From monthly to biweekly: a conversion-rate story

When I joined IT Park as head of mobile and then CPO, the release cadence was roughly every four weeks. Upsell conversion into new services within existing products was running at 1–5%. By the time I left, cadence was two weeks and that same conversion was 15–20%.

The cadence change isn't the whole story, but it's the piece that made the rest possible.

Why monthly releases hurt

A four-week cycle sounds fine until you look at what it does to decision-making:

• Every bet is expensive. If you're going to ship a feature only once a month, you can only test two features a quarter with any statistical seriousness. That pushes teams toward "safe" features and away from anything where the answer is genuinely unknown.
• Feedback lag eats learnings. By the time a feature's results are in, the team has already moved on to the next monthly block. The learnings get filed rather than acted on.
• The release itself gets heavy. A month of work in one ship turns every release into a minor crisis, which reinforces the argument that you shouldn't ship more often.

The shift

We cut the release cycle in half. Nothing fancy — the teams already knew how to ship; we just stopped asking them to stockpile.

What got possible

• Running two or three upsell experiments per quarter instead of one.
• Catching activation regressions within a fortnight instead of compounding them across a release.
• Retiring features that weren't earning their keep, without ceremony — a biweekly release absorbs a cleanup easily, a monthly one can't.

The conversion story

Conversion climbing from 1–5% to 15–20% had many contributing causes — a brand-new onboarding system from the mobile team increased the activation rate by about 30% on its own, and the upsell copy got rewritten three times. But every one of those wins came out of an experiment, and an experiment needs a release to happen.

What I'd tell a new CPO

The fastest high-leverage thing you can do in the first quarter is halve the release cycle. Before you rewrite strategy, before you reorganise teams, before you hire. The shorter loop is what makes everything else you want to do run faster.