How to Create a Successful Google Display Ad (and Why the Best Ones No Longer Look Like Ads)

Something has changed about the ads I see while reading. A lot of them no longer look like ads. They sit in the feed in the same typeface as the articles around them, and I have caught myself reading one before noticing the little grey Sponsored label. Ads that look like articles are the display industry reacting to a measured fact about human attention. This piece checks whether Google Display really is as cheap as everyone says (it is, and that turns out to be the least useful fact about it), whether Google's pile of data really makes it effective (a real effect, much smaller than the pitch), and what separates a display ad that works from one that quietly burns a budget. I use my own apps as the running example: Repeat Recorder, Paint Vlix and Camera to Clipboard. The Search side of Google is in How Google Ads Really Work.

A Display click costs about a quarter of a Search click

On the most widely quoted Google Ads benchmark set, the gap is not subtle:

• Search clicks average around $2.69. On Search, you are buying someone who typed the problem you solve into a search box.
• Display clicks average around $0.63, roughly a quarter of the price of a Search click.
• Display is usually bought by the impression, at around $3 per thousand impressions on the Google Display Network, with the common range running from about $2 to $10.
• Google puts the Display Network at more than two million sites and apps, reaching roughly 90% of global internet users.

Display clicks cost four times less and convert five times worse

The Google Ads benchmark set that puts Display clicks at $0.63 also gives conversion rates:

• Search ads turn roughly 2.81% of clicks into a conversion, such as a sale for ecommerce.
• Display ads turn roughly 0.59% of clicks into a conversion, such as a sale for ecommerce, about one fifth as often as Search.
• The cheaper Display click produces the more expensive customer. The same Google Ads benchmark set puts ecommerce cost per conversion at about $45.27 on Search and about $65.80 on Display.
• Search sells you intent that already exists, while Display interrupts people reading a recipe or checking a score.
• A Display click is cheap because the person clicking was not looking for what you sell.

Clicks from people who were not looking are why so many small advertisers report that Display "brought loads of traffic and nothing else". Display did exactly what it was priced to do.

Use Camera to Clipboard to shoot a real photo of your product

On Display the picture does all the persuading that intent does on Search, and a photograph of the actual product beats a stock image. Camera to Clipboard, one of my own apps, removes the annoying step of getting a phone photo into the tool you design in.

• A photo you take inside Camera to Clipboard is ready to paste on your computer the moment you take it. Camera to Clipboard sends the photo straight from the phone to the computer over your own local network, with no email, cable or upload.
• Pasting a Camera to Clipboard photo into Photoshop or GIMP drops it in as a new layer, ready to build the ad around.
• A Display customer costs about $65.80 against $45.27 on Search, so the ad's picture is the one thing left you can improve for free. A better product photograph costs nothing but twenty minutes of your time.

432 experiments found almost nobody can measure their campaigns

• The largest clean study is a meta-analysis of 432 randomised field experiments run on the Google Display Network. Those experiments used a technique called ghost ads, which identifies the exact people in the control group who would have seen the ad.
• In 432 Google Display experiments, the ads lifted site visits by a median of about 16% over people who saw no ad. The median lift in conversions was about 8%, far below the conversions Google Ads credits to a Display campaign.
• The sales an ad causes are tiny compared with the normal ups and downs in a company's sales. Across a landmark set of 25 large advertising field experiments, the typical margin of error on advertising return on investment was more than 100 percentage points wide.
• Proving a Display campaign worked can need ten million person-weeks of data (one person tracked for one week is one person-week), which a small advertiser does not have and never will.
• Even on Search, eBay's large-scale experiment found eBay's ads on searches for its own brand name had no measurable short-term benefit at all.

Google's data makes Display targeting better than random, but only by a small margin

On the claim that Google's data is what makes Display ads work, the research is unflattering:

• Choosing who sees an ad from their browsing history adds surprisingly little revenue per ad shown. At a large publisher, ads shown to visitors carrying a tracking cookie earned only about 4% more revenue than ads shown to visitors without one.
• Display targeting hits the right age and gender only about 63% of the time, according to Nielsen's digital ad ratings work, even though age and gender have the best data in the industry.
• Google reversed its plan to remove third-party cookies from Chrome. In October 2025 Google also retired most of the Privacy Sandbox proposals it had built to replace third-party cookies.

Google's data makes Display work better than random, and randomised experiments confirm a real lift. Google's data does not make an ad shown to a casual reader work like an ad shown to someone searching for your product, and the lift is small enough that your own Google Ads reports cannot reliably see it.

Banner blindness is a real reflex and native formats answer it

Being recognisable as an ad has a measurable cost, and the research on banner blindness is unusually old and unusually consistent.

