Table of Contents
Growth is rarely a straight line—neither in career trajectories nor in metric charts. Whether you are navigating the transition from sales to product or trying to figure out why a logically perfect onboarding flow is failing, the answer often lies in understanding the messy, human side of the equation. Laura Schaffer, the newly minted Head of Growth at Amplitude (and formerly of Twilio and Rapid), has built a career by looking past the obvious data to understand the psychology behind user behavior.
In her conversation with Lenny Rachitsky, Schaffer unpacks a wealth of tactical wisdom, ranging from how to advocate for your own promotion to why adding friction to a signup flow can actually increase conversions. Her approach challenges standard industry dogmas regarding statistical significance and developer marketing, offering a refreshing, psychology-first perspective on growth.
Key Takeaways
- Carve your own path: Do not rely solely on your manager to define your career trajectory. Instead, "un-gate" your knowledge and solve customer problems that become visible only to those on the front lines.
- Embrace "Good Friction": Not all friction is bad. In onboarding, asking the right questions can reassure users that they are in the right place, reducing anxiety and increasing conversion.
- Prioritize Velocity over Precision: In growth contexts, sticking to a 95% confidence interval can slow you down. Accepting higher risk in exchange for running more experiments often yields better net results.
- Understanding Developer Psychology: Developers avoid sales calls and marketing pages because they carry the burden of responsibility if a tool fails. To sell to them, you must provide a robust self-serve proof of concept.
Taking Agency Over Your Career Growth
A common pitfall for professionals in tech is relying entirely on the standard promotion cycle. Employees often work hard within their defined box, track their wins, and hope their manager advocates for them effectively. Schaffer argues that this approach limits your potential to your manager's influence and the company's existing org chart.
The "Voice of the Customer" Strategy
Schaffer’s tenure at Twilio began in product marketing, but she noticed a disconnect between executive perception and user reality. While leadership believed Twilio was incredibly easy to use, customers were telling Schaffer that getting started was a struggle. Rather than waiting for permission to fix it, she effectively "un-gated" her knowledge.
She began compiling a "Voice of the Customer" report—a digest of raw insights and pain points she heard from users—and circulated it widely. It wasn't long before senior leaders, including the CEO, began attending the sessions she hosted. By establishing herself as the authority on customer friction, she created the leverage to propose and eventually lead a dedicated growth team.
Your executive team and executive teams at companies are often very sharp but the nature of their day-to-day just does not link them with customers... That means that your superpower is in really pulling those insights in and bringing them to life.
The lesson is clear: executives are always hungry for ground-truth insights. By proactively identifying and sharing these insights, you build a brand that transcends your current job title.
The Psychology of Onboarding and Friction
One of the most counterintuitive lessons from Schaffer’s time at Twilio involves the concept of friction. Modern UX best practices often dictate that you should remove every possible barrier to entry—ask for as little as possible to get the user through the door.
When Friction Increases Conversion
Schaffer ran an experiment where she added more steps to the signup flow, asking users about their coding language, use case, and product interest. The hypothesis was that this data would be valuable enough to offset a drop in conversion. To everyone's surprise, conversion actually increased by 5%.
The reason was psychological. Users signing up for a complex developer tool often feel "imposter syndrome" or anxiety about whether they can handle the product. By asking "What language do you code in?" the interface signaled to the user, "We support your language; you are in the right place."
Ultimately the learning here was bad friction is bad and good friction is good. There's no such thing as being simple as just all friction is bad.
The "Pill in the Hot Dog" Method
In another experiment, Schaffer’s team found that developers were intimidated by the telecom aspects of Twilio (buying and configuring a phone number) but were very comfortable with coding. The original flow forced users to buy a phone number first—a high-anxiety step—resulting in drop-offs.
The solution was what Schaffer calls the "pill in the hot dog" method. They redesigned the flow to start with a familiar coding task (the "hot dog"). Inside that code, the user had to configure a phone number (the "pill"). Because the scary task was embedded within a comfortable context, users completed it without hesitation. Understanding the user's psyche at every stage is just as important as the functional utility of the product.
Rethinking Experimentation and Data Confidence
In academic and scientific fields, a 95% confidence interval (p-value < 0.05) is the gold standard because the cost of a false positive—like approving a dangerous drug—is catastrophic. In software growth, however, Schaffer argues that this standard can be counterproductive.
Velocity vs. Accuracy
If you strictly adhere to 95% confidence, you require larger sample sizes and longer run times, which drastically reduces the number of experiments you can run per year. Given that the industry average failure rate for experiments is roughly 80%, volume is your best defense against stagnation.
Schaffer suggests that growth teams should be willing to lower their confidence intervals (perhaps to 80%) to double or triple their experimentation velocity. While this increases the risk of false positives, the net result of finding more winning strategies often outweighs the noise. However, she warns that this must be paired with qualitative data. If the data looks good but user interviews suggest confusion, you shouldn't ship.
The closer you get to something that is... you go bury your head in the sand or go into an attic and build something for six months and ship it, the more likely it is that you are gonna ship the 80% wrong stuff.
Product-Led Growth and Selling to Developers
With experience at Twilio, Rapid, and now Amplitude, Schaffer has deep expertise in selling to a technical audience. The rise of "developer tools" has created a crowded market, but many companies fail to understand the unique purchasing behavior of engineers.
Why Developers Avoid Sales
Developers are notoriously averse to marketing pages and sales calls. They typically head straight for the documentation and the "Sign Up" button. This isn't just a personality quirk; it is a function of professional liability.
If a developer recommends a tool and it breaks production, they are the ones on the hook. They are the ones answering the pager at 2:00 AM. Because the stakes are personal and high, they cannot rely on a sales rep's promise that a feature works. They must verify it themselves via a Proof of Concept (POC).
This reality dictates that for developer-focused products, self-serve isn't just a convenience; it is a trust-building requirement. You must allow the user to reach a moment of value—and stability—before you ever ask them to speak to a human.
Conclusion
Whether navigating a career or optimizing a funnel, Schaffer’s philosophy centers on human behavior rather than rigid frameworks. Data is critical, but without the context of user psychology—why a developer fears a phone number, or why a new user feels anxious about their coding skills—numbers can lead you astray. By combining high-velocity experimentation with deep empathy for the user's emotional state, growth teams can uncover opportunities that logic alone would miss.
){kind=link}