Which programmatic planning mistakes repeat, and how do you fix them?
Part of Programmatic media planning
The short answer
Programmatic planning mistakes are failures made before launch that surface later as misplaced or wasted spend: a brief nobody confirmed, a site list padded to look bigger, an old list version in the platform, geography or language left at a default, and unchecked sellers. Each has a known fix, and a review after each flight stops repeats.
Key takeaways
- Several costly delivery problems start in planning, so a post mortem traces each to the brief, the list or a setting.
- Padding a list with weak fits to reach a target size spreads spend onto the sites that matched the brief least.
- Display & Video 360 says plan edits do not change the campaign and targeting changes can take a few hours.
- In Display & Video 360, a line item with no geography targeting bids on impressions from all geographic locations.
- A mistakes log with the trigger, the impact, the fix and the owner turns one flight's errors into the next plan's checks.
Which planning mistakes show up in programmatic post mortems?
A post mortem asks where a problem entered the plan, not only what the report shows. Spend in the wrong city or a line that never spent can come from a launch setting, but the root cause can sit earlier, in the brief, the list or the handoff. Launch settings are covered in our media plan QA checklist, and fees and logs in the programmatic campaign audit guide.
| Mistake | Where it enters | How it shows up later | Fix |
|---|---|---|---|
| Brief never confirmed | Brief | Sites on topics the client never asked for | Restate the brief as fields and get a written yes before searching |
| Padded site list | List building | Spend drifts to the weakest fits | Accept a shorter list and record why |
| Stale list version | Handoff | Removed sites still serving | Compare the platform's list with the approved version |
| Geography left at a default | Campaign setup | Impressions outside the market | Set geography and location options on purpose |
| Language assumed | Brief and setup | Ads in English on French pages, or the reverse | Separate audience language from page language |
| Sellers unchecked | List building | Spend through resellers nobody chose | Read ads.txt and sellers.json before the deal |
| Planning numbers treated as controls | Campaign setup | Overspend or a flight that runs past its dates | Set budgets and flight dates where they control serving |
Why do unconfirmed briefs cause failures downstream?
A brief written in an email carries the client's words, and each reader fills the gaps differently. "Parents in Ontario" can mean French and English or English only; "avoid news" can mean all news or only crime coverage. When the planner searches before the client confirms the reading, the list encodes a guess. The fix is to restate the brief as fields (product, audience, topics wanted and avoided, countries, languages, channels) and get a written yes. Our guide to avoid topics in a media brief shows how to write exclusions.
Why do padded site lists fail?
A list can get padded because a short list looks thin in a proposal and because delivery fears are real. Google's targeting overview (opens in a new tab) warns that line items "frequently can't purchase impressions because there isn't enough inventory that matches your targeting." The tempting response is to add sites until the list reaches a round number. The sites added last can be the ones that matched the brief least. When they are large general sites with plenty of inventory, they can take a share of spend out of proportion to their fit.
Padding also multiplies hidden reach. The same Help page on app, channel and website targeting (opens in a new tab) explains that when a targeted website has a matching mobile or connected TV app, the platform "will automatically target their app's inventory" unless environment targeting says otherwise. Each padded domain can bring app inventory nobody reviewed.
Four fixes for a list that feels too short
- Forecast before adding sites. Run reach forecasting on the confirmed list first, so additions answer a measured gap rather than a feeling.
- Widen the brief, not the list. If reach is short, ask the client to approve an adjacent topic, and search that topic properly.
- Put tests on their own line. Adjacent ideas go into a separate line item with a small budget, so their spend is visible.
- Record the decision. Note in the plan that the list is short on purpose and what the forecast showed.
How do stale list versions reach a campaign?
The approved list and the list in the buying platform are separate objects, and nothing keeps them in step automatically. Display & Video 360 Help on plans (opens in a new tab) says changes to plan settings "will not affect your campaign." Targeting changes on a live line item take time: "a few hours may pass before Display & Video 360 updates your line item's bidding accordingly." A client who removes a site on Monday morning can still see it in Monday's report.
Old templates are a second route. Google's Q1 2026 update notice (opens in a new tab), dated February 20, 2026, said targeting and inventory source settings would leave the campaign and insertion order levels and be "consolidated to partner and line item level only," estimated for February 23, 2026. A handoff document that still says to set the list at insertion order level is out of date. See our site list version control guide.
Which geo and language mistakes repeat across plans?
