Reach households before they search,
on inventory you can verify.
Display is the only paid channel where you can be billed for a placement and have no way to confirm anybody real saw it. The industry built a standard to fight exactly that, and almost no agency reads it. We check every domain against who is actually authorized to sell it, and exclude what cannot be verified before your budget reaches it.
Built for home services contractors, HVAC and roofing and plumbing companies, franchise systems and private equity portfolios — single location through to national scale, small business to enterprise. Run by an AI-first agency, so display feeds the same brand-entity signals AI platforms and search engines weigh when they decide which businesses to name.
Display runs inside an AI-first program here. AI-assisted placement analysis reads the whole delivery report rather than sampling it, exclusions compound into an asset that belongs to you, and the brand presence display builds feeds the entity that AI search engines and assistants draw on when they choose which businesses to name. Most agencies still buy display as impressions; we buy it as verified placement inside an AI visibility program.
Your ad ran somewhere. Can anyone tell you where?
A display report lists domains. That list is produced by the platform that billed you, and on most accounts nobody has ever checked a single line of it against anything independent. The check exists, it is free, and it takes minutes.

The mechanic is simple once someone explains it. When you buy display programmatically, your budget passes through several intermediaries before it reaches a page — a buying platform, an exchange, a selling platform, and finally the site itself. Each hop takes a share. Each hop is also, in a properly functioning market, identifiable.
The industry built a standard because it was not always functioning properly. The IAB Tech Lab describes ads.txt — Authorized Digital Sellers — as part of a broader effort to eliminate the ability to profit from counterfeit inventory in the open digital advertising ecosystem, providing a mechanism for content owners to declare who is authorized to sell their inventory. Its stated mission is to increase transparency in the programmatic advertising ecosystem, and version 1.1 added owner and manager domain values specifically to fight domain misrepresentation.
Read that carefully, because it is not our characterization. A standards body says, in its own published words, that counterfeit inventory and domain misrepresentation exist at sufficient scale to justify a global technical standard. If a display proposal you are reading does not mention that, the person who wrote it either does not know or would rather not raise it.
What this means practically is that every domain on your placement report can be checked. The site publishes who may sell its space. You compare that list against who actually billed you. Where the selling platform does not appear, the placement is unverifiable — not proven fraudulent, but not confirmable either, and unconfirmable is a perfectly good reason to stop paying for it.
None of this requires special tools or a vendor relationship. The files are published at predictable locations on the open web. The only scarce ingredient is somebody willing to spend an afternoon reading them, which is the same scarce ingredient as the search terms report on our paid search page.
The difference is one hop that declares nothing.
Both paths start with your money and end with an advertisement on a page. In one, every intermediary can be identified from published records. In the other, a link in the middle is invisible, and so is the answer to what your budget actually bought.

Nobody can confirm what you bought
- The selling platform is not in the site's file. It is offering inventory the owner has not authorized it to sell.
- The domain may not be what it appears. Misrepresentation is the specific problem the standard's owner and manager values were added to address.
- The page may exist only to carry advertising. Real URL, real load, no reason for a human to be there.
- The report still looks fine. Impressions delivered, domains listed, budget spent, nothing flagged.
- The exclusion list stays empty, so the same placements are bought again next month.
Every hop can be checked
- The selling platform appears in the site's published file, as a direct seller or an authorized reseller.
- The platform publishes its own identity, so the intermediary can be traced rather than assumed.
- The page has a reason to exist — editorial content, an audience, something a person came for.
- The report is the start of the work, not the end of it, because somebody reads it line by line.
- The exclusion list grows, and it travels with you to the next campaign and the next agency.
The transparency layer exists. Most buyers never open it.
These are open standards maintained by the industry's own technical body, plus the platform controls that let you act on what they tell you. All four are readable without an account, a login or a vendor.
A plain text file at a predictable location listing every platform authorized to sell that site's advertising, and whether each relationship is direct or through a reseller. The IAB Tech Lab specification exists to remove the ability to profit from counterfeit inventory.
Version 1.1 added owner and manager domain values specifically to make seller relationships more traceable and to strengthen the file as a tool against fraud in the buying and selling of advertising.
- Check each billed domain's file against the platform that actually sold the impression.
- Treat an absent seller as unverifiable and exclude the placement rather than argue about it.
The counterpart to ads.txt, published from the other side. Where ads.txt says which platforms a site authorizes, sellers.json lets a platform declare the identities of the sellers and intermediaries it represents.
