The 9x Effect

New products fail at rates between 40% and 90%. The 9x Effect (from Gourville's 2006 HBR paper) explains why: the people building the new product and the people being asked to switch to it are using wildly different scales to measure "better."

The math

Builder's perceived benefit  =  3 × actual benefit
User's perceived cost        =  3 × actual switching cost

Gap to overcome               =  3 × 3 = 9

• Builders overvalue the new product by ~3x. They're dissatisfied with the status quo (otherwise they wouldn't be building), convinced their innovation works (otherwise they wouldn't ship), and treat the new product as the benchmark because it's what they use every day. The IKEA effect compounds this — labor alone induces greater liking for the output of that labor.
• Users overvalue the existing product by ~3x. They've internalized the incumbent's quirks, built workflows around them, and paid the switching cost long ago so it doesn't register. They're subject to change aversion (loss aversion in disguise — they overweight what could be lost over what could be gained) and they don't know your product, so they discount its claimed benefits.

Multiply the two and your "2x better" product feels like "marginally better" to the user, which isn't enough to displace habit. To a user, the incumbent isn't just a product — it's a sunk investment, a habit, a mental model, and a set of workflows. The new entrant is an unknown with an unquantified upside. Beating that at 2x doesn't close the gap.

Why this matters for product strategy

Most founders I talk to pitch their product as "better" or "faster" or "cheaper" — often by a real but modest margin (1.5-3x). That margin feels decisive to them because they're measuring on the dimension they optimized. The customer, measuring across switching cost, trust, workflow disruption, and change-aversion, doesn't see a decisive product. They see a risk.

The 9x threshold isn't literally 9x on every metric. It's a rough heuristic for perceived advantage — the gap has to be obvious enough that the user overcomes their own inertia and risk aversion. Examples that cleared the bar:

• Google vs AltaVista/Yahoo (2000). Results demonstrably better on almost every query, not marginally. Within 18 months of launch, Google had displaced habit.
• Dropbox vs FTP/email attachments (2008). The "it just works" gap over the then-incumbent workflows was massive, not incremental.
• Stripe vs authorize.net (2011). Developer experience was 10x better, not 2x; the docs, the API, and the onboarding were all noticeably different tiers.
• Zoom vs WebEx/GoToMeeting (2015). One-click join vs. the incumbent's multi-step dial-in was an obvious gap to any user who tried both.

Examples that didn't:

• Google+ vs Facebook. Google+ had genuinely better features (Circles, photo uploads). Real users didn't care — the 2-3x edge couldn't beat Facebook's network and habit.
• Windows Phone vs iOS/Android. A respectable, sometimes better, OS. Not 9x better, and not enough to overcome ecosystem switching cost.
• Most ChatGPT-wrapper startups (2023-2024). A 1.2x better chat UX does not beat the habit of opening chatgpt.com.

The four paths to 9x

You rarely get 9x on a single dimension. In practice, sustained differentiation comes from stacking advantages:

1. 10x on the core job. Rare but decisive when it happens. Stripe's API, Google's search quality, early Zoom's video quality. If you actually have this, lead with it.
2. Orthogonal dimension. Win on an axis the incumbent can't easily compete on without breaking itself. Dropbox won on simplicity against enterprise file servers that couldn't simplify without losing enterprise customers. Substack won on creator monetization against platforms structurally committed to ads.
3. New platform shift. The underlying technology moved and the incumbent can't follow fast enough. Netflix vs Blockbuster, Instagram vs Facebook for photos, LLMs vs every legacy NLP product.
4. Lower activation energy. Bypass the switching cost entirely. Notion didn't ask you to migrate from Google Docs; it just became the place for new documents. Duolingo didn't compete with language classes; it filled idle time on the phone.

The best outcomes usually combine two or more. Stripe had (1) and (2). Zoom had (1) and (4). Dropbox had (1), (2), and (4).

Practical tests before you ship

Three questions to pressure-test whether your product clears the 9x threshold:

1. Would your target user recommend this to a friend unprompted, on day one? If they need to be asked, or need a week, the gap isn't wide enough.
2. If your biggest competitor added every one of your features tomorrow, would users still switch? If no, you're on a feature-parity path, not a differentiation path. That's a neutralizer strategy, which works at scale but not as a zero-to-one play.
3. Can you describe the advantage in one sentence that a non-expert understands? "10x faster," "works offline," "one-click onboarding." If the advantage requires a demo to see, the 9x threshold is probably not met yet.

The corollary for builders

The 9x Effect also explains why so many good products fail: the product was genuinely 2-3x better, but the perceived advantage was below the threshold. You don't always need to build something 9x better; sometimes you need to invest in closing the perception gap through onboarding, demos, free trials, and social proof. But those investments are substantial, and "marketing will close the gap" is a losing bet if the real gap is genuinely 1.5x.

If the product isn't 9x better, the honest answers are usually: (a) pick a segment where the existing product doesn't serve well and the gap is 9x for them, or (b) keep building until it is.

See also

• Change Aversion — the cognitive bias that drives the 3x over-weighting on the user side.
• MMRs, Neutralizers and Differentiators — the feature taxonomy that helps classify where your 9x advantage actually lives.
• What is a Market? — why segment selection is critical when the gap isn't wide enough for the broader market.