#7 - Franciska Dethlefsen @ Amplitude
#7 - Franciska Dethlefsen @ Amplitude
Introducing a self-service plan at Amplitude
“You can't just release the thing and expect that people will find it, use it, and pay for it. You have to educate them, make it visible, show what the feature is." - Franciska Dethlefsen
Audio links
Video
Podcast notes
[00:18] Meet Franciska Dethlefsen
[00:57] The Journey to Self-Serve at Amplitude
[06:45] The Evolution of the Self-Serve Plan
[10:26] Addressing Concerns and Building Confidence
[17:57] The Impact of the Plus Plan Launch
[21:22] Lessons Learned and Future Opportunities
[25:27] Closing Remarks and Preview of Next Episode
Transcript
We're going to have a look behind the scenes on how the new Plus plan at Amplitude came to life. Before we get to the details, can you start by giving a bit more context of this new self-service plan that you introduced? How is this different and why did you do it?
Amplitude has had a free plan for quite a while and that has been a big part of how we have been driving sales pipeline. So a lot of our customers start on our free plan, use it for a few weeks, a few months, a few years, depending on who they are. And we've been using that free plan to help sales look for the right opportunities at the right time.
But as we've been doing that for the past few years, we learned a lot from our customers around their needs and also our sales process. So we heard from our customers that Amplitude was somewhat expensive. Going from paying zero to paying $30,000 or more dollars a year was a big jump.
Simultaneously we also continuously went down market for especially the SMB segment and started selling our plans at a lower cost to meet the customer where they were. So from a sales perspective, it also didn't make much sense to have humans in the process selling these lower price plans that were hovering around $10, $20k a year.
So that was a big part of the motivation of introducing a self-serve plan. It was what customers wanted and it would drive efficiency internally, so we can have our sales team focus at the higher price points. So those were the main motivations internally and externally.
Besides that, we also made some significant changes to our pricing and packaging. We've been expanding our suite of products and actually have more of a platform offering now, and we wanted to bring that to our free and self-serve customer base as well. So with the introduction of this self-serve plan, we also made some fundamental changes to what we offer.
We're actually giving away a lot more for free now, on the self-serve plan. We want to give our customers what they need to be successful across experimentation, analytics, and customer data management. So we're giving that away for free in small increments so they can get started with what they need as an early company and can grow with our Plus plan.
When the original approach was put into place, I'm sure that the company would have considered that there was a segment of customers that wants to buy at a lower price point. What had changed that made the company reconsider?
That's a good question. I mean, we had different people in the company, including myself, that were advocating for this. We’d seen in the market, some of our competitors offering it. So it was really about internal pressure, also from Elena Verna who was our interim Head of Growth. She was a big champion of this, as well as myself and others.
We were putting this idea forward with the leadership team and slowly convincing them that this was the right approach. That involved lots of conversations, slide building and model building around the opportunity we had in front of us if we introduced this self-serve plan and the efficiencies we could drive as well.
So, I don't know if there was like a single point in time, but I've been working on this personally for over two years now. It's been through many different phases of testing and discovery and experimentation and many different phases of like leadership buy-in and convincing to conclude that this was the right approach.
You mentioned two things there, experimentation and model building. How did you build the confidence in the organization that this was the better approach, because some of these things you'll only learn afterwards…
I can talk a little bit about the experimentation we've gone through, testing the plan in market without promoting it officially. And then I can talk a bit about the other part of convincing, which is building a proper forecasting model around what this could bring in from a revenue perspective, including what possible cannibalization we would see from the sales-led business.
On experimenting with the plan: we've had three versions of the self-serve plan live in the product. The first version went live August of last year, so considering it’s December now, this has been a long time in the making. The goal initially was to show that it was possible for the product to sell itself if we made that available in the product and the first iteration was pretty basic.
We shuffled features around and introduced this V1 of the Plus plan in the product. We didn't tell anyone about it and we only showed the plan to very few of our free customers. We had sales tell us what customers to exclude, so we excluded anyone with an open opportunity and anyone that sales was trying to go after based on usage.
We exposed the plan to a very small group of customers and we also got very few purchases of that plan. But V1 was really mostly about showing the company that it was possible to put this offer in the product and that the product could, to some extent, sell itself.
The product didn't have a lot of bells and whistles around paid feature discovery, we simply put the plan in there. And without any work, we could show that it was getting purchased and people were using it. We slowly worked towards convincing our sales teams and our leadership team that this is the right approach and they got more and more bought in.
