Hiding the price in a sale email wins the click. Does it win the order?

KEY TAKEAWAYS

1. Hiding prices in email buys clicks, not orders. Two field experiments agree on that, and one found fewer purchases.
2. The effect depends on who reads the email. Purchase-ready customers buy more when they see the price. Light subscribers may explore more without it.
3. An ESP test that picks its winner on click rate will usually pick the no-price version. Switch the winning metric to orders.
4. Once you do show a price or a discount, Gmail can display it, and EU and some US rules decide how it must be stated.

Black Friday falls on 27 November this year, and sale emails for it are being built now. One design choice gets made almost by habit: put the price on each product tile, or leave it off so readers have to click.

Leaving it off does win clicks. The better evidence says it does not win orders, and sometimes loses them. This piece sets out what the field experiments found, where they disagree with other studies, and how to decide product by product and subscriber by subscriber.

What happens when a promotional email hides the price?

Clicks go up. Purchases do not follow.

The most recent test is a working paper by Daniel Schwartz, Martín Navarrete and Marcel Goic, revised in May 2026 and under revise and resubmit at Management Science. They ran a field experiment across 15 campaigns and more than 2.4 million emails. Withholding prices raised click-through rate by 22.3%. Conversions did not rise, whether measured directly, across channels or over four months.

A second experiment, published in the Journal of Consumer Research, ran at Fravega, a mid-tier Argentine appliance and electronics retailer. In December 2021 it sent seven daily sale flyers to 771,583 customers, assigned at random. One group saw prices on each product thumbnail. The other saw the image and description only. Both groups landed on the same product page with the price.

The no-price group clicked slightly more products, 2.92 against 2.81 on average. Its weekly purchase rate was lower, 2.31% against 2.61%. That is about 11% fewer purchases, and raw revenue per customer about 15% lower (our arithmetic from the paper's means).

An earlier thesis at the University of Chile, supervised by Schwartz, found the same pattern in randomised retail emails. Hiding prices raised clicks and site browsing but lowered sales of the advertised products. A Listrak example from 2016 pointed the same way, with no sample size published. The no-price version drew about 40% more clicks, and adding prices doubled the conversion rate.

Why do the extra clicks not turn into orders?

Because the click now does a different job. With a price on the tile, the reader decides in the inbox. Those who click have already accepted the price. Without it, the click becomes a way to find out. Some of those visitors see the number and leave.

Schwartz and colleagues found two groups moving in opposite directions. Light users, who buy rarely, explored more when prices were hidden and bought more. Heavy users, who were ready to buy, did better when they could see the price and pick what they wanted. Overall, the two effects cancelled. A strategy that showed prices to purchase-ready customers, picked by past spend or recency, frequency and monetary scores, raised revenue by up to 8.6% over hiding prices from everyone.

They also found a cost that shows up later. Withholding prices repeatedly lowered engagement with later emails. One plausible reading, ours and untested: readers who click and find a price they do not want learn that the next click is not worth it.

Does hiding the price ever sell more?

Yes, and the same JCR paper shows when. Its authors argue that a delay lets shoppers fill the gap with their own guess. If the guess is higher than the real price, the reveal feels like a deal. If it is lower, the reveal disappoints.

On Fravega's website in 2020, during inflation above 40%, hiding prices raised the share of days with at least one sale from 80.4% to 94.7%. In lab studies, the delay helped for a premium product or retailer and hurt for a discount retailer or a discounted product. Explaining the delay with a minimum advertised price policy hurt as well, because it signals a low price.

A StubHub experiment in 2015 also found that concealment can pay. Showing ticket prices without fees until checkout raised purchase propensity about 14% and revenue per user about 21%. That was fee concealment on a website, not a base price left out of an email. In the US, the FTC's fees rule has required since 12 May 2025 that any price shown for live-event tickets or short-term lodging be the total price.

So the studies do not contradict each other. They describe one mechanism with two outcomes. A sale email is sent to people primed to expect a lower price, which is the case where hiding it hurts. The email results fit that.

Sales trainers make a separate case for leaving price out of high-ticket, relationship-led marketing, so a salesperson can frame it first. That argument rests on experience, not experiments. The field tests above are all retail. Quote-based B2B pricing sits outside the evidence.

Why does your own A/B test keep picking the no-price version?

Because most email tests are scored on clicks. Klaviyo lets you choose open rate, click rate or placed order rate as the winning metric, and suggests click rate for content tests. Mailchimp offers open rate, click rate or total revenue, and revenue only works with a connected store. A test scored on clicks will crown the version without prices nearly every time.

The test is also too small to see the order effect. The sum below shows why. Klaviyo labels a result significant when its win probability passes 90% and each variation has more than 50 deliveries. That bar suits a click test. It is thin for orders, where you may have a few hundred per version.

Gmail adds a further pull towards clicks. In September 2025 Google announced a relevance sort for the Promotions tab on mobile, favouring senders people engage with most. Google has not said which signals count. A curiosity click might help placement. Schwartz's finding that repeated withholding lowers later engagement suggests the gain does not last.