Geography failures come from defaults. Google's targeting overview states that "if you don't set any geography targeting, your line item will bid on impressions from all geographic locations." Language failures come from treating three different things as one: the language of the page, the language of the creative and the language the platform thinks the viewer speaks. For YouTube and partners line items, Google describes language targeting as showing ads "based on the languages a user speaks," inferred from settings and signals, not from the page. Our programmatic geo targeting guide covers location settings in depth.
| Mistake | Documented behavior | Fix |
|---|---|---|
| No geography set on a line item | Bids on impressions from all geographic locations | Set the country or region on each line item |
| Postal codes uploaded without a country | Location lists accept postal codes with a two-letter country code to resolve codes shared by countries | Add US or CA beside each code |
| Proximity radius left at the default | "By default, the target radius is 20 mi" | Set the radius and unit for each location |
| Audience language confused with page language | YouTube and partners language targeting uses the languages a user speaks | Choose sites by page language and set creative to match |
| One video plan copied from the US to Canada | In-stream ads on Google TV "are only available for line items serving in the United States" | Rebuild inventory choices per country |
After launch, the geography help page (opens in a new tab) points to the Geography section of the line item Troubleshooter to check that targeted countries match the impressions that came in.
How do unchecked sellers slip into a plan?
A site name on a list says nothing about who sells it. The ads.txt 1.1 specification (opens in a new tab) defines DIRECT as an account the publisher "directly controls" and RESELLER as one the publisher has authorized another entity to control and resell. In the sellers.json specification (opens in a new tab), PUBLISHER means the inventory is on a medium owned by the named entity and the advertising system pays them directly. The fix is to read both files for each site before a deal is built and note the seller ID used, as our how to check ads.txt guide explains.
How do you build a mistakes review into each plan?
A review works when it is short, scheduled and written down. Display & Video 360's event history (opens in a new tab) records changes, lets users add notes to change descriptions and download a CSV of changes, and shows performance charts such as spend or impressions beside the change history. That gives a review its evidence: what changed, when, and what delivery did next.
Planning mistakes review, run after each flight
- The confirmed brief is attached, and each placement the client questioned has been traced to a brief field or a gap in it.
- The approved list version has been compared with the composite list downloaded from each line item.
- Delivery by country, region and city has been compared with the brief's markets.
- Page language, creative language and any language targeting have been checked against each other.
- Spend through reseller paths has been compared with the sellers chosen at planning.
- Each mistake found is logged with its trigger, impact, fix and owner.
- Each fix has become a line in the next plan's template, not only a note in the report.
What does a mistakes review look like on a real flight?
Fictional example. The company is unnamed and invented; no results or performance figures are implied.
A pest control company reviews its fall display flight in Hawaii
An unnamed residential termite and pest control company ran fall display ads in Hawaii, covering Honolulu, Hilo and Kahului, in English. The confirmed brief asked for home, garden and local news sites, and the approved list was short on purpose.
The review finds three mistakes. A trader copied a line item from a mainland campaign and left its geography at the country level, so the brief's markets were never set on that line. Two sites removed after a client call were still on the composite list, because the edit went into the plan document but not the platform. And a home improvement site had been added during setup to make the list look fuller, with no record of who approved it.
Each mistake goes into the log with an owner. The template gains three lines: set the state on each line item, compare the composite list with the approved version before launch, and route late additions through the same approval as the rest of the list.
- Geography: line item copied from a mainland campaign.
- Stale version: plan edited, platform not updated.
- Padding: one site added without approval.
- Three new template lines, each with an owner.
How does PlacementGraph keep a plan's list confirmed, unpadded and versioned?
Three of the mistakes above start before any buying platform is opened: the brief, the list size and the version.
Frequently asked questions
Who should run a planning mistakes review?
The planner who built the list and the trader who set up the campaign should run it together, since their handoff is where several mistakes enter. A lead who did not work on the flight can chair it, keeping the review on causes.
Should planning mistakes be shared with the client?
Mistakes that changed where the client's ads ran or what they spent belong in the client report, with the fix. Near misses caught before launch can stay in the team's log. Either way, the entry should say what changed in the template.
How soon after launch should a stale list be suspected?
Compare the composite list with the approved version on the day of launch and after any edit. Display & Video 360 says targeting changes on an active line item can take a few hours to reach bidding, so a removed site appearing in the same day's report is not yet proof of a stale list.
What if the same mistake appears in two flights in a row?
A repeat means the fix lived in a report instead of the process. Move it into the plan template, the handoff form or the approval step, give it an owner, and check it in the next review.
Sources
- About targeting in Display & Video 360 (opens in a new tab)Display & Video 360 Help
- App, channel, and website targeting (opens in a new tab)Display & Video 360 Help
- Create a campaign (opens in a new tab)Display & Video 360 Help
- Create a plan for a campaign (opens in a new tab)Display & Video 360 Help
- Coming soon: Display & Video 360 updates in Q1 2026 (opens in a new tab)Display & Video 360 Help
- Geography targeting (opens in a new tab)Display & Video 360 Help
- Targeting for YouTube & partners and Demand Gen line items (opens in a new tab)Display & Video 360 Help
- Event history (opens in a new tab)Display & Video 360 Help
- ads.txt specification version 1.1 (opens in a new tab)IAB Tech Lab
- sellers.json specification (opens in a new tab)IAB Tech Lab
Researched and written by PlacementGraph Research. Read how we research, source and update these pages. Corrections: diana@placementgraph.com.