Display retargeting lifted website visits by 17% and purchases by 11% for an online retailer, according to the Ghost Ads study by Johnson, Lewis, and Nubbemeyer. According to Stanford Graduate School of Business research, a randomized field experiment with an online seller of home-improvement products found retargeting caused 14.6% more users to return within four weeks. Those two results frame the retargeting ad performance statistics that follow.
Yet according to a Marketing Science Institute working paper, retargeting advertising had little impact on consumer response relative to the no-retargeting control group. Criteo reported Q2 2026 revenue of $428 million, down 11% from $483 million a year earlier, while Meta’s average price per ad increased by 12% year over year. Together, they show modest, conditional lift alongside rising average ad prices.
Key Takeaways
- Customers who received a cart retargeting ad within 30 minutes to one hour were less likely to make a purchase, while a late ad after one to three days had a positive incremental impact.
- 33% of the effect of the first week’s retargeting occurred on the first day, according to Stanford researchers.
- 42% of US online shoppers have abandoned a cart because they were just browsing or not ready to buy, according to Baymard Institute.
- Criteo’s Performance Media revenue fell about 10% year over year in Q2 2026, even as ad impressions across Meta’s Family of Apps increased by 14%, according to Meta.
- Google requires a minimum of 100 active visitors or users within the last 30 days before a data segment can serve on the Display Network, Search Network, or YouTube, per Google Ads Help.
- Criteo served 16,752 clients in Q2 2026, compared with 17,744 in Q2 2024.
Editor’s Choice
- Average cart abandonment rate: 70.22%, calculated from 50 studies.
- Returning users from retargeting: 14.6% more users returned within four weeks.
- Criteo media spend: $1.1 billion in Q2 2026 and $4.5 billion in the last 12 months.
- Criteo Q2 2026 revenue: $428 million, against $483 million in Q2 2025.
- Meta ad impressions: up 14% year over year in Q2 2026.
- Maximum list membership on Google: 540 days on the Display Network and Google search.
Retargeting Ad Performance Statistics From Field Experiments
- Display retargeting lifted website visits by 17% and purchases by 11% in the Ghost Ads experiment.
- Ghost Ads measured an online retailer’s display retargeting campaign, and its implementation records more than 100 million predicted ghost ads per day.
- Advertisers using Ghost Ads can measure ad lift just as precisely as with intent-to-treat or PSA experiments while spending at least an order of magnitude less.
- The effect of advertising in week 2 of the campaign was higher when the user was assigned to a nonzero level of advertising in week 1.
- Retargeting promotions lifted the purchase rate compared with retargeting advertising among consumers who abandoned their shopping cart.
- In an online travel firm’s field experiment, increased ad specificity was on average less effective than generic information.
- Specific ads beat generic ads only when consumers had well-defined product preferences and were actively engaged in the category.
| Study | Setting | Outcome measured | Result |
|---|---|---|---|
| Johnson, Lewis and Nubbemeyer (Ghost Ads) | Online retailer, display retargeting | Website visits | 17% lift |
| Johnson, Lewis and Nubbemeyer (Ghost Ads) | Online retailer, display retargeting | Purchases | 11% lift |
| Stanford GSB study (2019) | Online seller of home-improvement products | Users returning within four weeks | 14.6% more |
| MSI working paper (2021) | Shoppers who abandoned a cart | Response to retargeting ads vs control | Little impact |
| Lambrecht and Tucker | Online travel firm | Dynamic vs generic ads | Less effective on average |
Source: NBER 2016, Stanford Graduate School of Business 2019, Marketing Science Institute 2021, MIT Initiative on the Digital Economy 2011
About This Data
Compiled from eleven sources published between December 2011 and October 2026: five academic studies, three company earnings releases, one Google Ads Help page, one e-commerce research institute, and one advertising industry body. Figures are updated when sources publish new editions.
Key finding: Ghost Ads found display retargeting lifted website visits by 17% and purchases by 11%, while the MSI cart study found ads alone moved little and a Journal of Marketing study found cart ads sent too early lowered purchases, so measured lift depends on audience and timing.
What are retargeting ads, and how can they improve conversion rates?
Retargeting is a form of internet advertising in which banner ads are displayed to users after they visit the advertiser’s website. They raise conversion only when they reach people who would not have returned anyway, which only a randomized test can measure.
Criteo Performance Media Revenue by Quarter
- Criteo’s Performance Media revenue was $380.1 million in Q2 2026, against $421.8 million in Q2 2025.
