The Poor Man's A/B Test: How to Split-Test Your Price Without Any Special Software

Somewhere out there is a version of your product priced at $27 that converts twice as well as the $37 version you're currently selling — or maybe $37 actually feels more credible. You don't know, because you've never tested it. Big companies solve this with Optimizely, VWO, or a data team running multivariate experiments on thousands of visitors an hour. You have neither the traffic nor the budget for that, and you don't need it. What you need is two duplicate product pages, two differently-tagged links, and a plain spreadsheet. This is the split test creators have run quietly for years before 'growth experimentation' became a job title — and it works just as well on a few hundred visitors as a few hundred thousand.

Why a Homemade Test Actually Holds Up

Professional A/B testing tools mainly solve one problem you don't have: showing two different experiences to the same pool of anonymous visitors in real time. Your situation is simpler. You have identifiable channels — an Instagram bio, a TikTok bio, a newsletter, a handful of Stories — and you can consciously decide which price each channel sees. That's not a weaker version of A/B testing, it's a different (and arguably more controllable) flavor called a split-run test, exactly what direct-response marketers used for decades before software existed. The trick is discipline: one variable changes at a time (price), everything else stays identical, and you record results somewhere other than your memory.

Step 1: Build Two Identical Offers

Duplicate your product listing so you have two pages that are pixel-identical except for price. If you sell a $37 template pack, clone it and set the copy to $27. Same title, same cover image, same description, same delivery method — the only difference visible to a buyer is the number next to the checkout button. On store.fan this takes about ninety seconds: duplicate the product, rename it internally ('Template Pack — Price B') so you don't confuse it in your dashboard, and adjust the price field.

1. 1Duplicate your existing product listing so the description, images, and delivery are word-for-word identical.
2. 2Set Price A on the original page and Price B on the clone — pick a meaningful gap, like $27 vs. $37, not $27 vs. $29.
3. 3Give each one a distinct, memorable internal name so you never mix them up mid-test.
4. 4Grab the direct link for each product page — this is what makes the whole test possible.

Step 2: Send Different Traffic to Each Link

The single most important rule: never show two prices to the same person. If a follower sees $27 in your Instagram bio and later $37 in your email, you haven't run a test — you've confused a customer and torched their trust. Instead, assign each price to a channel or time window and keep it there for the whole test. Here's a placement map that works for most solo creators juggling two or three platforms.

If you only have one bio link, the day-alternating method above is your friend: odd calendar days point to Price A, even days to Price B, for two full weeks. It's not as clean as a true split, but it averages out day-of-week effects far better than running Price A for one week and Price B for the next.

Step 3: Track It Like a Scientist (Sort Of)

You don't need attribution software to know which link drove which sale — every sale on Price A's page is a Price A sale by definition. Check sales by product in your dashboard, not just total revenue, and build a simple sheet with one row per day and these columns:

• Date and which price was live on which channel that day
• Clicks to each link (your bio-link platform or story insights usually show this)
• Checkout starts (visits to the product page, if you track that separately)
• Completed sales for each product
• Revenue for each product — not just count, since a lower price needs more sales to break even on revenue

Reading Your Results Without Fooling Yourself

Run the test for at least ten to fourteen days, or until each link crosses roughly 100 clicks — whichever takes longer. Anything shorter and you're one lucky Tuesday from a false conclusion. When the window closes, calculate revenue-per-click for each price (total revenue divided by total clicks). That single number beats raw sale counts, because it accounts for conversion rate and price at once. If it's within about 10% between the two, treat it as a tie and let other factors — perceived value, positioning, how the number feels to say out loud — break it. If one price wins by 20% or more, that's your new default; check more guides on pricing psychology before your next test.

Mistakes That Quietly Ruin the Test

• Changing anything else mid-test — a new cover photo or edited description invalidates your comparison instantly.
• Announcing a 'sale' or discount code on only one price variant, which stacks a second variable on top of price.
• Ending the test the moment one price pulls ahead by a few sales — early leads flip constantly in small samples.
• Forgetting to check your delivery works identically on both — a broken download link on one page will tank its numbers for reasons that have nothing to do with price.
• Testing during an atypical week (launch week, holiday, viral moment) and assuming the result is your steady-state baseline.

The cheapest, most reliable pricing research tool a solo creator has isn't a survey — it's two live links and two weeks of honest tracking.

Once you've got a winning price, this becomes a repeatable habit rather than a one-off project. Test your ebook price this month, your coaching call next month, your membership tier after that. Every store.fan product page comes with its own checkout, instant delivery, and sales record in your dashboard, so the infrastructure is already sitting there — you're just pointing two links at it deliberately instead of guessing once and hoping.