Skip to content

Barrava Blog

5 Oct 262 min read

Incrementality before attribution: a practical marketing test plan

A practical guide to choosing an incrementality question, treatment, control, outcome and decision rule before trusting attributed conversions.
Start Reading
28 Sept 269 min read

Message match after the click: did the page keep its promise?

Follow the visitor from ad promise to click and landing page. A worked example shows how to spot a mismatch, check the mobile experience and measure ...
Start Reading
21 Sept 2610 min read

Conversion rate optimisation starts with the decision, not the button

A useful CRO programme starts with the visitor's job, finds the break in the journey and tests a change that could improve a real business outcome. ...
Start Reading
27 Aug 267 min read

Install a website toolbar with GTM or a direct script

The same Barrava script works through your site template or Google Tag Manager. Choose the owner and release route first, publish the domain config, ...
Start Reading
24 Aug 266 min read

When persistent CTAs help—and when they hurt

Keep a CTA visible when it supports a visitor's current task and leaves the page usable. This guide shows when persistence earns its space, when it ...
Start Reading
20 Aug 267 min read

Floating action button vs floating action bar: how to choose

A button gives one action the spotlight; a bar keeps several routes available. This guide shows when each earns its space and how to test the ...
Start Reading
17 Aug 267 min read

Which website actions deserve persistent visibility?

Not every useful action needs a permanent screen position. Use purpose, location, audience, consequence and evidence to choose the few that earn one.
Start Reading
13 Aug 267 min read

Accessible sticky and floating controls: a practical checklist

A sticky control can work perfectly on its own and still block a visitor's next step. Use these page-level checks before it goes live.
Start Reading
10 Aug 267 min read

Design for attention, not interruption

A website can make its next step clear without making every visitor stop and respond. This article shows how context, timing and a worked page ...
Start Reading
10 Aug 267 min read

Why 90% confidence can be enough for marketing experiments

A 90% confidence rule can suit a low-risk, reversible marketing change. This guide shows what the number means, how to plan the test and when the ...
Start Reading