Together they let a buyer trace a chain rather than assume one. Neither file on its own proves an impression was seen by a person; both together establish that the parties in the transaction are who they claim to be.
- Trace an unfamiliar seller back to a declared identity before its inventory enters a plan.
- Keep the finding, so the next campaign starts from what the last one established.
Mobile app inventory has the same problem and the same solution. The app-ads.txt standard extends the same declaration to applications, resolved through the developer URL listed in the app store.
In-app placements are where the least scrutiny usually happens, because they are furthest from anything the advertiser can look at directly, and because a screenshot of an app is harder to interrogate than a webpage.
- Apply the same standard to app inventory as to web, rather than treating it as a separate category.
- Exclude app placements with nothing published behind them, the same as unverifiable domains.
Knowing a placement is unverifiable is useless without the control that stops buying it. Google documents how to exclude specific webpages, sites and apps from a campaign, at the level the exclusion belongs.
An exclusion list is a compounding asset. It is built from your own billed placements, it grows every review cycle, and it should belong to you rather than to whichever agency happens to be running the account.
- Apply exclusions from observed placements and record why each one was added.
- Hand the list over on request, because it describes your spend rather than our method.
Five hops between your budget and a page somebody reads.
Programmatic display is not one transaction, it is a chain of them, and each link takes a share of what you paid. Understanding the chain is what makes the published records useful, because each record answers a question about a specific hop.
Where each check applies
Your budget enters at a buying platform, which bids on your behalf. The bid travels to an exchange, which runs the auction. The winning bid is filled by a selling platform, which is offering inventory on behalf of a publisher. The publisher's page renders the advertisement. Between each of those hops there is a share taken and an identity that can either be declared or left implicit.
The published records map onto that chain precisely. The publisher declares its authorized sellers, so the last hop is checkable. The selling platform declares its own identity and the sellers it represents, so the second-to-last hop is checkable. And the bid request itself can carry a record of the path it travelled, which the IAB Tech Lab covers across its OpenRTB specification and related supply chain work.
What none of it does is tell you a human being saw the advertisement. That is a separate question with a separate answer, and we are careful not to let the transparency layer stand in for it. Verified provenance means the transaction was what it claimed to be. It does not mean the impression was valuable.
- Buying platform — bids on your behalf, reports what it bought
- Exchange — runs the auction, passes the request along
- Selling platform — fills the bid, declares its own identity
- Publisher — declares which platforms may sell its space
- The page — where the advertisement finally renders
- Your exclusion list — the only part of the chain you control directly

Verify the inventory, then buy it, then keep verifying.
Display is usually run in the opposite order — launch broadly, gather data, optimize later. That works if somebody eventually looks at where the money went. Very often nobody does, and the campaign simply runs until the budget does.
Before the campaign launches
- Targeting scoped to the geography you actually serve, not the widest available
- Known unverifiable inventory excluded before a single impression is bought
- Creative sized and built for the placements the plan actually contains
- A measurement design agreed that does not depend on last-click
Every review cycle
- Every billed domain checked against the site's published seller file
- Unverifiable placements excluded and the reason recorded
- App inventory held to the same standard as web inventory
- Exclusions applied at the level they belong, not everywhere by reflex
Judged against the whole plan
- Branded search and direct traffic watched against flight windows
- A holdout geography kept out of the campaign where the market allows
- Display assessed on demand created, not on clicks it happened to catch
- The exclusion list handed over as an asset that belongs to you
Six things that happen whether or not the report looks fine
A display report almost always looks fine. Impressions delivered, domains listed, budget spent. This list is what happens after that, and it is the part that separates a campaign from a subscription.
Every billed domain checked against its published file
Not sampled and not limited to the top spenders. Each domain on the placement report is compared against the seller list that site publishes, and the finding is recorded alongside the domain.
Source: IAB Tech Lab — ads.txt
Unfamiliar sellers traced to a declared identity
Where a selling platform is not one we recognize, we look it up rather than assume, using the identity records the platforms themselves publish.
Source: IAB Tech Lab — sellers.json
App inventory held to the same standard
In-app placements get the same check as web placements, resolved through the developer record. Inventory nobody can look at directly is inventory that needs the check most.
Source: IAB Tech Lab — app-ads.txt
Exclusions applied and documented
Each exclusion carries the reason it was added, so a later manager does not remove a deliberate decision because they cannot see why it exists.
Source: Google Ads Help — exclusions
Geography kept inside the service area
Display targeting defaults are broad, and awareness bought outside the area you dispatch to is not awareness, it is waste. We name it as waste rather than counting it as reach.
