Growth Experiment #2: Enterprise leads are people, too!
Clearbit Growth Experiments: This series shares the process and results of our growth experiments — the good, the bad, and the ugly.
Sometimes growth isn't about trying to convert more visitors, but converting the right ones.
Like many of our peers in the tech industry, this has been our focus since the pandemic broke out. Recently we pivoted to selling to a narrower ideal customer profile (ICP), who we realized found more value in our products, and relatedly, had higher Lifetime Values (an LTV-ICP, if you will).
The vast majority of our website visitors don’t fall into this newly-narrowed ICP category and, as a result, were leads we didn't want our sales team to prioritize. We had to change how leads made it to sales.
The challenge: how could we design a sales flow that increased hand raises from our target audience and decreased hand raises from non-target visitors?
Our original conversion flow offered the same options to everyone — you could request a demo or sign up for a free account to kick the tires.
The problem was that our sales team was getting flooded with demo requests from non-ICP accounts. We would’ve rejoiced about this lead volume pre-pandemic, but these days we’re all about focus.
So, we figured, a quick fix would be to automatically send our non-ICP visitors into a self-serve flow, and send ICP visitors to a form for scheduling demos. Simple, right? Wrong. So wrong.
Hypothesis
Our objective was to reduce the number of unqualified leads scheduling demos with Sales to protect their time, and secondarily increase the number of target accounts signing up for demos.
Initially, we hypothesized that we could accomplish this by shepherding leads down two different paths: leading ICP visitors down a sales-assisted path and non-ICP visitors down a self-serve one.
The proposed signup flow used a single email field as an entry point:
Once a visitor entered their company email address and clicked 'Get Started,' Clearbit enriched the email address. If your company fell under our target accounts bucket, we’d direct you to a 'Request A Demo' page. Otherwise, you’d get routed to a self-serve lane where you could start a trial.
To help prioritize and scope experiments, our growth engineers always predict effort and outcome. In this case, we expected this to be a low-effort task yielding a high (< 60%) lift in conversions of target accounts signing up for demos.
Results
Ultimately we wanted to reduce the number of non-ICP leads talking to sales – but we ended up reducing the number of ICP leads, too!
We saw a 20% dip in ICP demo requests after removing the self-serve option. Instead of nudging people to talk to sales, we were nudging them off the page altogether.
Why it failed and how we fixed it
We have two hypotheses as to why this failed:
-
We’d created a bloated, disjointed experience for our highest-value visitors. When people entered their email address and clicked a button saying 'Get started,' they probably expected to be thrown into the next step, like a trial or self-serve flow. Instead, we sent them to ... another form — for the demo request.
-
We offered less choice. We thought that by eliminating the option to create a free account instead of scheduling a demo, the only action they’d take was to schedule a demo. We totally forgot that they could simply bounce as well. Funnily, all of us on the growth team prefer tinkering with a product before talking to sales – why should we expect a "high-value visitor" to feel differently?
As growth marketers, we can become so singularly focused on attracting email leads and converting them through the funnel that we lose sight of what those emails represent: real people.
As we went back to the drawing board, we recast the challenge. Instead of assuming no one fell under our ICP category until proven otherwise, we assumed everyone was an ICP account until proven otherwise. This meant designing an optimal experience for everyone, using Clearbit behind the scenes to categorize visitors and serve different experiences.
Quickly, we worked to make up for lost ground and found some success with the following steps:
- Personalize CTAs on the backend. We nixed the form field method of diverging paths altogether and created a catchall 'Get Started' button that everyone sees.
Non-ICP visitors are asked to enter their email address, Clearbit enrichment confirms that they are in fact not in our ICP, and then they are sent through to the self-serve flow.
- Show scheduler for ICP leads to book demos immediately. After a visitor enters their email to schedule a demo, we enrich this data with our Enrichment API and ICP visitors can immediately book a demo. (We use a Chili Piper integration to do that.)
With these two changes, we drastically reduced demo requests from non-ICP visitors: bringing down the amount of sales time spent with non-ICP visitors from 70% to 10%!
Learnings
We took away three key design principles from this run:
- Optimize your website for your ICP first, and filter non-ICP second. Initially we did this backwards.
- Give visitors the option to self-serve.
- Reduce as much friction as possible for getting users on board. For us, that meant letting people schedule a demo immediately.
That’s great, but what if I don’t have Clearbit?
The foundation of this experiment is based on redirecting your target audience to a different signup flow, so it’s pretty difficult if you don’t use data enrichment and reverse-IP products like Clearbit Reveal and Enrichment. But without it, here are a couple of ideas:
- Move "Contact Sales" CTAs away from your homepage and onto your ICP pages.
- If you have a self-serve flow, ask questions during the signup process to determine if someone is in your ICP, and route them accordingly.
Scorecard
Our first experiment was a big fail, but here’s how we’d rate the three updates we made afterward.
- Effort: LOW
- Impact: HIGH, as it helped focus our sales team focus limited resources on high-value accounts.
Shout outs to Brian, Colin, and Julie who ran this experiment!Interested in another growth experiment? Check out Does social proof work?
