Should you raise prices before Black Friday?

• Don't raise prices now so that your Black Friday discount looks bigger. In the EU, a price reduction has to be measured from your lowest price in the 30 days before it. In the US, the FTC's pricing guides call a price set up only to advertise a reduction from it a false bargain.
• If your prices are too low for your costs, raise them for real, and keep the new price long after the sale. Ideally do it in a normal month, not five weeks before Black Friday.
• Use a margin calculation, not a test, to judge a price change at small-store traffic. You can work out how many sales you could lose and still earn the same.
• Leave prices alone from Monday 16 November (the freeze) until the sale ends.
• This is information, not legal advice. Rules differ by country. If you sell into the EU or have a specific question, check with an adviser.

The wider calendar is in our general Black Friday testing guide.

The idea is familiar: a $40 product goes up to $50 in early November, then shows "was $50, now $40" on Black Friday. Shoppers see 20% off. You lose nothing. It's also the specific thing price-reference rules exist to stop.

The EU: lowest price in the last 30 days

Directive (EU) 2019/2161 added Article 6a to the Price Indication Directive (98/6/EC). Any announcement of a price reduction has to show the prior price, and the prior price is the lowest price the trader applied in a period of at least 30 days before the reduction. Member states can set some exceptions (for example perishable goods or products on sale for less than 30 days), so the details vary by country.

For a reduction that starts on Friday 27 November, the window runs from Wednesday 28 October to Thursday 26 November. If the product sold for $40 on any day in that window, $40 is the price your reduction has to be measured from. Raising it to $50 on 1 November doesn't change that: your "now $40" would be a 0% reduction.

The US: a former price has to be real

The FTC's guides against deceptive pricing (16 CFR 233.1) say a former price is a legitimate basis for a comparison when it is "the actual, bona fide price at which the article was offered to the public on a regular basis for a reasonably substantial period of time." They give the inflated-then-reduced price as the example of a false bargain. Some states have their own rules on top. A price that lasted three weeks before Black Friday and that nobody paid is a weak "was" price.

Your customers notice

Regulation aside, your repeat customers remember what they paid. A small store lives on them. A "sale" price that matches last month's normal price is the kind of thing people screenshot.

Some small stores do need higher prices: shipping, packaging and product costs rose and margins got thin. That's a real reason to change prices. Three rules keep it honest and useful:

1. 1.Make it permanent. Set the new price because it's your price, and keep it through December and January.
2. 2.Set it early, or after the sale. Changing it on 1 November leaves your Black Friday discount measured from your old, lower price in the EU. If you want the new price to be the base of your sale, change it at least 30 days before the sale starts, which for most stores means now or not this year.
3. 3.Remove stale compare-at prices. In Shopify the compare-at price is the crossed-out price on product pages. Shopify shows a sale only when the compare-at price is higher than the price, and only the sale price appears at checkout. Check every product with a compare-at price and ask: did we actually sell at this price, recently, for a meaningful time?

Judge it on margin, not on a test

At small-store traffic a test can't detect the change in orders a modest price change causes. Margin arithmetic can tell you how much room you have.

The arithmetic is the useful part. "Would I lose more than one sale in seven at $44?" is a question you can answer from what customers tell you, what similar products cost and what your sales did the last time you changed a price. It isn't proof, but it's a better basis than a test that can't finish. The price elasticity entry in our glossary explains the idea behind it.

Price tests can work for a store with steady traffic, run in a normal month. Before Black Friday they're a poor fit for a small store, for three reasons:

• The timing. A test started now would run straight into the EU 30-day window and, at small-store traffic, into the sale.
• The sample. As the example shows, a test of a modest price change needs tens of thousands of visitors per version.
• The season. Sale-week shoppers react to prices differently from your normal customers. A result from November describes November.

If you advertise your products in Google Shopping, there's one more constraint: Google Merchant Center asks that the price in your product data match the price on the landing page and at checkout, and it can disapprove products that don't. A price test shows some shoppers a different price, so plan any future price test with that in mind.

What to do instead before the sale:

• Decide your discount with break-even orders (part 4 of this series).
• Compare offer formats at the same cost (a gift against a dollar amount) in January, not depths in November.
• Read test prices or discounts first? for the trade-off once your traffic allows a test.

• Raising prices in November and calling the old price a sale. Under the EU rule it isn't a reduction; under US guidance it's a fictitious former price.
• Leaving last year's compare-at prices on products. An old crossed-out price can be as misleading as a new invented one.
• Changing prices during the freeze. A price change on 20 November also changes what every banner, email and discount promises.
• Dropping the "new" price right after Cyber Monday. That shows it was never the regular price.
• Judging a price change on revenue for one week. Look at what each sale earned, over a normal month, before and after.

• [ ] Any real price change made by 1 November, and kept after the sale
• [ ] No price raised just to make the sale discount look bigger
• [ ] Compare-at prices checked: each one is a price you actually sold at
• [ ] Margin per sale worked out for the products you discount
• [ ] For EU shoppers: the prior price for each reduced product noted from 28 October
• [ ] Prices left alone from 16 November until the sale ends

Advising clients on this? See the client price-test risk checklist. For a larger brand's view, see price tests and BFCM.