• Banner blindness, people's habit of ignoring anything that looks like an ad, was documented in 1998. A Rice University study by Benway and Lane found users missed banner-style elements even when those elements contained exactly the information they were hunting for.
• Eye-tracking has confirmed banner blindness for nearly three decades. Studies by Nielsen Norman Group, a usability research firm, find readers' eyes skip the top banner zone, the right-hand sidebar, and anything with a rectangular border, high-contrast colour and stock photography.
• Banner blindness is selective attention, not hostility toward advertisers. An ad-shaped rectangle has been irrelevant to readers for twenty-five years, so the brain learns to ignore it.
• Around 30% of internet users worldwide run an ad blocker at least sometimes, so Display ads never reach those users at all.

Native ads, which look like the content around them, exist because they get past banner blindness.

Native ads were looked at 53% more often than banners in an eye-tracking study

• Native ads are read like content because they are shaped like content. A study by IPG Media Lab, an ad agency group's research lab, and Sharethrough, a native ad company, which surveyed 4,770 people and eye-tracked 200, found consumers looked at native ads 53% more frequently than display ads.
• Native ads make people more likely to want to buy, not just more likely to look. In the IPG Media Lab and Sharethrough study, native ads registered an 18% higher lift in purchase intent than banner ads.
• Native ads are also more often actually seen on screen, with around 81% of native ads counted as viewable in 2026 data against roughly 64% of desktop banners.

Feed Google complete asset sets instead of designing banners

Google assembles native-looking ads itself, so your job is to give Google enough images and text to build one instead of a plain rectangle.

• A responsive display ad is a set of headlines, images and logos, not a finished layout. Google recombines your headlines, descriptions, logos and images to fit whatever ad slot it wins.
• Supply images in all three aspect ratios Google accepts, or Google can only build a plain generic banner. Horizontal at 1.91:1 (1200x628), square at 1:1 (1200x1200), and vertical at 9:16 (900x1600).
• Google asks for 5 to 10 images per aspect ratio, up to five headlines and five descriptions, and 3 to 4 distinct ads per ad group. Adding a video widens the places Google can show the ad.
• Demand Gen, Google's campaign type for feed ads, puts your ad inside the feeds of YouTube, Discover and Gmail, where video that looks shot on a phone beats polished commercials.

Design a Display ad for the ad slot you want to win, not for the ad you want to make.

Use Repeat Recorder to rehearse the phone-shot video you upload

Video widens the placements you can win, and feed placements want video that looks shot on a phone by a real person. In a phone-shot video, how you speak is the entire production. Repeat Recorder, another of my own apps, records you and plays the take straight back without you touching the phone.

• Repeat Recorder replays each take automatically and then starts recording again, so you can run the same opening line twenty times without picking the phone up.
• Google asks for up to five headlines and five descriptions carrying genuinely different messages, so you need several scripts rather than one reworded five times. Saying each headline out loud shows which headlines are the same sentence in different words.
• Repeat Recorder saves takes into folders you name yourself, so one folder per ad group keeps that group's scripts together.

Keep the brand loud, because ad size decides if you were seen

• Keep your brand unmistakable in a native ad, because looking like content must never mean hiding who the ad is from. In Reddit's own analysis of its ads, simply having a visible logo roughly doubled the conversion rate.
• The vampire effect is an ad so striking that nobody remembers the advertiser. With the vampire effect, viewers remember the joke in the ad but cannot say which company ran it.
• Native formats make the vampire effect worse, because a faintly branded native unit is read as just another piece of content. Make a native ad blend in with the page's layout, never hide your brand's identity, as I explain in Do Meme Ads Actually Work.
• Supply a clean logo at 1:1 or 4:1, with no flair or coloured background, and use one idea per image with the product as the subject.
• Choosing a Display ad size decides how often the ad is actually seen on screen, not just how it looks. Roughly 82% of 300x600 ads count as viewable, against 71% of 300x250 ads and 56% of 728x90 ads.

The two headline fields are not interchangeable

In a Google responsive display ad, the long headline has the most room and shows up in the fewest placements.

• The 30-character limit on the short headline is the real constraint, not the 90 characters of the long headline. The 30-character short headline is the only version of your message that every placement can display.
• The long headline has to stand on its own, because Google warns the long headline may be shown with no description beside it.
• Headlines get clipped on narrow mobile placements, so front-load the words that matter.
• A literal statement of what the product does survives cropping better than a clever headline, because wordplay needs setup and setup costs characters.
• Write the 30-character short headline first. "Snap on phone, paste on Mac" is 27 characters and survives everywhere. "Never email yourself a photo again" is 34 characters, so it gets cut.

Retargeting reports a $38 customer that really costs $138

Retargeting, showing ads to people who already visited your site, is the one thing almost every Display guide tells you to start with. This guide does not recommend starting with retargeting, because holdout tests show retargeting ads cause far fewer sales than Google Ads credits them with.