- Criteo attributed the Performance Media decline to soft performance in Commerce Growth, partially offset by improved year-over-year trends in AdTech Services.
- Over two years, Performance Media revenue fell about 8.7%, down from $416.5 million in Q2 2024.
- The highest quarter in the nine-quarter series was Q4 2025 at $464.8 million.
- Criteo’s media spend was $1.1 billion in Q2 2026, up 9% year over year at constant currency.
View the data
Platforms such as rtb.com compete for the same budgets across the programmatic advertising market.
Recent Developments
- July 22, 2026: Alphabet reported that consolidated revenues increased 24% to $119.8 billion, with 17% growth in Google Search & other and 13% in YouTube ads.
- July 29, 2026: Meta reported revenue of $60.80 billion, an increase of 28% year over year, with ad impressions up 14% and average price per ad up 12%.
- August 5, 2026: Criteo reported Q2 2026 revenue of $428 million, against $483 million a year earlier, and appointed Connor McGogney as Chief Financial Officer, effective August 10, 2026.
- August 5, 2026: Criteo now expects fiscal 2026 Contribution ex-TAC to decrease 12% to 10% at constant currency.
- August 5, 2026: Criteo said over 2,000 brands now advertise on ChatGPT through it across seven countries.
Retargeting Timing and Frequency Effects
- Customers who received an early cart retargeting ad within 30 minutes to one hour were less likely to make a purchase than those who did not receive it.
- A late cart retargeting ad after one to three days had a positive incremental impact on customer purchases.
- A second study replicated this double-edged impact for ads delivered by mobile SMS.
- 33% of the effect of the first week’s advertising occurred on the first day.
- The impact of retargeting decreased as the time since the consumer first visited the website increased.
- Shoppers who abandoned more recently and had a single category of products in their cart exhibited higher purchase lift.
- The lift in website revisits was moderated by the elapsed time since cart abandonment in an inverse U-shaped relationship.
| Study | Timing window | Effect on the outcome |
|---|---|---|
| Journal of Marketing study (2020) | 30 minutes to one hour after cart abandonment | Lower purchase rate than no ad |
| Journal of Marketing study (2020) | One to three days after cart abandonment | Positive incremental impact on purchases |
| Stanford GSB study (2019) | First day of the campaign | 33% of the first week’s effect |
| Stanford GSB study (2019) | Week 2, after week 1 exposure | Higher effect than without week 1 ads |
| MSI working paper (2021) | Moderate time since cart abandonment | Website revisit lift, better promotion only |
Source: American Marketing Association 2020, Stanford Graduate School of Business 2019, Marketing Science Institute 2021
Why it matters: Delivering cart retargeting ads too early can engender worse purchase rates than not delivering them, thus wasting online advertising budgets, and both the negative impact of early ads and the positive impact of late ads can be amplified when the products in the retargeted carts are larger in quantity. Timing is a budget decision.
Cart Abandonment Behind Retargeting Audiences
- The average documented online shopping cart abandonment rate is 70.22%, based on 50 different studies.
- 42% of US online shoppers have abandoned a cart because they were just browsing or not ready to buy.
- Excluding just browsing, 40% cited extra costs that were too high, and 20% said delivery was too slow.
- 19% did not trust the site with their credit card information, and 18% left because the site wanted them to create an account.
- 13% said the returns policy was not satisfactory, and 12% could not see or calculate the total order cost up front.
- 10% abandoned because the credit card was declined and 9% because there were not enough payment methods.
View the data
A shopper put off by fees needs a price answer rather than a product reminder, and e-commerce conversion rate benchmarks show how much of that pool converts at all.
Google Ads Remarketing List Requirements
- Google Display Network segments must have a minimum of 100 active visitors or users within the last 30 days.
- Google Search Network and YouTube segments carry the same minimum of 100 active visitors or users within the last 30 days.
- Customer lists uploaded before February 1, 2024 must still meet the previous minimum of 1,000 matched users.
- On April 22, 2025, Privacy Sandbox VP Anthony Chavez said third-party cookies will remain on Chrome, dropping plans to introduce a standalone prompt.
| Google network | Minimum active users (last 30 days) | Default membership | Maximum membership |
|---|---|---|---|
| Display Network | 100 | 30 days | 540 days |
| Google search | 100 | 30 days | 540 days |
| YouTube | 100 | Not listed | Not listed |
Source: Google Ads Help 2026
Cookie-based lists feed display retargeting, so Chrome’s share of browser usage makes its cookie policy decisive.
