On the market for 15 years, Kissmetrics is by no means a new product nor was it built for the self-serve journey. Rolling out a free trial four months after I joined the company in 2022, it didn’t take long to see that running a Product-Led Growth (PLG) motion in an older product was anything but easy. 

Here are six lessons I’ve learned, along with practical guidance for other PLG leaders wrestling with similar restraints inside their older products 👇

1. Don’t run a PLG motion unless you have a reason to

If you’re debating whether to build a free trial or layer a PLG motion onto your product, make sure you have a good reason to. No matter the company or tool, a PLG motion isn’t something you can randomly throw against the wall just to see if it’ll stick. Running a PLG motion, and running it well, requires a lot of planning, patience, and commitment. 

This is especially true for an older product or a smaller company that can only take on one big bet at a time. At Kissmetrics, we converted our entire team to support the PLG motion—because the upside of our “why” outweighs any downside of potential risks. 

For us, our decision was based on market opportunity: Our biggest competitor is Google Analytics, which is forcing its customers to transition to a new product version, Google Analytics 4, in July. Since it won’t be a quick (or easy!) transition for Google customers to make, we expect folks will become frustrated and search for alternative tools.

To get ahead of things, and to compete with a free product head on, we built a free offering to let folks try out Kissmetrics, which has led to a massive uptick in the people evaluating our tool. 

2. Narrow the barriers across the user experience

Working with an older product, the challenge we have isn’t a lack of awareness or low sign-up volumes. (On the contrary, I get 100s of sign-ups each month!)

It’s that getting a PLG motion to work in an older product feels a bit like playing Whac-A-Mole:

Where are they getting stuck now?

What’s happening?

How do we fix it? 

For example, we’ll see 15% of folks implement the JavaScript on their website and never come back. Why? Did they lose interest in the tool? Find something different? Run into product friction? What can we improve to get them back?

Because unlike newer tools built for self service, Kissmetrics was designed for a guided onboarding experience, which inevitably leads to quirks within the self-service flow. In this way, it’s less about removing giant roadblocks in the product and more about narrowing the barriers so users can do what you want them to. 

Early on, this meant optimizing areas along the onboarding journey to deliver more value for our free trialers. For example, we axed a pop-up that required folks to immediately configure their domain name and time zone—this is something that can be done later on if they decide to add a second domain to track. 

We also used to drop new trialers into a blank instance, which made it hard to figure out what steps to take or reports to build. To remedy this, we created pre-built reports to show them the art of the possible inside the product. By narrowing these types of barriers, and helping folks experience value by creating a metric or report early on, you can drive faster momentum and stickiness inside your product. 

3. Evaluate (and then adjust!) the length of your free trial

The length of your free trial is something you should evaluate on a regular basis. Right now, Kissmetrics offers users a 30-day free trial, but I’ve been debating shrinking it down to 21 days.

Part of my reasoning behind this is based on the speed of implementation when working cross-functionally. But mostly, it’s that throwing people into a month-long trial leaves me working with an eight-week sales process. 

In other words, a spike in demo requests and free trials in mid-March won't close until May. Quite frankly, my product isn’t expensive enough to drag the sales cycle out for two months, especially given today’s economic conditions.

To evaluate trial performance metrics and figure out the ideal length to offer, we’re leaning into our own analytics in Kissmetrics, which measures both website and product data. B

ecause I can see what actions our free trialers are taking inside the product (and when), I can analyze the potential impacts of trimming the trial by a week.

Looking inside your own product, what timeframes do you see the biggest drop-offs in usage or activity? Do people need more or less nurturing in your product? Would reducing (or expanding) your trial positively or negatively impact on the deal cycle?

4. Dig into your data to uncover new selling insights

In a similar vein, consider what other data you have at your disposal to uncover new insights and enhance the sales motion. For us, pulling data such as referral traffic from Kissmetrics helped us identify new buyer behaviors, as well as validate different parts of the sales cycle. 

For example, Sales recently noticed that the demo requests coming from Google Analytics users seemingly doubled over the past few months. Digging into our web traffic data in Kissmetrics, I looked into the referral sites for both our demo requests and scheduled demos. 

Sure enough, my sales team was right—100% of the folks who scheduled demos with us in that time frame had previously visited the “Google Analytics vs. Kissmetrics” page. Whereas before, more than 50% had not looked at that page prior to scheduling a demo. 

These types of insights can be used to enhance your highest-traffic webpages, drive more focused marketing activities or campaigns, or prioritize what leads sales reps should be reaching out to.

5. Consider what other stakeholders still need your support

More often than not, one person can’t get to value alone in an older product. At Kissmetrics, our free trial lets business folks track some website behavior without any coding, but more complex scenarios, such as tracking in-product behavior or multiple web properties, requires engineering support.

This is also where we’ve historically seen the biggest stalls in the onboarding process, as well as in our free trial. 

I recently had an a-ha! moment when thinking about the onboarding experience from an engineering perspective. Our knowledge base has no shortage of context documentation and technical guides, however; we didn’t have a simple one-pager outlining the three basic steps Engineering needs to take to set up the product.

Up until now, they’ve had to dig through documentation in our knowledge base to scope out the work themselves. 

Consider what other stakeholders are traditionally involved in helping your primary user unlock value from your product. Are there opportunities to provide them better support or product enablement?

For us, giving marketers a one-pager to send to their engineering team turned out to be a simple win for remediating long-standing friction in our onboarding process.

6. Lean into sales to cover the early bumps and bruises 

Finally, don’t be afraid to bring your sales team in early on to smooth out any rough parts. Since we have the advantage of using Kissmetrics to see what folks are doing inside our product, we can look at our active users, or those getting stuck at known areas of friction, to prioritize who Sales should reach out to. While we may reach out to folks a bit earlier than other PLG teams, we do so because we know the experience is still a bit bumpy and it gives us an opportunity to provide more value. 

In terms of sales resources, we run light with one sales rep who’s taking point. Because he deals with all of the trials—which requires a ton of work—I comp him the same regardless of where the sale comes in from or whether he touched it directly. The way I see it, we’re early in our PLG journey, and still determining what a product-qualified lead actually looks like for us, so I want him to make money while he helps me figure it out. 

Will Kissmetrics continue its PLG motion with a free trial? 

Right now, I don’t have a definitive answer. Honestly, there are still so many bumps in the trial process, I’m not sure whether we’ve had anything “real” and powerful for users to play with yet to prove it’s a huge value-add. 

But as we dig into more and more trial data and keep smoothing out the bumps and bruises, we’ll iterate and evolve our findings. Luckily, we have a lot of data to draw a conclusion from. 

And for now that trial will stay up and running….but will not be something stagnant – a ‘set and forget’ offer. We’ll either eventually do it well or scale it back.

More frameworks like this one?

Follow Calixa to be notified when new expert frameworks and guides are published.