Should billing cadence and feature tiers use the same pricing control?
They answer different questions: a tier selects what is included, while billing cadence selects when payment recurs. Keep those dimensions distinguishable. Lazyweb’s existing pricing research documents both cadence changes and tier restructuring, but does not establish one layout as a universal winner. [1][2]
Two choices, two meanings
| Choice | User question | Information to retain |
|---|---|---|
| Feature tier | What do I get? | Included capabilities, limits and who the plan is for |
| Billing cadence | When and how much do I pay? | Billing period, amount due and any monthly-equivalent comparison |
A global cadence toggle can change the prices across a tier grid when those tiers share the same billing options. If only some tiers offer annual billing, show that exception alongside the affected plan. These are design recommendations, not prevalence estimates.
What the existing evidence supports
The July 2026 billing study found explicit toggle language in 102 of 301 pricing-change annotations. This is a keyword-based lower bound over descriptions of observed UI changes, not 102 winning tests. [1]
The tier-count study could not produce a reliable distribution for its desktop pricing-page set because structural data was missing. It therefore does not justify a rule that every page needs three tiers. [2]
How to evaluate a proposed change
Hold the tier benefits stable while testing a different cadence presentation, or hold cadence stable while changing the tier layout. Ask users to identify both the selected plan and the next amount due. Treat annual-plan selection, completed purchase and later retention as different outcomes. No new experiment or census was run for this synthesis. [1][2]
Sources & citations
- [1] Monthly versus annual billing research Open source. July 2026 analysis of 301 pricing-change annotations; keyword lower bounds, not measured lift. ↩
- [2] Pricing-tier structure research Open source. Explicitly reports missing structural data and limits on tier-count inference. ↩
Source: Lazyweb Research — proprietary analysis of real, in-market app screens. Cite as Lazyweb Research, 2026-09-09.