Criteo Revenue by Region
- APAC revenue fell to $80.7 million in Q2 2026, a drop of about 17% year over year.
- Americas revenue was $176.0 million in Q2 2026, down about 12%.
- EMEA revenue was $171.3 million, down about 8%.
- Americas markets generated about 41.1% of Criteo’s Q2 2026 revenue.
View the data
Every region shrank, so the decline was not confined to a single market.
Criteo Advertiser Client Count
- Criteo counted 16,752 clients in Q2 2026, up from 16,528 in the prior quarter, about 2% fewer than a year earlier.
- Criteo became OpenAI‘s first advertising technology partner in March 2026.
- Criteo had 992 fewer clients than in Q2 2024.
- The client count was 17,142 in Q2 2025 and 17,269 in Q4 2024.
View the data
Criteo Contribution ex-TAC by Segment
- Contribution ex-TAC was $255 million in Q2 2026, compared with $292 million a year earlier, a decline of about 13%.
- Retail Media Contribution ex-TAC fell, reflecting a $21 million headwind from scope changes with two specific Retail Media clients, about 21% below the prior-year quarter.
- Excluding that headwind, Retail Media Contribution ex-TAC grew 20% in Q2 across the underlying client base.
- For fiscal 2026, Criteo expects an Adjusted EBITDA margin of approximately 30% of Contribution ex-TAC.
- Criteo expects third-quarter Contribution ex-TAC between $237 million and $241 million, or 15% to 14% lower year over year at constant currency.
View the data
By the numbers: Performance Media contributed $208.3 million of Criteo’s $255.5 million Contribution ex-TAC in Q2 2026, about 81.5% of the total, against $47.2 million from Retail Media (about 18.5%). Performance Media still supplies most of Criteo’s Contribution ex-TAC.
Google Advertising Revenue by Channel
- Google advertising revenue was $81,629 million in the quarter ended June 30, 2026, against $71,340 million a year earlier.
- Google Network revenue was $7,303 million, compared with $7,354 million a year earlier.
- Google Services revenue increased 15% to $94.5 billion.
- Google Network revenue slipped by $51 million year over year.
- Google Services revenue grew 17% in Google Search & other and 13% in YouTube ads.
- Total traffic acquisition costs rose by $1,474 million year over year.
View the data
Google Network covers ads on partner sites and apps, where Display remarketing runs, while search pricing sits in the Google Ads revenue and CPC data.
Meta Ad Impressions and Price per Ad
- Ad impressions delivered across Meta’s Family of Apps increased by 14% year over year in Q2 2026.
- Average price per ad increased by 12% year over year.
- Revenue was $60.80 billion, an increase of 28% year over year.
- Total costs and expenses were $42.03 billion, an increase of 55% year over year.
- Daily active people averaged 3.60 billion for June 2026, an increase of 3% year over year.
View the data
Meta reports one average price across all of its ads, and the Facebook ad cost data follows how that price moves over time.
How much do retargeting ads cost?
Retargeting has no fixed price, because past-visitor audiences bid in the same auctions as other ads. Meta’s average price per ad increased by 12% year over year in Q2 2026, an average across all of Meta’s ads rather than a retargeting price. Browsers that block ads never enter those auctions, as ad blocker usage data shows.
What is the ROI of retargeting?
Retargeting ROI depends on incremental lift, not attributed conversions. Ghost Ads found retargeting lifted purchases by 11%, while retargeting promotions lifted the purchase rate compared with retargeting advertising. In the home-improvement experiment, retargeting caused 14.6% more users to return within four weeks, a revisit gain that still has to convert into sales. Attributed conversions overstate that return.
Conclusion
In the Ghost Ads experiment, an online retailer’s display retargeting campaign lifted website visits by 17% and purchases by 11%. A separate Journal of Marketing study tested cart-ad timing, and cart ads sent within 30 minutes to one hour reduced purchases. Cart reminders did better one to three days after abandonment than within the first hour, while the overall effect of retargeting faded as time since the first site visit grew.
Criteo’s Performance Media line is shrinking while Meta’s average price per ad rises, while the experiments point to timing as the lever advertisers control.




















