So that was phase one. Phase two, we reworked pricing and packaging and we opened it up more. We released it to all customers in the product, no more sales exclusions or restrictions on who could see it. We still weren't able to promote it or put it on our website, it was still only in product. And slowly but surely we showed again that having more people have access to it means we have more people purchasing it.
We kept iterating on that version for quite a while, and then we had some broader changes around pricing and packaging, as we were going to market more as a platform. So we had the opportunity to make a V3 of the offering, which is what it is now. That includes more features from our additional products, combining them into that self-serve platform that we have today.
From there, we set up an official launch date and got it out the door. We had a soft launch of V3 as well, making sure everything was working as expected. Then we slowly rolled this out, again, without a big splash. And then on October 17 is when we officially launched this in the market and made a big splash about it as well from a marketing perspective, got the pricing up on our website, and officially brought it out.
But yeah, that was a, almost a year and a half after the first version of the plan was conceived in the product. So a really long journey, which shows how much convincing and how much testing and how many iterations we had to go through before we felt good about what we were officially launching.
Yes, that sounds like a diligent release process. Can you talk a bit about the concerns that were coming from the sales team or the broader organization? What was the tipping point to get people onboard with this plan?
Yes, I can speak to that also in relation to the model building and forecasting, which I didn't touch on earlier. We spent a lot of time building out the revenue opportunity of the self-serve plan with very realistic numbers based on the data that we were gathering through our alpha and beta versions of the plans.
And with that same lens, we assessed the impact the plan had on our sales pipeline now that it was in the product, to ensure that we weren't cannibalizing sales pipeline. We wanted the Plus plan to be additive, not a realization of what we were making through the sales motion already.
We knew that would be some cannibalization and that would be ok. We actually do not want sales to sell the $15k a year plan. We want them to focus on those more lucrative deals and have the product sell that $15k plan. So understanding cannibalization was a big concern.
The fear was that we were going to sell this plan at a lower price point, which is great for customers, but weren’t going to make the revenue that we need to grow at the rate that we need to. So in all the models that we built, we focus a lot on cannibalization. We mapped out what cannibalization could be from net new customers that would now go on a self-serve plan, as well as existing customers that would potentially downgrade to this new plan.
We were able to level set with leadership and the sales org that while some cannibalization would happen, net-net this would be additive and show that over time how this could become a significant part of our revenue - a healthy self-serve business that runs 24-7. It's not relying on humans to be out selling or not on vacation or whatever else. This is going to be an engine that will power the business for the long run and help sales team become way more focused and efficient.
Can you say something about the main levers you have in the model you created?
We're still fine tuning the model as we get more data, which is great as we learn more about how the plan is acting out in the real world. We can continue to fine tune what the forecasting will look like in the coming years.
The main inputs we are thinking about in this model are really the full PLG journey. We can influence how many new users come to our website every week, every month. We can influence how many of those users sign up for our free plan. So how many of the visitors that we have end up signing up. And can we increase that conversion rate? That's the lever in itself.
Once they've signed up, how many of them activate? How many of them actually get started with Amplitude? Amplitude is a relatively difficult product to get started with. You generally, not always, but you generally need engineering resources. It's not a single player tool unless you're an engineer. If you're a PM, you gotta ask your engineer for help. If you're a growth person, you also have to ask an engineer, and maybe you need permissions to other tools where the data is. It's not easy.
So our activation rate has been hovering around, I would say 20%, 25%. And so already there, that's a huge lever. Can you improve the number of users who activate in your product? That's something that will have downstream effects on monetization.
You also have to keep these users engaged over time. So as they activate and start using the product, do they continue using the product? And do we have what we need to keep them engaged? That's another lever.
Then there's monetization. How many of the customers that are active in our product or of our total customers will buy the Plus plan? There will always be a big percentage that won't buy. And how can we improve that percentage? We're looking at best practices around 3% conversion rate of total customer base, which we're not at just yet. But working towards that and having that as another lever in the model.
And then we have our existing free customer base, which is a big part of our user base today. We have been having the free plan for years. So what percentage of these existing customers can we attract with this new offering as well.
Those are some of the main levers we have, that we can apply different levels of conviction and investment in. And that's also how we're structuring our teams, to go and focus on these various levers and improving the conversion rates between each stage in the journey.
I can imagine that when you're for example talking about the setup rate, which might require multiple people and systems, that you might also consider introducing a human touch here to improve that conversion rate. Is that something you've considered?