Retargeting advertises to people who came to you and left

Retargeting does not mean re-serving an ad to someone who already saw that ad.

• A retargeting audience is built from actions on your own site or app, not from ad exposure. A retargeting audience is people who visited your site, viewed a product, started a signup, installed your app or abandoned a cart.
• A retargeting ad does not repeat how someone first found you, it follows up on the moment they left without buying.
• Visitors from Display ads shown to strangers convert at about 1.5%, so 98 in 100 of those visitors leave without doing anything. Most of those visitors left for temporary reasons, and a prospecting campaign (the trade word for advertising to strangers) does nothing to bring them back.
• A retargeting audience needs nobody to have clicked an ad, and you do not even have to be running ads yet. The retargeting tag records each visitor when your page loads, so a site with decent unpaid traffic is already building an audience it could advertise to tomorrow.

The analytics tag you already have is the whole setup

• Linking your existing Google Analytics 4 (GA4) property beats installing a separate Google Ads remarketing tag. Link your Google Analytics 4 property to your Google Ads account and use its audiences, editing no pages at all.
• On a hand-built site, a second tracking tag has to be pasted into every page, including every page you write afterwards.
• Turn on Enable Personalized Advertising while you link Google Analytics to Google Ads, because that setting is the one everybody misses. The Google signals setting in Analytics must be on too, or your retargeting audiences look healthy but never fill with people.
• Build retargeting audiences in Google Analytics, not in Google Ads. Audiences built in Google Analytics export automatically for remarketing on Search, Display and YouTube, within up to two days.
• The website tag cannot see people who installed your mobile app and drifted away. Reaching lapsed app users needs the app's own analytics SDK (the analytics code built into the app) linked to your Google Ads account.

Build each retargeting audience from people who viewed a product page and did not buy

• Define each retargeting audience by what visitors did on your site and how recently they did it. For example, an audience of people who visited the pricing page in the last 30 days but never reached the thank-you page.
• One site-wide retargeting audience mixes readers in with buyers, so on a site that publishes articles as well as products, the retargeting ad that follows a visitor is close to random.
• Check each retargeting audience reaches Google's minimum of 100 active users in the last 30 days. A narrow retargeting audience on a low-traffic site can sit below that line and never serve.
• Do not assume a product feed is the better retargeting ad. Marketing professors Anja Lambrecht and Catherine Tucker, in their field experiment in the Journal of Marketing Research, found dynamic remarketing, which shows the exact item somebody looked at, performed worse than a generic brand ad on average.
• Set a membership duration for each retargeting audience, because somebody who left your site 89 days ago is not the same prospect as somebody who left yesterday.

The goal is the right moment, not wearing anyone down

• Retargeting does need each person to see the ad roughly three or four times, with a cap. Common display practice caps an ad at three to five impressions per person per week.
• Showing a retargeting ad repeatedly works by catching somebody on the day they are ready to buy, not by nagging them.
• Showing the same person a retargeting ad with no cap is how retargeting earns its reputation for being creepy, and the annoyance it causes never shows up in your Google Ads reports.
• Remove people who have already bought from the retargeting audience immediately. Advertising a product to somebody who already bought it is the single most common retargeting mistake.

Every dashboard number favours retargeting and all are measured

• Retargeting converts roughly two and a half times better than advertising to strangers. The median retargeting conversion rate is about 3.8%, against about 1.5% for ads shown to strangers.
• The gap in click-through rate between retargeting and standard display ads is bigger still, around 0.7% of retargeting ads clicked against roughly 0.07% of standard display ads.
• Retargeting clicks are not cheaper than cold Display clicks. Retargeting clicks cost near $0.76 against the $0.63 Display Network average, and the retargeting impression price runs 20% to 70% higher.

All of them measure the wrong thing, so $38 becomes $138

• Retargeting overstates the sales it causes more than any other ad format, because retargeting aims ads at the people most likely to buy without seeing an ad.
• A typical holdout test, which hides the ads from a random group, finds retargeting ads caused about 27.5% of the sales credited to them. So every reported retargeting cost per customer has to be multiplied by about 3.6.
• Corrected for the sales the ads really caused, Google Display is the most expensive of the three main places to retarget. A Google Display retargeting customer reported at $38 really costs about $138, against $95 on Google Search and $80 on Meta.
• Install the free retargeting tag on day one, but fund retargeting ads only as a test with a holdout group attached. Keep a random slice of the retargeting audience from seeing the ads for a few weeks, and compare their purchases with everyone else's.
• Cart abandonment is the strongest case for retargeting anywhere. A cart-abandonment retargeting customer reported at $13 really costs about $47 once corrected, as shown in what retargeting actually costs once you correct for what it caused.