Before you trust any click-based result, also check how much of it is machines. Security scanners click links too, as covered in which funnel decisions click data still supports.

What does the trade look like on a real send?

A hypothetical example, with made-up figures.

A homewares store sends its Black Friday email to 120,000 subscribers. With prices shown, 3.0% click and 0.50% order: 3,600 clicks and 600 orders.

Without prices, apply the 22.3% click lift from the Schwartz study. Clicks rise to about 4,400. If orders stay flat, as in that study, the store still gets 600. If they fall by the 11% seen at Fravega, it gets about 534. Either way, a click-scored test declares the no-price version the winner.

Can the store measure the difference itself? To detect 0.50% against 0.445% with 80% power at the usual 5% level needs about 244,000 recipients per version (our arithmetic). The whole list is half of one arm. A single send cannot settle it. Pool the result across several sends, as the researchers did with 15 campaigns, or use the published evidence as your default.

Which emails should show the price?

If you do show prices, you can mark them up for Gmail. Google's Promotions tab documentation, updated 8 September 2026, accepts an optional price and discount on each product card. A price of 100 with a discount of 25 displays as $75. Google says annotations may not appear for every recipient.

How do you test price display without fooling yourself?

1. Split the list by purchase history first. Use recency, frequency and spend. Send prices to the purchase-ready segment without testing.
2. Randomise only within the remaining subscribers. Do not compare segments with each other.
3. Set the winning metric to placed order rate or revenue per recipient. Never click rate. Give buyers a longer window than a click test needs.
4. Size the test on orders. Work out recipients per version from your order rate. If one send is too small, repeat the same split for several sends and pool the results.
5. Hold the landing page constant. Both versions should land on the same product page with the same price.
6. Track the next three emails. Compare later clicks and orders between the two groups to catch the engagement cost.
7. Check the rules before the send. Base any EU discount on the 30-day lowest price. Show total prices where US fee rules apply.

The same principle runs through judging AI-written nurture emails on meetings, not opens. Score an email on the outcome you sell, not the step before it.

The short version

Leaving the price out of a sale email is a reliable way to raise clicks and an unreliable way to raise sales. Two large field experiments found no gain in orders, and one found fewer. Show prices to people ready to buy. If you want to test withholding, do it on light subscribers, score it on orders, and accept that one Black Friday send is too small to prove much.

Questions

Should you show prices in promotional emails?

Usually yes, at least to recent and frequent buyers. Two large field experiments found that hiding prices lifted clicks but not purchases. One, with 771,583 customers of an Argentine retailer, found purchase rates fell from 2.61% to 2.31% when prices were removed from the email.

Why does hiding the price increase clicks?

Without a number, readers who are curious have to click to find out. Some of those clicks come from people who would have skipped a price they disliked. The extra clicks are real, but many end on the product page once the price appears.

Is there any case where withholding the price sells more?

Yes. Research in the Journal of Consumer Research found that delay helps when shoppers expect a higher price than the real one, for example with premium brands or during high inflation. It hurts when shoppers expect a bargain, which is the usual mindset for a sale email.

How should I A/B test price display in email?

Pick placed order rate or revenue as the winning metric, not click rate. Split by purchase history. Expect to need far more recipients than a click test, often over 200,000 per version to detect a 10% change in a low order rate, so pool results across several sends.

Sources

1. SSRN, Schwartz, Navarrete and Goic, Selection versus exploration. How price information frictions decouple clicks from purchases in email marketing papers.ssrn.com
2. Martín Navarrete, Research martinnc.cl
3. Journal of Consumer Research, Affonso et al., Concealing prices. How delayed price disclosure influences consumer purchase decisions newswise.com
4. Iowa State University via Newswise, Researchers find when retailers wait to reveal prices, shoppers fill in the blanks newswise.com
5. Universidad de Chile, González Ham, Estudio de técnicas de economía conductual en email marketing repositorio.uchile.cl
6. Marketing Science, Blake, Moshary, Sweeney and Tadelis, Price salience and product choice faculty.haas.berkeley.edu
7. Listrak, Five email split-tests that really move the needle listrak.com
8. Google for Developers, Annotate emails in the Promotions tab developers.google.com
9. Google for Developers, Email promotions reference developers.google.com
10. Google, Introducing a new purchase tracking view and more relevant promotions in Gmail blog.google
11. Klaviyo Help Center, How to A/B test campaign emails help.klaviyo.com
12. Klaviyo Help Center, Review your email A/B test results help.klaviyo.com
13. Mailchimp, About A/B testing campaigns mailchimp.com
14. Court of Justice of the European Union, Press release 152/24, Aldi Süd, C-330/23 curia.europa.eu
15. Federal Trade Commission, Rule on unfair or deceptive fees. Frequently asked questions ftc.gov
16. Sandler, Leave price out of marketing go.sandler.com

NEXT STEP

See where your brand stands.

Your Digital Authority Score, who AI recommends for your target questions, and the placements that close each gap. Your first report is free.

Get your free report →