Related: the same territory discipline runs through local visibility.
Creative built for the placement, not the deck
A display unit is small, briefly seen and frequently ignored. We build for legibility at actual size, and we check the destination is ready before we send traffic to it.
Related: landing page construction.
We will show you the placements we could not verify.
Display reporting is the easiest reporting in the industry to make look good, because impressions are abundant and cheap and nobody has to defend where they came from. We report the domains we excluded and why, which makes the numbers smaller and the campaign better.
We will also tell you when display is the wrong spend. If demand for your service already exists and goes uncaptured, awareness advertising is solving a problem you do not have yet. That finding costs us the campaign and we report it anyway.
Display sits alongside paid search, remarketing and the rest of the digital program, and the accounting discipline is the same across all of it.
A shorter list, and a reason for every line on it.
The output of the work is not a dashboard. It is a list of places your advertising will no longer appear, each with the finding that put it there — and a list of places it will, each of which somebody checked.

| What you see on the report | What it is usually taken to mean | Checked against the record What it actually establishes |
|---|---|---|
| A domain you recognize | The ad ran on that site | Confirmed only where the selling platform appears in that site's published file |
| A domain you do not recognize | Long tail, probably fine | Checkable in minutes — the file is public and the answer is unambiguous |
| An app placement | Nothing to check | Same standard applies, resolved through the developer record |
| Impressions delivered | People saw the ad | Delivery is not viewing, and provenance is not attention — separate questions |
| No exclusions on the account | Nothing needed excluding | Far more often it means nobody has looked yet |
The last two rows are where display goes wrong quietly. Everything above them can be established from published records. Neither of those two can, and treating them as settled is how a campaign runs for a year without anybody asking the question this page opens with.
Four display line items you can stop paying for.
One is a number that measures abundance rather than value. One is a service sold against a metric it does not control. One is inventory nobody can vouch for. The fourth is a promise the medium cannot keep.
Impressions reported as reach. An impression is a delivery event. It is not a person, it is not attention, and on the open web it is the cheapest thing in the transaction. A proposal that leads with impression volume is leading with the number easiest to produce in quantity.
Broad targeting sold as scale. Reaching more people is trivially achievable by loosening every setting, and it is what an account does by default. Scale that includes households outside your service area is not scale, and it should appear in the plan as waste rather than as reach.
Unverifiable inventory left in the plan. Not because it is necessarily fraudulent — most of it is not — but because the check is free and nobody ran it. Paying for a placement whose provenance was never established is paying for an unexamined assumption.
Display judged on last-click conversions. Display creates demand that gets captured elsewhere, usually by search. Measuring it on the clicks it happens to catch systematically understates it, and it is why display budgets get cut in the accounts where they were working. Measure it on demand created, or do not run it.
The pattern beneath all four: the buyer is sold abundance when the thing worth buying is provenance.
- IAB Tech Lab — ads.txt, Authorized Digital Sellers
- IAB Tech Lab — sellers.json, supply chain transparency
- IAB Tech Lab — app-ads.txt
- IAB Tech Lab — OpenRTB specification
- IAB Tech Lab — ads.cert
- IAB — industry guidelines and standards
- Google Ads Help — About Display ads and the Google Display Network
- Google Ads Help — Exclude specific webpages and videos
- Google Ads Help — Google Ads policies
- Google Ads Help — Google Ads best practices
- Google Ads Help — Different ways to track conversions
- FTC — .com Disclosures: how to make effective disclosures in digital advertising
- FTC — Advertising and marketing business guidance
- FTC — Advertising FAQs: A Guide for Small Business
Display advertising, answered against the published records
Every answer below links to the standard or platform documentation it rests on, so you can check it without taking our word for it.
How do I know my display ads actually ran where the report says?
Check the domain against the file the site publishes. ads.txt — Authorized Digital Sellers — lets a content owner declare which platforms are authorized to sell its inventory, and the IAB Tech Lab states plainly that the standard exists to eliminate the ability to profit from counterfeit inventory. Compare the platform that billed you against that list. Where it does not appear, the placement is unverifiable. That is not proof of anything wrong, and it is a good enough reason to stop buying it. The same "check the record rather than the claim" habit runs through our paid search work.
What is a made-for-advertising site and why should I care?
