Before you start
Set the test up like this and the evidence you need will exist when you are finished. Set it up any other way and your ad platform will not report a separate cost per acquisition for each page, which means neither of us can tell what happened.
- Duplicate the ad set that is already running to your control. Change only the destination URL, to your Landra page.
- Run both ad sets in the same campaign, at the same time, from the same start date.
- Keep the same conversion event and the same attribution setting on both.
- Let both run until eachhas spent at least 5× your control’s cost per acquisition.
If you send both URLs from a single ad set, or use dynamic destinations, your platform will report one blended number and there is nothing to compare. Ask us first if you are unsure — we would much rather help you set it up than turn down a claim afterwards.
The promise
Run a real test. Take a page you built with Landra and run the same paid traffic to it and to your current landing page at the same time, spending at least five times your control’s cost per acquisitionon each. If the Landra page comes out more expensive per acquisition than your control, show us, and we refund your first month’s subscription.
A worked example.Your control converts at a $50 CPA. Five times that is $250. So you spend $250 driving traffic to the Landra page and at least $250 driving the same traffic to your control, over the same dates. If the Landra page’s CPA lands above your control’s, the refund is yours.
This is a performance promise, not a general refund policy. It is offered by Pow Organics LLC, doing business as Landra, and the Terms of Service— including their governing law and dispute resolution provisions — apply to it.
Who can claim
- You are on a paid Pro or Pro Plus subscription and have been charged at least once. Free trials are not covered, because nothing has been paid yet.
- You were already running paid trafficto an existing landing page before the test — see what counts as a control.
- The claim covers your first paid month only. One claim per account, per business, and per payment method.
What the words mean
Four words decide every claim, so here is exactly what each one means.
- Control— the landing page or product page you were already sending that paid traffic to before the test.
- Acquisition — a purchase, as your ad platform reports it. If your business runs on leads rather than purchases, tell us before you start and we will agree the event in writing.
- CPA— the ad platform’s reported spend for that page’s ad set, divided by the acquisitions it reports for that ad set, over the test window. We read the number your ad platform reports, not one reconstructed in a spreadsheet afterwards.
- Test window— the start and end dates over which both ad sets ran. The same dates for both.
What counts as a control
To be a usable control, the page must:
- have been live and receiving paid traffic for at least 14 days before the test began;
- have recorded performance in your ad platform for that period; and
- sell the same product at the same price as the Landra page you test against it.
A page with no prior traffic is not a control — there is nothing to beat. If you have never run paid traffic before, this guarantee does not apply to you, and we would rather say so up front than take your money and argue about it later.
If your control recorded no acquisitions in the 14 days before the test, it has no CPA and there is no threshold to compute. Talk to us before you start and we will agree a spend figure in writing.
What counts as a fair test
- Both pages, at the same time.The Landra page and your control run together, over the same start and end dates. Sequential tests are not eligible — a page run this week against a page run last week compares two weeks as much as it compares two pages, and neither of us can tell which.
- At least 5× your control’s CPA, on each page. If the control converts at $50, that is $250 to the Landra page and at least $250 to the control, inside the window. Below that the result is noise, and we would rather you kept the test running than took a payout on a coin flip.
- Same traffic. Same platform, same campaign objective, same audience or targeting, and the same ads. Duplicate the ad set and change only the destination URL.
- Same product and price on both pages.
- Same measurement. The same conversion event, attribution setting and attribution window on both ad sets, unchanged for the whole test.
On sizing the test.Use your control’s CPA from the 14 days before the test to decide what to spend — that is the number the 5× threshold is measured against. CPAs move, so if yours drifts up during the test, keep the test running until your spend clears the new figure. We would rather you had a real read than lost a claim on a technicality.
Beating it, or not
The metric is cost per acquisition, and the number to beat is your control’s CPA over the same test window.
If the Landra page’s CPA comes out higherthan your control’s, you can claim. If it comes out equal or lower, the page did its job — you keep the subscription, and you now have a second page that performs at least as well as the one you had.
If the Landra page recorded no acquisitions at all at the required spend, that counts as worse than your control. You can claim.
CPA is the number this guarantee is written against, because it is the one every ad platform reports the same way. If you steer on something else internally — ROAS, contribution margin, subscription rate — send it anyway; we genuinely want to see it, and it may change what we build you next. It is not a separate route to a refund.
What we need to see
Screenshots from your ad platform are the minimum, and usually enough. For the same date range, show us:
- Spend, acquisitions and CPA for the Landra page— enough to see you spent at least 5× your control’s CPA;
- Spend, acquisitions and CPA for the control, over those same dates — this is what proves your control CPA and that both pages carried comparable spend;
- Your control’s CPA for the 14 days before the test, which is what the 5× threshold is measured against;
- the URL of each page; and
- the campaign and ad set names, so we can see the traffic was the same.
One export covering both ad sets is ideal, since it shows the dates line up. If something does not add up we may ask for a CSV export or a second screenshot with the date range visible. We will never ask for access to your ad account, and we will never ask for your login.
We keep what you send only as long as it takes to decide the claim and to keep a record that we paid it, and we use it for nothing else.
When to claim
Start your test within 30 days of your first charge. Claim within 30 days of the day your test ended, and in any case no later than 90 days after your first charge. Your first charge is the one on your billing page dated soonest after your trial ended.
What you get back
The full amount actually charged on your first invoice after any discount, in the currency you were charged, refunded to that same payment method within 10 business days of us approving the claim. Later months are not included. There is no cash equivalent and no credit alternative.
Cancelling is your call, not a condition. Tell us and we will cancel the subscription for you, effective immediately, with no retention conversation. Say nothing and it carries on exactly as it was. Taking the refund does not end your plan, and keeping your plan does not cost you the refund. Either way you keep every page you have built and every bonus from the First Winner Kit.
What is not covered
- Tests below the 5× threshold on either page, or where the two pages carried very different spend.
- Sequential tests — the two pages have to run over the same dates.
- Tests where the product or the price differed between the two pages. How the offer is presentedis expected to differ — rewriting that is the whole point of the Landra page.
- Tests where the ads differed. Duplicate the ad set and change only the destination URL; if you change the ad, you are testing the ad, not the page.
- Organic, email, affiliate, or influencer traffic — this is written for paid.
- Periods where tracking was broken, attribution settings changed mid-test, or the ad account was restricted.
- Accounts suspended for a Terms of Service or advertising-policy violation.
- Second and subsequent claims on the same account, business, or payment method.
How we decide
A person reads your claim — there is no form that auto-rejects you. We decide against the conditions on this page, not against how we feel about the result. If the numbers meet them, we pay. We will acknowledge your claim within 2 business days and give you a decision within 10 business days, or tell you what is still missing.
We may ask follow-up questions, and if a test was close but under-spent we will usually suggest you keep running it rather than close the claim. If you think we got it wrong, reply once and a second person will look; that review is final.
Evidence that has been altered, or figures that contradict what your ad platform reports, void the claim and are treated as a Terms of Service matter.
The small print
This guarantee is personal to your account, is not transferable, and is void where prohibited. It does not limit any rights you have under law. If any part of it is unenforceable the rest still stands, and not enforcing a condition once does not waive it.
We can change or withdraw this guarantee for future customers at any time. If we do, it does not affect a test you have already started — the terms that applied when you began are the ones we will judge your claim against, and we will not apply a later, stricter version to it. The effective date above tells you which version you are reading.
How to claim
Email support@getlandra.comwith the subject line “Guarantee claim” and the evidence listed above. You can also start it from the support chat in the app.