It's always a balance. I think there is generally a sentiment from the company that when we say self-serve, we mean self-serve. But my teams are always looking for ways that we can use humans in a scalable way. So this would be things like a one-to-many implementation with office hours where it's not a one-to-one experience,
We get lots of people in and they can ask their questions on a group call with a specialist that will guide them on how they get tracking started. We also have invested in education. The Amplitude Academy is a big resource where we have lots of courses and education resources. So that's another way where we try to scale humans in a way.
And we have our community efforts as well, where we run a lot of “How to” style webinars to show how you use Amplitude, how you expand your use cases, how you use the product essentially, and what the best practices are.
We're considering other experiments like what if we introduced a 30-minute call when you buy the self-serve plan or a check-in or anything like that we can do to ensure that you are adopting the features in the in the premium plan that you pay for, so you will see the value and continue to pay us for that plan.
One thing you mentioned is that even though you've been testing this plan for one and a half years already, you only went public with it recently. Even though it's only been a few months since the announcement, what have you learned since then?
Well, having it be officially out makes a huge difference. Having it out in the world and on your marketing website, on the pricing page, front and center, definitely makes a huge difference in terms of people signing up in the first place, but also how many end up buying it.
We actually have a good amount of customers that get started and immediately upgrade to this new plan, which was something that was surprising to us. We thought everyone was going to start on the free plan for a while before reaching one of our limits and then upgrading to the paid plan.
Another learning we've had so far is that about the tiers customers buy in self-serve. Some of the sales leadership and leadership at large were skeptical around the product being able to sell $20k of $30k a year price points. That seemed like a big investment that people would want to talk to sales or human to do those things instead of putting down a credit card.
But when we launched the latest version of Plus, we increased the tiers that you were able to purchase. And we've actually seen a decent amount, I think north of 10% of customers, buying these higher tiered plans that some of us were skeptical anyone would go and do.
That was proof that this could also work in the tier where sales potentially could still operate. And obviously there are benefits to self-serve. You can pay monthly versus yearly. And a lot of people just don't want to talk to sales. They just want to go their way and use the thing. Maybe they've used Amplitude before. They don't need a demo. They don't need help. They just want to get started. So that was another big learning.
The other learning and muscle that we still have to build out really is around retention and churn. Monthly payments, very low priced items, we go as low as $60 a month, also increases the chances of customers churning. They're just testing it out or maybe they didn't like it or maybe they didn't see the ROI.
Churn is something that we anticipated to be high, but are still navigating and figuring out how we best address it. How do we make sure that when someone starts paying us that they feel that it's worth it and continue to feel that it's worth it?
This is really about education and making sure that our customers adopt this new plan, which is not something that we've done in the past because we didn't have the plan and we didn't invest a lot of resources in that part. So that's an area of opportunity for us to make sure that we retain these customers at a higher rate.
Before we close, is there anything you want to share?
Yes, I think I didn't touch on something that changes when you go from sales-led to product-led., which is what we’ve done. We've had a really amazing sales team that has been able to sell our product and now platform. They've been talking to our free customers every day, convincing them that they should upgrade to our paid plans for these extra features, added benefits, etcetera. So they've done all this selling.
The product hasn't had to do any selling. It just had to be there and be useful. So as you move towards the product doing the selling, that's a massive work stream that you need to deploy dedicated resources to. You can't just release the thing and expect that people will find it, use it, and pay for it. You have to educate them, make it visible to them, show what the feature is, maybe even give them access to the feature in some way, shape or form.
That stream of paid feature awareness, discoverability, education, that cannot be underestimated and it cannot be separate from releasing a self-serve plan because you might release it and then not get the results that you want.
And that's definitely something we saw in the early days. We weren't really seeing a huge increase in free to paid conversion, but we had also not invested in that paid feature discoverability or awareness. So that's something for anyone who is exploring this route, going from adding product-led to a sales motion, you've got to invest in this.
You need dedicated growth marketers, you need dedicated engineers, dedicated PMs, maybe even dedicated data analysts to understand. the user journeys and what works and what doesn't, and continuously optimize this experience because this is going to be a key lever as you grow as well.
It's an efficient investment and there's compounding effects that you don't get with sales team. But it's an investment nonetheless. So don't forget to do that, because otherwise you're going to come knocking six months from now and be like, “Franciska, we did this, but we didn't get anyone to buy it” and that would be because you didn't invest in that experience that sales has been doing really well one-to-one for years. So that's a huge learning and an area I would double down on.
I could not agree more. Thanks for highlighting that and for sharing your learnings today.