Switch off app placements and unseen impressions first

For a small advertiser, the largest single improvement to a Display campaign is usually switching things off.

• Mobile app placements are the classic Display budget hole. Ads inside games, flashlight apps and children's apps generate plenty of clicks and almost no conversions, so exclude app categories first.
• Roughly 60% of mobile banner clicks are accidental taps, according to survey work, and you paid for every one.
• Made-for-advertising sites, built only to show ads, took 21% of ad impressions in 2023, according to the ANA (the Association of National Advertisers, a US trade group). The ANA found that share fell to about 6.2% in 2024, but only for advertisers who actively exclude those sites.
• Roughly a quarter to a third of Display impressions are never actually seen. Across ad networks, about 72% of ad impressions count as viewable in 2026.
• Google's optimised targeting setting is on by default and treats your chosen audience as a hint, not a rule. Optimised targeting will stop showing ads to your chosen segments if it finds cheaper conversions elsewhere.

View-through conversions are a diagnosis, never a receipt

• A view-through conversion credits an impression that was never clicked. Somebody was served your ad, ignored it, converted later for their own reasons, and the impression takes the credit.
• At the scale of Google Display, view-through conversions the ad did not cause are guaranteed, and some of the impressions counted were never on anyone's screen for a full second.
• Diagnose with view-through conversions, but decide budget with clicks and holdout tests.
• Give Google a real conversion event, such as a purchase, or Google's automatic bidding will chase people who only scroll.

Display traffic converts on landing pages far worse than search or email traffic, so the landing page needs one clear idea, one action, and no assumption that the visitor remembers who you are.

Use Camera to Clipboard to put a product photo on the landing page

A landing page for cold Display traffic has to show the product within a second of arriving, which means a real photograph rather than an illustration of a concept. Camera to Clipboard, one of my own apps, gets that photograph from the phone that took it onto the computer that builds the page.

• A photo you take inside Camera to Clipboard is on your computer's clipboard the moment you take it, so reshooting an image for a page becomes a thirty second job.
• Camera to Clipboard can send several photos at once from the phone's library, which suits the images Google requires in three aspect ratios.
• Google prices the Display auction, but the photograph at the top of your landing page is entirely yours.

Standalone Display is being retired into Demand Gen in 2026

Google is not shutting down the Display Network. It is changing how you buy it.

• Google announced in May 2026 that standalone Display campaigns are being retired into Demand Gen, with the Display Network becoming an inventory channel inside it.
• Through the second half of 2026, new standalone Display campaigns stop being creatable, after the migration tool started rolling out in June 2026.
• In 2027 Google migrates any remaining Display campaigns automatically.
• Migrating a Display campaign yourself with Google's migration tool preserves up to 42 days of performance history, cutting the time Google's bidding needs to relearn to roughly a day or two.
• The same two million Display Network sites and apps are still available inside Demand Gen, and Demand Gen's channel controls let you run on the Display Network alone.
• Google cites an average 9.5% increase in ROI for advertisers adding Display Network inventory to Demand Gen campaigns, a vendor figure to treat with suspicion.

Four ways to buy Google inventory, compared on what differs

No way of buying Google ads wins on every measure. Cold Display buys the cheapest attention and the most expensive customers. Search buys the most expensive attention and the cheapest customers. Retargeting wins every measure the Google Ads dashboard reports, but loses on cost per customer once the reported $38 is corrected to $138. Native feed ads get looked at the most and demand the most creative effort.

Create the demand cheaply on Display, then let Search close it

• Display buys cheap reach from people who were never going to search for you. If you need someone to buy today, that ad budget belongs on Search.
• Judge Display on how many people it reaches and the extra sales a holdout test shows it caused, never on the sales credited to the last ad clicked.
• Install the retargeting tag from day one, but do not start by funding retargeting ads. If you do fund retargeting, fund it as a test with a holdout group attached.
• Give Google complete asset sets and aim at native placements, with plain, real-looking images and video and a loud brand.
• Exclude mobile app categories and cap how often one person sees your ad before improving the ad's images and text.
• Then run Google Search ads alongside Display to catch the people Display made interested. Display and native feeds create the demand, and Google Search captures it when somebody types your category into a search box.

Display advertising rewards the advertiser who knows they are buying attention, and quietly punishes the one who thought they were buying customers at a discount.

Conclusion: how to make a Google Display ad work, in one glance

• A Google Display click is genuinely cheap and a Google Display customer is not. A Display click costs about a quarter of a Search click and converts about five times worse.
• A retargeting customer the Google Ads dashboard reports at $38 really costs about $138. Holdout tests credit retargeting with only about 27.5% of the sales it claims.
• Exclude app placements and cap frequency before spending anything on better Display creative. Cap each Display ad at three to five views per person per week.