TL;DR: PostHog fits product teams already sending events β experiments sit on a public usage path, not a CRO quote. Convert fits teams that want printed monthly plans tied to tested users or tests. VWO fits CRO teams that will accept named plans (Growth/Pro/Enterprise) whose dollars often sit behind sales. Optimizely fits enterprises buying a modular quote.
A/B testing software splits traffic and reports a winner. Feature flags can power the same idea (see the feature-flag comparison). This page is the experimentation/CRO buy.
Every entry uses the same fields: published pricing, unit (events, tested users, tests), visual editor and the main limitation. The order below is not a quality ranking. Public meters first. Quote-led last.
A/B testing software compared at a glance
| Provider | Published pricing | Unit | Best for | Link |
|---|---|---|---|---|
| PostHog | Experimentation on posthog.com/experiments β free allowance then usage. Separate from product-analytics events | Product events + experiment usage as printed today | Product-engineering teams already on PostHog | Visit β |
| Convert | Public monthly plans on convert.com/pricing (2026 reads: Growth-style bands from a few hundred dollars / tested-user caps; some pages also show per-test bundles). Confirm live | Tested users and/or concurrent tests | A printed CRO bill without Optimizely paper | Visit β |
| VWO | Plan names on vwo.com/pricing. Dollars frequently sales-gated in 2026 | Tested traffic / MTU + modules | Visual editor + research suite buyers | Visit β |
| Optimizely | Quote on optimizely.com/plans | Contract, modular products | Enterprise experimentation programmes | Visit β |
Checked 16 September 2026. Convert is the vendor in this set that still prints dollars on a public page more consistently than VWO/Optimizely. Treat third-party $299/$490 screenshots as stale until you open the live card.
One tool rides an existing product-analytics bill. One still prints plans. Two quote.
Tested users are not pageviews
A tested user is someone in an experiment. Ten million pageviews and 40k tested users are different invoices. PostHogβs experiment module is also not βincluded in the 1 million analytics eventsβ by slogan β read that productβs meter.
The four providers in detail
1. PostHog
EXPERIMENTS ON THE PRODUCT STACK
Best for: product teams that already capture events in PostHog and want flags + experiments in the same warehouse-friendly tool.
| Published pricing | posthog.com/experiments and the main pricing page β free monthly buckets then usage |
| Unit | Platform events plus experiment-specific entitlements as printed |
| Visual editor | Not VWOβs marketer-first editor. Engineering-led |
| Free plan / trial | Free cloud allowances |
| Main limitation | If nobody in engineering will instrument, this is the wrong CRO tool |
- Turn experiments on only for the pages you will test.
- See the feature-flag and product-analytics comparisons for sibling meters.
- Do not use PostHog to avoid Convert if the buyer is a no-code marketer.
2. Convert
PRINTED CRO PLANS
Best for: mid-market teams that want a public price tied to tested users or concurrent tests.
| Published pricing | convert.com/pricing β named paid plans. 2026 third-party reads listed Growth/Pro dollar bands; confirm the current unit (users vs tests) |
| Unit | Monthly tested users and/or number of tests |
| Visual editor | Yes. Dev resources still help on complex sites |
| Free plan / trial | Trial. Not PostHogβs lasting free events |
| Main limitation | Flicker and SPA support are implementation work, not a plan name |
- Export last monthβs unique testers, not total sessions.
- If the page still shows dollars, Convert is easier to finance than VWO.
- Enterprise remains quote on large caps.
3. VWO
SUITE, OFTEN SALES-GATED
Best for: CRO teams that want testing plus heatmaps/surveys in one vendor and will complete a sales path.
| Published pricing | vwo.com/pricing lists Growth / Pro / Enterprise. Public dollars were often absent in 2026 checks |
| Unit | Tested traffic plus module packs |
| Visual editor | Core pitch |
| Free plan / trial | Explore/free paths exist; treat limits as unpublished until you log in |
| Main limitation | You may not self-serve a year-one number from the homepage |
- Send VWO the same tested-user scenario you used for Convert.
- Unbundle Insights modules if you only need tests.
- Do not reuse a January dollar screenshot as current list.
4. Optimizely
MODULAR QUOTE
Best for: enterprises standardising on Optimizely Web/Feature/Content and willing to contract.
| Published pricing | optimizely.com/plans β talk to sales |
| Unit | Product line + traffic + support |
| Visual editor | Yes, on Web experimentation |
| Free plan / trial | Demo / POC |
| Main limitation | Wrong shortlist if the brief is β$300/mo CRO toolβ |
- Scope Web vs Feature Experimentation as different SKUs.
- Procurement time is part of cost versus Convert.
- Do not pick Optimizely to beat PostHogβs free bucket.
Kurums Match: which one fits your situation
We already send events to PostHog
Use PostHog experiments with a sample. Do not buy VWO for the same flag.
We need a marketer visual editor and a public price
Convert. Bring tested-user counts.
We want testing + heatmaps in one sales cycle
VWO quote, compared to Convert + Hotjar.
Enterprise programme
Optimizely or VWO Enterprise.
Buying tip: write concurrent tests Γ monthly tested users. That pair is the RFP, not βwe need A/B testing.β
How this comparison was put together
Vendor pages, 16 September 2026. Where dollars are unpublished we say so. No paid placement. Kurums did not run a live A/B on production traffic for all four.
Order is public meter first. No badge reflects a commercial relationship.
Frequently asked questions
Is Google Optimize still an option?
Google shut Optimize down. This set is what teams buy instead.
Are feature flags enough?
For shipping, often. For CRO stats and a visual editor, not always. See the feature-flag article.
Why wonβt VWO show a price?
Usage + modules. Budget a sales call or pick Convert.
Did we test flicker?
No. Install the snippet on staging.
Discover more from Kurums | Business Intelligence
Subscribe to get the latest posts sent to your email.