Skip to main content
Solutions/Problem Aware/Saas
Problem Aware · Web Application

Ship variants, measure outcomes, keep what works.

A/B testing provides evidence for product decisions. Two versions of a headline, a checkout flow, or an onboarding step — measure which converts better. Implementing A/B testing requires variant assignment, consistent user experience, and statistically rigorous measurement.

150+
Projects shipped
99%
Client retention
~12wk
Average delivery
The problem
Product decisions made without experimentation — wanting to test multiple variants of UI, copy, or flows to measure which performs better

A/B testing has three requirements that are frequently implemented poorly:

Consistent assignment: A user should see the same variant on every visit. If the assignment is random on each page load, the user sees different variants — corrupting the data. Assignment should be based on user ID or a stable anonymous cookie, hashed to a variant.

Mutual exclusivity: Users in experiment A shouldn't also be in experiment B if the experiments affect overlapping behavior. Experiment interaction effects corrupt results.

Statistical significance: An experiment with 20 users per variant can't produce reliable conclusions. The required sample size depends on the expected effect size and the baseline conversion rate. Running an experiment for 2 days and calling it "significant" at p < 0.1 is invalid.

Implementation options:

Posthog experiments: Built on top of Posthog feature flags. Assigns users to variants, tracks conversion events, reports statistical significance. If using Posthog for analytics, experiments come naturally.

Vercel Edge Middleware (for Next.js): Assign variants at the edge before the page renders. No client-side flicker. Store assignment in a cookie; report events with the variant attribute.

GrowthBook: Open-source experimentation platform. Connects to your existing analytics to pull metrics. Self-hosted or cloud.

What can be A/B tested:

  • Onboarding flow (step order, required vs optional steps)
  • Pricing page (price points, feature emphasis, CTA copy)
  • Trial conversion prompts (timing, messaging)
  • Email subject lines and send timing
  • Feature discovery UX
What we build

A/B testing infrastructure with consistent variant assignment, conversion tracking, and result reporting

Posthog experiments

or Vercel Edge Middleware setup

Variant assignment

with consistent user-level assignment

Conversion event

tracking per variant

Experiment results

reporting with significance indicators

Guardrail metrics

to detect negative side effects

Engagement

One honest number to start.

Fixed-scope, fixed-price. The number below is the starting point — final scope is built from your brief.

Tier · Web ApplicationFixed scope
From$25,000

A/B testing infrastructure with consistent variant assignment, conversion tracking, and result reporting

99% client retention across 40+ projects
Process

Three steps, every time.

The same repeatable engagement on every project. No surprises, no mystery, no billable ambiguity.

01Week 0

Brief & discovery.

We send you questions, then get on a call. Output: a written scope with every step, feature, and integration listed.

02Weeks 1–N

Build & ship.

Fixed schedule, weekly reviews. No scope creep unless you change the scope — and if you do, we reprice it transparently.

03Post-launch

Warranty & retainer.

30-day warranty on every launch. Most clients stay on a monthly retainer for ongoing features and maintenance.

Why fixed-price

Why Fixed-Price Matters Here

A/B testing infrastructure is the assignment mechanism, event tracking, and reporting setup. Fixed-price.

FAQ

Questions, answered.

Until it reaches statistical significance, or a predetermined minimum sample size. A test that reaches significance in 2 days either has an enormous effect size or was underpowered. Plan for 2-4 weeks minimum for most conversion metrics.

Yes — feature flags with event tracking provide the basics. A dedicated experimentation platform adds sample size calculators, significance testing, and interaction detection. For early-stage products: feature flags + Posthog. For growth-stage: a dedicated platform.

Next step

Tell Ryel about your project.

Describe what you’re building and what outcome you need. You’ll have a written, fixed-price scope within the week.