Display Creative That Earns Its Impressions

A display placement that never gets noticed is not cheaper media, it's wasted media with a lower sticker price. Most display underperformance on client accounts traces back to creative decisions made once at launch and never revisited: one static size uploaded to a responsive slot, one message running for the length of the flight, one set of assets expected to carry a campaign that runs for a full quarter. The targeting can be flawless and the placement can still fail if the creative itself stopped registering weeks ago.

Specs are the floor, not the strategy

Getting the technical specs right is table stakes, and it is still worth getting precisely right because a rejected or downscaled asset never gets the impression at all. A responsive display ad needs a defined set of headlines, descriptions, images, and logos for Google's system to actually assemble into working combinations, and the most common image sizes that clear the most inventory are the 1.91:1 horizontal and 1:1 square formats, not whatever crop happens to be sitting in a client's brand folder. The IAB's standard ad portfolio shifted years ago from fixed pixel sizes to aspect-ratio-based units specifically so creative could flex across the range of screens a modern buy actually reaches. An account still building single-size static banners is leaving reach on the table before targeting even enters the conversation.

Format variety is what keeps an audience from tuning out

Format variety matters more than any single spec. A campaign running one static banner and one motion version across a whole flight is still, functionally, one creative repeated at scale, and repetition is exactly what audiences learn to filter out. A working format set covers the standard IAB sizes, a couple of native-style units that blend into content rather than interrupting it, and enough headline and image variation within the responsive asset group that the algorithm actually has real combinations to test against each other instead of serving the same pairing by default.

Refresh cadence is the discipline most accounts skip

Creative fatigue is a documented, measurable pattern, not a vague creative-team complaint: the same asset shown to the same audience produces a declining response curve as familiarity turns into indifference and then into active avoidance. The fix is not a single new banner dropped in mid-flight, it's a refresh cadence built into the media plan from the start, with new creative variants scheduled to enter rotation before performance actually starts declining rather than after a client notices the numbers slipping. An account that treats creative refresh as a scheduled part of the plan, the same way budget pacing or bid adjustments are, catches fatigue before it shows up in a monthly report.

• Rotate in new creative variants on a fixed schedule tied to the flight length, not only when performance visibly drops
• Track impression frequency per user alongside click-through rate, since fatigue shows up in frequency data before it shows up in click data
• Keep a message and a counter-message in rotation together rather than one dominant creative running unopposed
• Retire a top performer before it decays instead of running it until the numbers force the decision

Frequency, not just CPM, is where fatigue shows up first

A creative shown too often to the same narrow segment burns through goodwill faster than the aggregate CPM suggests, because the average masks a subset of the audience seeing the same banner a dozen times while the rest of the buy is still under-exposed. Google Ads' own frequency capping controls let a campaign limit impressions or interactions per user directly, at the campaign level, which is the cheapest lever available for controlling fatigue before it costs real performance. Capping frequency per creative, not just per campaign, and rotating variants against that cap costs less than the retargeting spend an account often reaches for once the fatigue has already set in.

Frequency caps are also not one-size-fits-all across a media buy. A cap set uniformly at the campaign level treats a small, high-intent retargeting pool the same as a broad top-of-funnel awareness audience, when the two need very different exposure limits: a narrow retargeting audience saturates fast and needs a tighter cap, while a broad prospecting audience can typically absorb more repetition before fatigue sets in. Setting frequency caps per audience segment, not just per campaign, is a small adjustment that keeps the same creative pool working harder across more of the account.

Getting client sign-off on the cadence, not just the creative

Refresh cadence is also a conversation an agency needs to have with the client up front, not something buried inside a media plan a client never actually reads. A client who does not know a refresh cadence is baked into the plan reads every new creative variant as scope creep or an unplanned production cost, when it is actually the maintenance the placement needs to keep performing at the rate the original plan promised. Naming the cadence in the same document that sets the flight dates and the budget, alongside the format mix, turns refresh from a surprise mid-flight request into an expected part of the deliverable.

• State the refresh cadence in the media plan itself, tied to specific dates or performance thresholds, not left implicit
• Build the creative production cost of the refresh into the original quote, not as a change order once fatigue shows up
• Show frequency and click-through trend lines in reporting so a client can see fatigue setting in before it's argued about
• Agree in advance on what triggers an off-cycle refresh, a hard performance drop versus a scheduled swap

Dynamic creative optimization takes the manual work out of variation

Format variety and refresh cadence still require someone to produce each new asset by hand. Dynamic creative optimization automates a layer underneath that: a single template pulls modular pieces, headline, image, offer, price, from a feed and assembles them at serve time based on who is seeing the ad, what device they're on, and what has performed best so far. That is not a replacement for the format and refresh discipline covered above, it's what makes maintaining that discipline affordable at scale, since a template with ten headline variants and five image variants produces fifty working combinations, each one earning its own click-through rate rather than a single blended average, from a fraction of the production effort of building fifty static assets one at a time.

DCO earns its setup cost once an account has a real feed to draw from, a product catalog, location-specific offers, a price that changes by market, not for a single evergreen brand message that never varies by audience. Setting it up also requires more structured asset planning up front than a standard responsive ad group: the headline and image variants have to be built as genuinely interchangeable modules from the start, not repurposed after the fact from a set of static banners that were never designed to be split apart. Conduit's Wisconsin Beef Council case study shows that same discipline paying off on a lean budget: a 2x higher click-through rate came from treating format and creative variety as a first-class part of the media plan, not an afterthought bolted on once the flight was already live.

Accessibility is a creative-quality issue before it's a compliance one

A banner that fails basic contrast standards does not just risk a complaint, it fails to register for a real share of the audience seeing it. WCAG's contrast guidance sets a 4.5:1 minimum contrast ratio between text and its background for standard-size text, a standard that applies directly to any headline or call-to-action placed over a photographic or colored background, which is most display creative. A CTA button that reads fine against the mockup's background can fail that ratio entirely once the responsive system swaps in a different image, exactly the kind of failure a pre-launch QA pass should catch before launch, not after a client's own accessibility review flags it.

The same logic applies to motion. A creative with fast flashing or rapid animation is not just a design choice, it's a real access barrier for users with vestibular disorders or photosensitivity, and it's worth building a simple internal standard, a cap on flash rate, a static fallback frame for any format running animation, into the creative spec the same way sizing and file-weight limits already are. None of this is a large lift once it's a standard checklist item. It's a much bigger one once it surfaces as a client-side complaint after launch.

• Check contrast ratio for every headline and CTA against its actual background image, not just the mockup
• Cap flash and rapid-motion rate in any animated format, and keep a static fallback frame available
• Test legibility at the smallest served size, not just the largest asset in the set

A quick QA pass before anything launches

A handful of checks before a creative set goes live catch most of the problems that otherwise surface as a mid-flight scramble: confirming every required asset size actually renders correctly rather than getting auto-cropped by the ad server, checking that headline and description copy still reads correctly across the shortest and longest character-count variants the responsive format allows, and verifying the call to action is legible against every background image in the set rather than just the one used in the mockup. None of this is complicated, and all of it is cheaper to catch before launch than after a client asks why one variant looks broken on a specific placement.

Specs and formats are the part of display creative an agency can standardize once. Refresh cadence is the part that has to be run every month on every account, which is exactly the kind of operational discipline that is easy to promise in a pitch and hard to sustain without a team built around it. Conduit's white label display advertising program builds refresh scheduling into every campaign from the start rather than reacting to fatigue after a client flags declining numbers.