How to Run an Appcues A/B Test on In-App Copy That Means Something
Most in-app A/B tests are decided by noise. This covers sample sizes, the metric to pick, and how to stop after one honest result instead…
Appcues is a no-code builder for in-app onboarding: tours, tooltips, modals and checklists that ship without a deploy. It is free for 100 monthly active users, and paid plans start at $299/month — a jump that surprises most teams.
Appcues is a Boston company founded in 2013. It focuses narrowly on the moment a user first meets your product and every announcement after that. The workflow is: point the visual builder at a page, drag a content block onto the real interface, set a target audience, publish.
Compare it to Pendo and you find less analytics, more message-crafting. Appcues is happier being the layer that makes onboarding feel designed than the layer that reports on feature adoption.
Speed is the whole argument. A marketer can publish a retargeted welcome tour in an hour without touching the release train. In companies where a copy tweak means a two-week sprint, that alone pays for the subscription.
The pricing cliff is real. Free covers 100 MAU; the Essentials plan is often quoted near $299/month once you pass that, and enterprise agreements land in the tens of thousands per year. Small startups frequently install it for free, then uninstall at scale.
It is not an analytics tool. You get flow-level metrics (views, completion, goal conversion), but no funnel or retention analysis beyond that. Teams expecting a Pendo replacement are disappointed. Pair it with Mixpanel, Amplitude or PostHog for the numbers.
Teams buying analytics. Use Pendo, PostHog or Amplitude. Also skip it if your app is a marketing site with no login — there is no account model to target against, so you would be paying for a fancy banner system.