1 / 7
Tap any image to enlarge, then swipe or use the arrows to page through every screen.
Sales attributed to advertising are not always sales caused by advertising. Advertising Profit & Experiment Lab helps you separate campaign economics from the evidence that a campaign made a difference.
Enter the real margin after fulfilment, fees and refunds, then choose a supported comparison design. Review profit per participant, conversion and uncertainty. The tool checks whether your data includes non-buyers and whether the test groups were formed as planned.
Compare budget scenarios using assumptions you can see and change. When the data does not support a conclusion, the result says so. Observational comparisons are labelled separately from randomised experiments.
What you get
- Contribution model: real cash margin per order after product cost, delivery, fees, handling and refunds — the value of your time shown separately.
- Design checks: randomised customer-level tests, randomised region/time-cluster tests analysed per cluster, and observational comparisons labelled “Association only — not a causal estimate.”
- Sample planning: participants needed per group for your minimum useful effect at 5 % significance and 80 % power.
- Uncertainty intervals: Welch interval for contribution per participant, Wilson and Newcombe intervals for conversion, Holm correction for several variants, sample-ratio mismatch check.
- Budget scenarios: spend, expected participants, conversion, margin and capacity — no response curve invented from two points.
- Reports: printable result, profit model and scenario reports (A4 or US Letter), CSV exports and complete backups.
How it works
- Define profit. Enter price, product cost, delivery, fees, handling and refunds. Your time is valued separately, only if you choose to.
- Choose a design. Random split by customer, random split by region or period, or an observational comparison — with the main metric and end date fixed in advance.
- Import participants and review validity. Import one row per participant, including people who did not buy. The tool checks the split, missing non-buyers and sample size.
- Compare economics. See incremental profit with its uncertainty range and an honest status, then export a report and keep a backup.
Limits
No campaign execution; no automatic attribution; insufficient traffic can prevent a reliable conclusion. Significance is evaluated once, after the planned end date — this version has no sequential method for stopping a test early. Budget scenarios are your assumptions, not forecasts.
Questions
Does it use AI?
No. The tool uses explicit calculations and rules. Results depend on the records and assumptions you enter.
Do I need a subscription?
No subscription is required for the one-time purchase of this product. Access through TidyCents PRO follows the current PRO terms.
Where is my data stored?
Working records are stored on your device. Keep a backup, especially before changing browsers, clearing browser data or moving to another device.
Can several people edit the same workspace at once?
This version is designed for one local workspace. It does not provide simultaneous shared editing.
Can I move my records to another device?
Yes. Export a full backup and restore it in the same product on the other device.
Does it connect to my bank or accounting software?
This version works with direct entry and supported file imports. It does not automatically connect to your bank or accounting system.
Can I try it before buying?
Yes. The demo contains fictional records and prepared scenarios so you can inspect the screens and results before buying.
Does it run or change my ad campaigns?
No. It does not connect to ad platforms and does not start, pause or change campaigns. You run the test in your own tools and import the results.
Can it prove that my ads caused the sales?
Only a randomised comparison with a control group can estimate a causal effect, and even then within an uncertainty range. Before/after comparisons are labelled “Association only — not a causal estimate.”
What if my traffic is small?
The tool shows how many participants your test needs. With too little data the result says “Insufficient data” or “Inconclusive” instead of declaring a winner.
Business license for use in your own business. No resale or redistribution. Opens in your browser right after payment, on phone and computer. Your figures stay on your device. No physical item.
Advertising Profit & Experiment Lab
Check whether a campaign adds profit, with costs, refunds and experiment uncertainty shown beside the result.
See what a campaign really adds — buy once for $24.
- One-time purchase
- Full version of this tool on the site plus an offline version to download. One payment. The offline copy works without a subscription.
- Access with PRO
- Full version on the site while your PRO is active. Export of your results and backups is included. The offline program is bought separately.
Pay by card, Apple Pay or Google Pay
- Works in any country · your own currency
- Download link on the confirmation page and by email
- What you need: Any modern browser · phone and computer · offline HTML file · CSV & printable reports · 8 languages
- Refund within 14 days, for any purchase
- Formats: Interactive tool in your browser — phone and computer; PDF & CSV export
- Interactive tool · modern browser · phone and computer · downloadable offline version.
- Your working records stay on your device. Export a backup to move them to another device.
- Use your own currency and enter the rates, costs and contract terms that apply to your business.
- Explicit calculations: exact decimal money, published statistical formulas, formula version shown with every result.
- CSV and XLSX import of participant rows or cluster totals; CSV exports and complete backups.
- 8 languages: English, Español, Deutsch, Français, Português, Italiano, Nederlands, Русский.
Any purchase is refunded if you ask within 14 days of payment. Files that will not open, and anything that does not match its description, are refunded in full at any time, with no time limit.