It is a page that exists to carry advertising rather than to be read — real URL, real load, real impression, no reason for a person to be there. Your report shows a domain and a delivery, so nothing looks wrong. The defense is provenance rather than instinct: check the seller declaration, look at whether the page has editorial content and an audience, and exclude what fails. Google documents how to exclude specific webpages and apps once you have decided. This is the same silent-failure shape described on our maintenance page — everything returns a healthy result while the outcome is worthless.
Is display advertising just brand awareness, or does it drive leads?
Mostly the former, and that is not a criticism. Display reaches people before they are searching, which is exactly when a home services business is invisible to intent-based channels. Google's own description of the Display Network frames it around reach and audience rather than capture. The mistake is judging it on last-click conversions, which understates it systematically. Measure it against branded search volume and direct traffic across flight windows instead, the way we describe on the measurement page.
Our display campaign has thousands of placements. Is that normal?
It is common and it is rarely deliberate. Broad targeting plus automated placement produces enormous domain lists, most of which nobody has examined. The number itself is not the problem; the absence of an exclusion list beside it is, because it means the length was never a decision. Start by checking the highest-spending domains against their published seller files and work down. The sellers.json standard helps where the selling platform is unfamiliar. It is the same principle as reading every charged query rather than sampling, described on our PPC page.
What about ads inside mobile apps?
Same standard, different file. The app-ads.txt specification extends the authorized-seller declaration to applications, resolved through the developer URL listed in the app store. In-app inventory usually gets the least scrutiny precisely because it is hardest to look at directly, which is an argument for checking it more rather than less. Hold it to the same bar as web inventory and exclude what has nothing published behind it. Where the destination matters as much as the placement, see landing pages.
Do we need a separate fraud-detection vendor?
Not to do the work described on this page. The seller declarations are published on the open web and readable without any account or vendor relationship, and the exclusion controls are inside the ad account you already have. A verification vendor can add signals beyond provenance, and at larger budgets that may be worth buying — but nobody should be selling you a tool to do the free check they have not been running. Ask first whether the basic check has been done, using the industry's own published guidelines as the reference. We take the same line on tooling in CRM work.
What is the difference between display and remarketing?
Display reaches people who have not encountered you; remarketing returns to people who already have. They often run on the same networks and use the same creative sizes, which is why they get conflated, but the audience question is completely different and so is the discipline. Remarketing carries obligations about the signals a browser sends regarding being left alone, which is its own subject — see our remarketing page. Provenance checking applies to both, using the same published seller declarations.
Does a verified placement mean a real person saw my ad?
No, and we are careful not to let one stand in for the other. Provenance establishes that the parties in the transaction were who they claimed to be and that the site authorized the sale. Whether the advertisement was visible on screen long enough for a person to register it is a separate question with separate measurement. The technical specifications that carry supply chain information do not answer it. Treat any agency that blurs those two things the way you would treat one that reports platform conversions without reconciling them, as described on our paid search page.
Who owns the exclusion list if we change agencies?
You should. The list is built entirely from your own billed placements, so it describes your spend rather than an agency's method, and rebuilding it means paying again for findings you already paid for. Ask where it lives and whether the account sits under your own billing and administrative access — Google's own policy documentation covers the account requirements. We build in the partner's account as a matter of course, which is the position we take on data generally, including in the CRM.
Is display worth it for a local service business?
Sometimes, and we will tell you when it is not. Display earns its place when demand for your service is being created rather than merely captured — a new market, a new service line, a seasonal push ahead of the season. It is the wrong first purchase when intent-based demand already exists and goes unanswered, because awareness spending cannot fix an unanswered phone. Claims in display creative carry the same substantiation expectation as anywhere else, covered in FTC guidance for small business. Start with an honest read of where demand sits, which is what the A.R.C. Report is for.
Find out where your display budget has actually been going.
The A.R.C. Report covers your whole marketing position, and where display is in the plan we pull the placement report, check the billed domains against the seller files those sites publish, and tell you what share of the spend went somewhere nobody can vouch for. Findings are yours whether or not we work together.
- Billed domains checked against each site's published seller declaration
- Unfamiliar selling platforms traced to a declared identity
- App placements held to the same standard as web placements
- Geographic targeting checked against the area you actually serve
- An exclusion list you keep, with the reason recorded for each line
- A straight answer if display is the wrong spend for you right now
Explore the wider program: all services, paid search, remarketing, local services ads and the A.R.C. Report.
Tell us the campaign and the markets you serve, and we will tell you what the display spend is buying before you change anything.
No cost, no commitment. We will follow up by email or phone to walk you through the findings.

