When a Competitor's Pricing Page Shows Two Prices
Modern pricing pages are split-tested live. Open one in incognito twice — or compare two of your colleagues' screenshots — and you can sometimes catch a competitor running an experiment in real time.
Visit a competitor's pricing page twice in incognito and the numbers might differ. That's a live A/B test you can reverse-engineer if you catch it early.
The first time someone on your team says "wait, when I opened their pricing page it said $49, not $39" you should stop and pay attention. They didn't misread it. They got bucketed into a different variant of a live pricing experiment. Most B2B SaaS companies above a certain size are running pricing tests continuously, and the surface area where those tests run is right in front of you for free.
The catch is that experiments end. If you spot one in flight, you have a narrow window to learn from it before the company picks a winner and the page settles.
How to spot it
The easiest detection is the cheapest: open the competitor's pricing page in incognito, then again in a different browser (or from a phone on cellular). If you see different prices, different default toggles (monthly/yearly), or different "Most Popular" badges across the variants, you've caught a test.
Other tells worth bookmarking:
- Two teammates compare screenshots in Slack and the prices don't match.
- The pricing page has a feature comparison matrix that contradicts itself between visits.
- A specific tier disappears in one variant and reappears in another.
- The CTA copy changes ("Start free trial" vs "Talk to sales") for the same tier across visits.
You can also surface a test by clearing cookies and changing your country (a VPN to the US, EU, and an APAC city often reveals geo-bucketed price testing too).
What variants tell you about hypothesis
Every A/B test has a hypothesis embedded in it. If the variants are $39 vs $49, the test is "can we charge more?" — and the answer will arrive when one variant becomes permanent. If the variants are "Pro" vs "Team" as a tier name at the same price, the test is about positioning, not price. If the variants change the default toggle from monthly to annual, they're testing whether they can move customers into longer commitments without lowering revenue per month.
Read the axis of variation and you've read the question the company is asking itself. That tells you what they're uncertain about — which is information they'd much rather you didn't have.
Plans that disappear, reappear, get renamed
A different flavor of pricing experiment is the plan-shape test. A new tier appears for a few weeks and then vanishes. The "Hobby" plan becomes the "Starter" plan becomes the "Free" plan over the course of a quarter. A "Team" tier splits into two separate tiers ("Team" and "Business"). All of those are companies feeling out segmentation, and each iteration tells you which customer slice they've decided is most worth carving out separately.
If you can reconstruct the sequence (in flight, or after the fact via the Wayback Machine), you've reconstructed their pricing strategy without anyone telling you.
When the experiment "settles," that's a decision
The most valuable observation isn't the test itself — it's the moment the variants collapse into one. When everyone in your team starts seeing the same price again, the company has decided. The winning variant is now the company's commercial position, and it carries information: this is what they believe they can charge. That floor (or ceiling) is now public and you can plan around it.
If the winning variant is higher, the market just accepted a higher price. If it's lower, demand at the higher tier wasn't strong enough. Either way, you've watched a pricing strategy converge in public.
How Seeto handles this
Catching a pricing test requires visiting the same page repeatedly from different sessions, which nobody actually does on a schedule — it's the kind of work that gets described as a habit and quietly skipped. Seeto snapshots competitor pricing pages on its own cadence from neutral infrastructure, so variant differences and post-experiment settlement both surface as discrete change events. The experiment-in-progress shows up as a pricing surface that won't sit still; the resolution shows up as the moment it does.
The two-minute version
For each of your top three competitors, once a week:
- Open their pricing page in incognito. Note the prices, tier names, default toggle, and "Most Popular" badge.
- Compare to last week's note. Any difference is worth a second look — either you caught a test in flight or the company has just made a decision.