How to Master SaaS Onboarding and Activation Rate

TL;DR

A lower activation rate in the 5 to 15 percent range often beats a higher one because it correlates more strongly with long-term retention, meaning you are pushing most users toward a meaningful state they weren't reaching before. Overhauling Airtable's onboarding raised its activation rate by 20 percent, and not every change needs an A/B test.

Transcript

an activation rate that falls in a lower percentage range maybe for most companies 5 to 15 percent is better than one that falls in a higher percentage range because it means that there's likely much higher correlation with long-term retention and you're really working hard to get most of your users to reach a state that they're not reaching today ... Read More

Key Insights

• Activation rate in a lower 5 to 15 percent range can be healthier than a higher one because it signals stronger correlation with long-term retention and shows you are working hard to move most users to a state they don't currently reach.
• Experimentation serves two purposes: understanding the precise metric impact of what you build, and risk mitigation when making dramatic changes that could be very good or very bad before all customers see them in production.
• A one percent precision gain, like activation moving from six to seven percent, often doesn't help the business much beyond letting someone claim credit in a performance review, so that precision may not justify the cost.
• Experiments are expensive because engineers, analysts, and product managers spend time understanding results instead of roadmapping, doing foundational analysis, or shipping, so teams should experiment only when they genuinely need to.
• The default culture of A/B testing everything grows as growth teams scale into organizations, but an intentional culture focused on doing right by customers and the business is the only way to escape that trap.
• Rewarding talent should not depend solely on pointing to A/B tested numbers, because a numbers-only culture biases teams toward constant experimentation, which is not always the right thing for the business.
• Airtable added a feature letting form submitters request an emailed copy of their submission, gated on creating an account, and rolled it out without an A/B test because customer research and data work already gave conviction it was valuable.
• Overhauling Airtable's onboarding led to a 20 percent increase in activation rate, showing onboarding is one of the biggest and most undervalued growth levers a product team can work on.

Questions & Answers

Q: Why can a lower activation rate be better than a higher one?

An activation rate in a lower percentage range, maybe 5 to 15 percent for most companies, can be better than one in a higher range because it likely means much higher correlation with long-term retention. A lower rate suggests you are working hard to get most of your users to reach a meaningful state they are not reaching today, rather than counting an easy-to-hit milestone. The tougher, more selective bar tends to track real engagement and retention more closely.

Q: What are the two main reasons a growth team wants to experiment?

Lauryn identifies two reasons growth teams experiment. The first is to understand more precisely the metric impact of what they are building and putting in front of customers. The second is risk mitigation, used when making so many big, dramatic changes that a launch could be really great or really bad for the business, and it would be good to understand that before every customer experiences it in production. She advises treating experimentation primarily as a risk mitigation tactic.

Q: When should you not run an experiment?

You often don't need to experiment when the added precision doesn't meaningfully help the business. Lauryn gives the example of activation going up six percent versus seven percent, where that precision mainly lets you say in a performance review that you increased activation, without changing decisions. Because engineers, analysts, and product managers spend costly time understanding results that could go toward roadmapping, foundational analysis, or shipping, low-risk changes with strong prior conviction can simply be rolled out.

Q: How can a growth team avoid the trap of A/B testing everything?

Lauryn says the only way to escape experimenting on everything is to build a very intentional culture about doing right by customers and the right thing for the business. That means creating other ways to motivate, reward, retain, and develop talent, such as recognizing impact measured by qualitative customer feedback, deals closed, or deals not lost, rather than forcing engineers within growth to point to A/B tested numbers to prove their impact, which biases teams toward constant experimentation.

Q: What is an example of a feature Airtable launched without an A/B test?

Airtable noticed a feature parity gap in Airtable Forms where a submitter could not request a copy of their own submission, for example remembering which T-shirt size they had ordered. The team built that feature, which was gated on creating an Airtable account, and simply turned it on. They skipped the A/B test because customer research and data work already showed it was valuable, and the impact on top-line signup metrics was big enough to see directly.

Q: How much did redoing Airtable's onboarding improve activation?

According to the episode description, overhauling Airtable's onboarding flow led to a 20 percent increase in activation rate. Lauryn discusses onboarding as her favorite topic and one of the biggest and most undervalued growth levers, drawing on her work redoing Airtable's onboarding. The conversation covers what she saw work, common pitfalls, and the company's unique segmentation process, alongside advice on figuring out the right activation metric.

Q: Why does Lauryn say experiments are expensive?

Experiments are expensive because they consume scarce team capacity. Having engineers, analysts, and product managers on the ground spending time designing experiments and understanding their results takes time that could otherwise be spent on roadmapping, foundational analysis, or shipping product. When the precision gained from an experiment doesn't change business decisions, that cost isn't justified, so Lauryn recommends experimenting only when needed and leaning on a rigorous product development process instead.

Q: What should teams do instead of relying heavily on experimentation?

Instead of defaulting to experiments, Lauryn recommends letting the product development process do more work. That means spending more time with customers, being more rigorous about understanding precisely what problem you are solving, getting mockups in front of people to see how they react, and building enough conviction that you are comfortable shipping to every customer tomorrow. With that rigor, the experiment often doesn't matter as much because you already trust the change is right.

Summary & Key Takeaways

• Lauryn Isford, most recently Head of Growth at Airtable and formerly a product growth lead at Facebook and growth lead for Blue Bottle's e-commerce, argues onboarding is one of the biggest and most undervalued growth levers, and that a lower activation rate of 5 to 15 percent can correlate better with long-term retention.

• She holds a contrarian view that growth teams experiment too often. Experiments exist to measure precise metric impact or to mitigate the risk of dramatic changes, but small precision gains rarely help the business enough to justify the engineering, analyst, and product time they consume.

• Escaping the experiment-everything trap requires an intentional culture that rewards doing right by customers through qualitative feedback and closed deals, not just A/B tested numbers. Airtable shipped a form-copy feature gated on account creation without an A/B test, relying on prior customer research and post-launch attribution analysis.