Should we use price anchoring, compare-at prices or bundles?
Our answer
Yes, with real anchors only. Show a compare-at price only for a price you really charged, show bundle savings against the sum of the parts and test each against no anchor on profit per visitor.
- Confidence
- Medium
- Strongest proof
- 1 controlled test
- 6 official rules, 2 datasets, 10 practitioners, 3 opinions or claims
- Sources
- 18 sources
- 22 statements
MO
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentA former price must be real, and the rules differ by place. The FTC allows one only if it was offered regularly for a reasonably substantial period. California requires the three months before the ad. EU rules use the lowest of the past 30 days.
Percentage badges move shoppers more than a strikethrough does, and percentages from an inflated reference price mislead them most. A European Commission experiment with 4,015 participants tested both.
Bundles can raise order value and still cost you profit. One brand saw order value rise 22% on a collection page of bundles (Marketing Operators podcast). Eightx (fractional CFO firm) and Intelligems (price testing software maker) warn that deep bundle discounts can cancel the profit.
Our judgment
- Follow the strictest former-price rule among the places you sell. The FTC, California, the EU rules and Amazon's and Google's checks each define a real former price differently, so one product can qualify on one surface and not another (PPC Land).
- Base any percentage you show on the genuine prior price. Use it rather than a recommended or list price, given the European Commission's finding on percentage badges and inflated reference prices.
- Judge each test on profit per visitor, not order value. That follows Fudge's advice to judge order value and conversion together and Eightx's scorecard of contribution per order.
- Keep bundle discounts shallow enough that a bundle order earns more contribution than a single item. In Eightx's model a 35% discount left a three-item bundle no better than one item, and Intelligems warns about deep bundle discounts. Marketing Operators podcast, Andrew Faris and HexClad show bundles can lift order value.
Our judgment, built from US Federal Trade Commission, California Legislative Information (California Law), Competition and Consumer Protection Commission (CCPC), Amazon, Google, PPC Land, European Commission, Fudge, Eightx, Intelligems, Marketing Operators podcast, Andrew Faris and HexClad: our reading of these sources, not a finding from any one of them.
What the sources say
22 statements from 18 sourcesIreland's regulator, applying EU price rules, says a sale's reference price must be the lowest price charged in the 30 days before the cut; successive cuts in one campaign can use the price before the first cut.
“lowest price […] applied to a product no less than 30 calendar days before applying a price reduction”
Google shows a struck-through sale price in US Shopping ads only if the discount is over 5% and under 90% and the base price was charged 5 of the last 30 days or 15 of the last 200.
“Sale price annotations aren't always guaranteed to show, even if you meet the requirements above.”
Shopify says a sale price shows when the compare-at price is set above the price; themes decide where it displays, and crossed-out savings from discounts need a third-party app.
“Your theme controls whether sale prices are displayed on your storefront, and where they're displayed.”
A California federal judge let a class suit proceed against Brooks Brothers over a struck-through $89.50 price allegedly not charged in the prior three months. The order does not rule on the price itself.
“Plaintiff plausibly alleges all claims under all three causes of action.”
Fudge, which sells a Shopify theme tool, says compare-at prices must be real past prices (the EU's lowest in 30 days), bundles should show both totals, and anchoring tests must track conversion as well as AOV.
“A change that raises AOV while lowering conversion can reduce total revenue.”
US FTC pricing guides say a former price must be a genuine price offered regularly for a reasonably substantial time; if it is fictitious, the sale price is probably just the regular price.
“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”
California law bars advertising a former price unless it was the prevailing market price within the three months before the ad, or the date it applied is clearly stated in the ad.
“unless the alleged former price was the prevailing market price as above defined within three months next immediately preceding the publication of the advertisement”
Two European Commission experiments with 4,015 people on a simulated shop found percentage discounts drive choices, strikethrough prices barely matter, and inflated reference prices led many to pick the wrong deal.
“The presence of a percentage discount strongly influences selection, even when it does not reflect the largest actual price reduction.”
HexClad's Connor Rolain says selling complementary products together in the first purchase gets much higher take rates than selling them separately.
“we'll sell them on both products out of the gate”
CTC's Joy says Black Friday discounts below 20% don't move shoppers: keep offers simple (20%+ sitewide), raise order value, add a high-priced bundle, and turn off first-order discount pop-ups.
“Anything below 20%, people don't care.”
On Scalability School, Brad Ploch reports an Intelligems split test on a $279 kitchen appliance: a $199 sale price beat the usual $229 by 44% revenue per visitor, so $199 became the standing price. No sample given.
“cutting a $279 product's sale price from $229 to $199 lifted revenue per visitor 44%”
Growth Suite, which sells discount pop-up software, says show the real original price with a clear reason for the discount, and never inflate it. Permanent 30% off makes the struck-through price meaningless.
“If everything is permanently 30% off, the struck-through price stops meaning anything.”
On Scalability School, Brian Ng says a losing supplement brand moved to a subscription-first page with a $74.95 list price at 50% off; revenue per user rose 15 to 18% and monthly profit went from about -$24K to +$10K.
“We raised MS, MSRP to like $74.95 and then we gave it a 50% discount.”
PPC Land says reference price rules differ: the FTC uses an undefined period, Google 30 days in 200 or 5 in 30, Amazon a 90-day median and the EU the lowest price over 30 days.
“One product can therefore show a legitimate discount on one surface and none on another.”
Neil Patel says that in the data his team has seen, brands that lean on discounts get lower customer lifetime value than those that use bundles to win the first purchase.
“It actually reduces your LTV from the data we've seen.”
Andrew Faris says a targeted product bundle beats the brand's other products on average order value by 10 to 20%.
“that bundle is crushing other products we have in terms of AOV to the tune of 10 to 20%.”
On Marketing Operators, a brand reports that sending traffic to a bundles collection page instead of a single product page raised AOV 22% and revenue per user 27%, with conversion flat.
“we're seeing CVRs flat. It's actually up slightly on the collection page variant, but AOV is up 22%, revenue per users up 27%.”
A lawsuit, with no ruling yet, alleges iRobot showed a Roomba as discounted from the same $449.99 reference price from June 2025 on without ever selling it at that price, in breach of California law.
“run perpetually, so the sale price is really the regular price”
Eightx, a fractional CFO firm, says bundles lift AOV about 15 to 30% for consumer brands but should be judged on contribution per order, with discounts of 10 to 20%, or 5 to 10% below 50% margin.
“The right metric is contribution dollars per order.”
Eightx, a fractional CFO firm summarizing older pricing studies, says .99 endings lift unit sales about 24% on average, but premium brands should keep round prices on hero products. Treat the figures as priors.
“Odd endings signal value and deal; round prices signal quality and conviction.”
From April 23, 2026, Amazon validates a List Price only if it was offered at another retailer recently or bought as the Featured Offer on Amazon. Typical Price is the 90-day median non-promotional price.
“Typical Price is the median non-promotional price customers paid for a product over the last 90 days.”
Intelligems says data from over 1,000 price tests shows a 5% price change typically moves conversion about 10%, so whether a price change raises profit depends on margins.
“price changes create roughly 2x the conversion impact (a 5% price change typically drives a 10% conversion change)”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer