💻Digital & Web

A/B tests that give you the right answer.

Most A/B testing tools stop at p-values and conversion rates. Decision Process gives you Bayesian posteriors, credible intervals, causal adjustments, and the statistical rigor to make decisions you can stand behind.

Experiment Templates

Ready-to-run experiments

Checkout Flow Simplification

Reducing checkout from 3 steps to 1 page increases completion rate.

Conditions

  • 3-step checkout (control)
  • 1-page checkout

Metrics

  • Checkout completion rate (%)
  • Revenue per visitor ($)
  • Time to complete (sec)
Website sessions+18% completion rate

Homepage Hero Copy Test

Outcome-focused headline outperforms feature-focused headline for trial signups.

Conditions

  • Feature headline (control)
  • Outcome headline
  • Problem headline

Metrics

  • Trial signup rate (%)
  • Time on page (sec)
  • Scroll depth (%)
Anonymous visitorsOutcome: +11% trial signups

Onboarding Flow Comparison

A guided onboarding checklist improves 7-day activation vs. self-serve exploration.

Conditions

  • Self-serve (control)
  • Guided checklist
  • Video walkthrough

Metrics

  • 7-day activation rate (%)
  • Time to first key action (hrs)
  • Support tickets (7d)
New user signupsChecklist +23% activation

Email Subject Line A/B

Personalized subject lines with the recipient's company name improve open rates.

Conditions

  • Generic subject (control)
  • Personalized with name
  • Personalized with company

Metrics

  • Open rate (%)
  • Click-to-open rate (%)
  • Unsubscribe rate (%)
Email recipientsCompany personalization +8% opens

Pricing Page Layout Test

Highlighting the middle tier increases Business plan selection vs. a flat grid.

Conditions

  • Flat pricing grid (control)
  • Middle tier highlighted
  • Annual toggle default

Metrics

  • Business plan select rate (%)
  • Upgrade intent signals
  • Session duration on page
Pricing page visitorsHighlighted mid-tier +14% Business

Worked Example

Checkout flow A/B across 4,200 sessions

An e-commerce team tests a simplified single-page checkout against the existing 3-step flow. Traffic is split 50/50 via the SDK. Checkout completion and revenue per visitor are tracked over 3 weeks.

Results: checkout_completion_rate (%)

3-step checkout (control)

mean: 61.4%

95% CI: 59.8–63.0

1-page checkout

mean: 72.5%

95% CI: 71.0–74.1

P(1-page better) = 99.8%

The single-page checkout produces an 18% lift in completion rate (d = +1.1, very large effect) with 99.8% posterior probability. Revenue per visitor also increases from $8.40 to $9.90 (+18%). At this confidence level, the team can ship with high confidence. Recommendation: deploy 1-page checkout to 100% of traffic.

Run your first digital & web experiment

Private beta — tell us about your use case and we'll get you set up.