also features like we have additional products that are available only through our sellers, as well as specific features and functionality that isn't available on our plus plan, our self-serve plan. And so continuing that product-led sales work

We've been building a lot of feature gates and upsell moments inside the product to have users hand raise if they're interested in something that is on our enterprise plans. Just from a logical standpoint, if it's cheaper for someone to be on the enterprise plan based on high volume, what's the logic of converting them to the higher plan then? What's the mechanical benefit of having them on a sales plan?

Sure, yeah. It's not cheaper per se in terms of what they're paying now, but as they're growing. So the plus pricing is fixed based on your MTUs, monthly tracked users. On the sales side, we offer, you know, as you grow, like...

way beyond what the free plans or the self-serve plan supports, sales pricing becomes more affordable as you scale. So it's not like the price point is going to be cheaper. You're going to pay for more features and functionality. You get more support. You get a lot of extra, but you get more scalable pricing as you grow, which most enterprises need because they have lots of events or lots of users or both. Thank you so much for giving us a deep dive into how you made this happen at Amplitude.

If you were to give advice to our SaaS listeners for them implementing their self-serve plans, what would be one do and one don't?

Yeah, there's so many do's and don'ts. I think one do is maybe test something. If you don't have a free plan today, if you don't have a free trial, test it out. Is there demand for a free trial? Is that going to open up your acquisition further? You don't necessarily need a free plan. You don't necessarily need a self-serve plan. It's not for everyone. But thinking about putting the product first.

more front and center in the buying journey is becoming more and more critical for SaaS companies to succeed. End users demand it. They want to see, they want to try before they buy. So letting users do that, I think is going to be critical, especially if your competition is already doing that. And then I think the don't,

is I don't know if it's counterintuitive to what I just said, but before you launch a self-serve plan, you've got to make sure that you have healthy metrics in terms of your acquisition metrics, your signup metrics, your activation rate, and your retention rate. If those are not healthy or trending towards healthy, it doesn't make sense for you to launch a self-serve plan. Like we were discussing, like if it's really hard to get started by yourself,

How would someone get to kind of value and buy an option for you self-serve? Doesn't really work. So before dabbling in self-serve, make sure that you have healthy activation rate. Make sure you have a healthy retention rate. Users aren't leaving you after a few months. They actually continuously use your product for the long term. And those are kind of prerequisites before you go into the self-serve world.

If beyond these do's and don'ts, you have any other burning tips, would love to hear them as well. Oh, burning tips.

I mean, diversify your channel mix. This is something that we're doing and investing in right now. Being heavily reliant on your brand or on word of mouth is great, but you got to find ways to reach more audiences and new audiences. So take a look at your channel mix. Do you have a healthy percentage coming from non-branded organic products?

from referral, from social? And what are the ways you can scale those channels to get more users in at the very top of the funnel? Thank you so much, Francesca. Where can people learn more from you and learn more about Amplitude, obviously, if they want to do that?

Yeah, sure. They can learn more about our self-serve plan at amplitude.com slash plus. So take a look there. And then if you're a startup, go to amplitude.com slash startups. We have lots of offers for both small startups as well as large ones. And for me personally, you can find me on LinkedIn. I post irregularly, but we'll try better, but definitely connect and would always love to chat PLG and self-serve with anyone. Yeah.

Thank you so much. Wishing you good developments with the self-serve program in the future and have a wonderful rest of your week. Thank you so much. Thanks for having me. Thanks for listening. You can find a written recap for this episode at usaless.com slash podcast. Please help us grow by leaving a review on iTunes.