Quick answer
In StubHub’s randomized fee-presentation experiment, shoppers seeing fees later selected tickets 4.40 percentage points more often at the first measured step. Among shoppers who reached checkout, their completion rate was 14.89 points lower. A recovery reply should state the shopper’s total and show which fees are already included.
Blake, Moshary, Sweeney and Tadelis, Price Salience and Product Choice, Marketing Science
Key findings and sources
Later fee disclosure had a 4.40-point higher selection rate
In the 2015 StubHub experiment, 27.96% of event-page arrivals selected a ticket when fees appeared later, compared with 23.56% when shown upfront. Said Maybe’s comparison gives a 4.40-point difference.
Blake, Moshary, Sweeney and Tadelis, Price Salience and Product Choice, Marketing Science
U.S. StubHub web visitors assigned by cookie in the August 2015 fee presentation experiment Subtract the upfront rate from the later-fee ticket details transition rate.
Later fee disclosure had a 14.89-point lower final-stage completion rate
Among StubHub visitors who reached the final checkout stage, 18.52% completed with later fee disclosure and 33.41% completed with upfront disclosure. This was a 14.89-percentage-point gap.
Blake, Moshary, Sweeney and Tadelis, Price Salience and Product Choice, Marketing Science
U.S. StubHub web visitors assigned by cookie in the August 2015 fee presentation experiment Compare checkout-to-purchase rates among people who reached the preceding stage.
The stage-level comparison changed direction by 19.29 points
The later-fee minus upfront gap moved from plus 4.40 points for ticket selection to minus 14.89 points for checkout completion. The 19.29-point shift compares different stage-arrival populations.
Blake, Moshary, Sweeney and Tadelis, Price Salience and Product Choice, Marketing Science
U.S. StubHub web visitors assigned by cookie in the August 2015 fee presentation experiment Subtract the negative final-stage gap from the positive early-stage gap.
Switching the denominator yields 17.11% lower 10-day revenue for upfront fees
Blake and colleagues reported 20.64% more 10-day revenue for later fee disclosure among assigned visitors. Expressed relative to the later-fee group, Said Maybe calculates 17.11% less revenue for the upfront-fee group.
Blake, Moshary, Sweeney and Tadelis, Price Salience and Product Choice, Marketing Science
U.S. StubHub web visitors assigned by cookie in the August 2015 fee presentation experiment Set upfront to 100 and later to 120.64. Divide upfront by later and subtract one.
Compare the results
The better-performing fee display switched by checkout stage
The two fee groups traded places between selecting a ticket and completing checkout.
A checkout percentage can fall while total sales rise
The group that initially saw ticket prices without fees selected tickets at 27.96% versus 23.56% when the full price was shown upfront. Among people who reached the final checkout stage, the rates were 18.52% and 33.41%. Expressed per 1,000 stage arrivals, that is 44 more initial selections and about 149 fewer final-stage completions for the later-fee group, on separate denominators.
Those two changes cannot be subtracted to estimate an order effect. The pool at checkout already reflects who advanced earlier. Table 3’s overall results use the randomized assignment comparison and show 14.10% greater probability of purchase and 20.64% more 10-day revenue for the later-fee group in this marketplace. The result is historical and the source does not provide a DTC revenue forecast.
A recovery reply comes after the shopper has chosen something
When the conversation begins at an abandoned checkout, the shopper may already have seen a lower number on a product page, a discount in an earlier visit, or a total before entering a destination. Verify which price they last saw and which fees actually apply. Quote the current out-of-pocket total if the store has enough information, then explain what changed in plain language.
If shipping depends on location, say when the final amount will be calculated. If the store can resolve it without collecting more information, include the actual fee in the reply. A buyer comparing a $65 item with a $72 checkout total has a different question from someone comparing two discount options.
Measure the final outcome on an eligible-shopper denominator
Track the last price the shopper saw, the actual checkout total, the stage of the first fee disclosure and the final completed order. Preserve the original event and refresh the amount immediately before sending a reply or returning the shopper to a cart link. A coupon, destination change or expired offer can alter the same cart’s total.
Evaluate a reply that explains the total against a comparable holdout. Record paid orders, refunds and margin across all eligible assignments. Then use conditional stage reports to diagnose where the shopper stopped. A lift in cart return clicks alone can miss a second stop when the final total appears again.
Study data
| Measure | Value | Context |
|---|---|---|
| Ticket selection, fees shown at checkout | 27.96% | Published experimental estimate or descriptive stage rate |
| Ticket selection, fees shown upfront | 23.56% | Published experimental estimate or descriptive stage rate |
| Checkout completion, fees shown at checkout | 18.52% | Published experimental estimate or descriptive stage rate |
| Checkout completion, fees shown upfront | 33.41% | Published experimental estimate or descriptive stage rate |
| Published relative difference in ticket selection | 18.67% | Published experimental estimate or descriptive stage rate |
| Published relative difference in checkout completion | -44.58% | Published experimental estimate or descriptive stage rate |
| Overall purchase probability difference across the test | 14.1% | Published experimental estimate or descriptive stage rate |
| Overall 10-day revenue difference | 20.64% | Published experimental estimate or descriptive stage rate |
| Session purchase probability difference | 12.43% | Published experimental estimate or descriptive stage rate |
| Session revenue difference | 18.96% | Published experimental estimate or descriptive stage rate |
| Revenue difference among buyers | 5.42% | Published experimental estimate or descriptive stage rate |
| Average seat price difference among buyers | 5.73% | Published experimental estimate or descriptive stage rate |
What changes when you compare the figures
| Comparison | Result | Calculation |
|---|---|---|
| Later minus upfront fees, selection gap | 4.4 percentage points | = 27.96 - 23.56 |
| Upfront minus later fees, final-stage completion gap | 14.89 percentage points | = 33.41 - 18.52 |
| Shift between the two stage-level gaps | 19.29 percentage points | = (27.96 - 23.56) - (18.52 - 33.41) |
| Extra selections per 1,000 event-page arrivals with later fees | 44 selections per 1,000 arrivals | = (27.96 - 23.56) * 10 |
| Fewer final-stage completions per 1,000 checkout arrivals with later fees | 148.9 completions per 1,000 arrivals | = (33.41 - 18.52) * 10 |
| All-assigned 10-day revenue, upfront relative to later fees | -17.11% | = (100 / 120.64 - 1) * 100 |
Methodology and sources
Transcribe Table 5’s conditional transition rates and compare groups within each step, in percentage points and per 1,000 stage arrivals. The difference between those two gaps is descriptive because reaching checkout can depend on assignment. Separately, Table 3 reports overall purchase and revenue for randomized visitors. Reversing the reported relative revenue difference requires dividing the original reference index by the treatment index. Figures reflect rounded published values.
Download the source data (CSV)
Download the calculations (CSV)
Related reading
Online discount statistics from a 314-trader pricing sweep
Shopping cart coupon statistics from Alibaba’s randomized promotion study
Equal-Value Checkout Promotions Can Get Different Responses
Countdown timer statistics and how to verify an ecommerce offer
Free shipping threshold statistics for basket size, orders and profit
Check the real price before you promise two Shopify discounts
Sources and definitions
Blake, Moshary, Sweeney and Tadelis, Price Salience and Product Choice, Marketing Science
The source data includes the population, period, denominator and limitations for every figure.