Guides · Conversion
Will my landing page convert? A checklist with benchmarks
Last updated November 11, 202611 checksRecords from 47 companiesBy Every Signal
You cannot know a landing page's conversion rate before traffic hits it, and no benchmark will tell you. What you can do before launch is check the page against what companies that convert actually do: how long their trials are, whether they take a card first, which plan they preselect, how they anchor the price, what their timers do and where they ask for an email. Every Signal records those fields for real companies, logged out and without paying, and tags each one.
Below is an 11-point checklist built from those records, each item with the figure behind it and a link to where it was read. Then how to run a free audit of your own page at /try/, and a plain account of what benchmarks can and cannot tell you.
The checklist
What do converting pages do, and how many of them do it?
Counts are over the 47 subscription, dating and software companies in the free trial and pricing guide (records dated October 19 to November 8, 2026), plus named examples from public reports. Tags: verified seen on the page, in the store or in code; estimate third-party or partial; inferred our reading.
| Check | What the records show | Tag | Source |
|---|---|---|---|
| 1. Ad and page say the same offer | FashionPass ads headlined $75 while the landing pages said $39 for the first month, then $125 | verified | fashionpass |
| 2. Price appears before the ask for an account | Calm's quiz shows the paywall after 36 to 43 screens and a first name, before any account is created | verified | calm |
| 3. Trial, and its length, is stated | 10 of 47 offer a free trial; 7 days is the most common length (6 of 10); 3 sell a paid trial for $1 to $4.99; 15 offer none; 19 do not say | verified count | free-trial-and-pricing |
| 4. Card up front is a choice, said plainly | 5 of the 10 free trials take a card first; both web checkouts that do show a $0.00 due todayline and promise a reminder (Superhuman 7 days before, Calm 2 days before) | verified | superhuman calm |
| 5. One plan is preselected on purpose | Of 9 records that capture a default, 4 preselect annual or longer (all work tools) and 5 preselect weekly, two-weekly or monthly (all entertainment) | verified count | free-trial-and-pricing |
| 6. The annual discount is real arithmetic | Annual costs a median 58% less than 12 months of monthly (n=15, range 10 to 75%); the struck listprice is usually 12 x monthly (Calm $203.88 = 12 x $16.99) | verified arithmetic | free-trial-and-pricing |
| 7. If there is a timer, it does something | 5 of 47 paywalls show a countdown; only Calm's changes the price at zero (monthly $13.49 back to $14.99); the others keep the price, and most are held in the browser | verified | calm moonly |
| 8. The email gate sits where it earns its keep | Calm and Moonly show the paywall before asking for email; myIQ puts the paywall after an email gate; Shortical shows an email sheet before the paywall; NetShort asks for no email at all | verified | myiq netshort |
| 9. The renewal and refund line is specific | myIQ states it in full: $1.00 trial for 7 days, then $29.99 every 28 days. 5 of 47 state a money-back guarantee (Truity 60 days, 16Personalities 30, Optery 30 on first purchase, Leave Me Alone 14, Moonly 30 with conditions) | verified | myiq truity |
| 10. The paywall is placed, not dropped | 28 of 47 records say where it sits: after a quiz, on the free result, at an episode (FlexTV 8, NetShort 6 to 9, ReelShort 9 or 15, Shortical 13 to 14), after sign-up, or on a usage limit | verified | free-trial-and-pricing |
| 11. The page can be measured | FashionPass fires 7 ad pixels, Nuuly 11; High Purity Labs fires none, so nothing it does can be attributed | verified | fashionpass high-purity-labs |
The 47-company dataset is data.json (CC BY 4.0), one object per company with every field and tag.
Reading the checklist
Which checks matter most before launch?
The offer on the ad and the offer on the page
The cheapest fix in the list. A visitor who clicked on $75
and lands on $39 then $125
has to re-decide, and the page carries the cost. FashionPass's record is the clearest public case fashionpass verified. Open your own ads and your page side by side; the Meta ads guide shows how to list the ads.
Trial length and the card
Free trials are rarer than the word suggests: 10 of 47. Where a length is stated, 7 days leads (6 of 10), with 3 days (Nebula, Hily), 14 days (Calm, yearly plan only) and 30 days (Superhuman) also present verified count. Superhuman and Calm take the card first and say Due today $0.00
; SaneBox says No credit card required
and takes it later in the dashboard sanebox verified. Neither is wrong. A card filters for intent; no card fills the top of the funnel. Decide which number you need and state the terms in the small print.
The default plan and the anchor
Work tools preselect annual (Superhuman at 17% off
, Clean Email's 5-account yearly plan, SaneBox's 2-year term); entertainment apps preselect the shortest plan (the four micro-drama apps, Optery's monthly toggle). That grouping is ours inferred; the defaults are the records' verified. If you show a struck price, make it something a customer can check: Calm's $203.88 and Clean Email's $119.88 are exactly 12 times the monthly price verified arithmetic. Shortical's struck $25.00 is its own renewal price, which a careful reader can find in the small print shortical verified.
Timers that do nothing
Of the five timers in the records, Calm's is the only one that changes the offer at zero: the annual plan disappears and monthly goes from $13.49 back to $14.99. Its end time is held in the browser, so a new browser shows the full discount again. Moonly's 92% discount reserved for: 10:00
kept the same price at 0:00, and Leave Me Alone's 12-hour timer restarts in each new browser verified. Whether any of this lifts conversion, the records cannot show. What they show is that a timer is a claim your customer can test in one reload.
Limits
What can benchmarks tell you, and what can they not?
They describe practice, not your rate
Every figure above is a count of what companies do, read from their pages. None is a conversion rate. The rate depends on the traffic, the offer, the device and the step measured, which is why one company shows different funnels to different visitors: Calm's Meta ads send 480 of 575 to a quiz with no trial and a quarterly default, while calm.com offers 14 days free on yearly calm verified.
The few rates that appear in Every Signal's public reports are context, each with its tag, and none is a target:
- Micro-drama apps: the Shortical report tests the company's profit claim against a
1-5% industry norm
for the share of new users who pay shortical estimate - Email tools: an old founder interview put SaneBox's organic trial-to-paid at 20-25% sanebox-vs-superhuman rumour
- A small online store: the High Purity Labs report models revenue from 15,000 visits at a 1.5% order rate and a $170 average order, as an assumption, not a measurement high-purity-labs inferred
Measured rates in the knowledge base carry the step, the n, the period and the traffic source, and a median is only computed over three or more verified values. The kb_benchmarks tool returns them that way through the API and MCP server. Use them as a range to sanity-check a plan, not as a promise.
Free audit
How do I check my own page against these records?
1. Paste the page copy at /try/
Open /try/ and paste the text of the page, between 40 and 3,000 characters: the headline, the offer, the price, the trial terms and the small print. No sign-up, two free audits a day; what you paste is kept for 30 days to run and improve the audit, so do not paste personal data.
2. Choose the type, sector and goal
Pick Landing page
as the type. Pick the closest sector or Not sure
, which lets the audit choose from the copy. State the goal if you have one, such as start a 7-day trial
.
3. Read the findings against the records
In about 40 seconds you get a verdict (ship, revise or rework), a score out of 100, and findings in severity order. Each finding names the knowledge base record it is judged against, with the evidence and a fix, and ends with a rewrite you can paste back to your AI or your editor.
4. Apply the rewrite and run it again
Fix the findings you agree with and run once more. For a live URL rather than pasted copy, the analyze_landing_page job on the API (100 credits, $10) adds the tech stack and pixels, funnel signals and a hero rewrite. The same audit as the free page is audit_work (30 credits, $3), and the first API key gets 50 free credits.
Questions
FAQ
What is a good landing page conversion rate?
There is no single number, and anyone quoting one without the traffic source, the offer and the step being measured is guessing. The public figures in Every Signal's reports are context, not targets: a 1-5% industry norm for new users paying in micro-drama apps (estimate), a founder's claim of 20-25% organic trial-to-paid at SaneBox (rumour), and a 1.5% visit-to-order assumption used to model a small store (inferred).
Should I ask for a card before the free trial?
Both choices are common. Of the 10 companies with a free trial in Every Signal's 47-company records, 5 take a card first and show a $0.00 due today line with a reminder promise, 1 says no card is needed, and 4 are unclear. Card up front filters for intent and raises trial-to-paid; no card raises trial starts. Pick based on which number you need.
Do countdown timers help a landing page convert?
The records cannot say whether they help, only how they are used. Five of 47 paywalls show a timer, and only one, Calm's quiz paywall, changes the price when it reaches zero. The others keep the same price, and timers are usually held in the browser, so a new browser resets them. A timer that does not change anything is a claim your customer can test.
Can Every Signal tell me my page's conversion rate?
No. Nobody can before traffic hits the page. The free audit at everysignal.com/try/ checks the page against what real companies do and against the measured rates in the knowledge base, and returns a score, findings with the record behind each one and a rewrite. Your own analytics give you the rate.
Is the data behind these benchmarks available?
Yes. The 47-company table is published in the free trial and pricing guide with a data.json file under CC BY 4.0, one object per company with every field and its tag. The records themselves are available through the Every Signal API and MCP server.
Check your page
Audit your landing page against the records, free
Paste the page into /try/ for a score, findings with the record behind each one and a rewrite in about 40 seconds (no sign-up, two a day). The full table of 47 companies is in the free trial and pricing guide; API and teardown prices are on /pricing/.