What is a makegood in advertising, and when does a publisher owe one?
Part of Curated deals and PMPs
The short answer
A makegood is replacement advertising a publisher runs to fix a delivery failure it is responsible for, such as missing guaranteed impressions, omitting an ad or running ads where the order excluded them. Buyer and publisher agree its terms in writing, and if no makegood can be agreed, the standard terms allow a credit instead.
Key takeaways
- Makegoods answer failures the publisher is responsible for: under delivery of guaranteed volume, omitted ads, adjacency breaches and some reporting and ad serving failures.
- Under the 4A's and IAB standard terms, deliverables bought on CPC, CPL or CPA have no makegoods at all.
- In programmatic guaranteed, a shortfall can be the buyer's doing when its line item does not bid, so the cause decides whether a makegood is owed.
- Record every makegood with its reason, the original order or deal, the volume, dates, placements and who approved it, so the remedy can be audited later.
What is a makegood in advertising?
Definition
- Makegood
- Advertising a publisher provides, at no extra media charge, to remedy a failure to deliver what an insertion order or guaranteed deal promised. Its volume, placements and dates are agreed between buyer and publisher before it runs.
Digital buying writes the idea into contract templates. One is the 4A's and IAB Standard Terms and Conditions for Internet Advertising, Version 3.0 (opens in a new tab), which uses the makegood as the default remedy in several clauses, not only the one about under delivery.
A makegood is different from bonus impressions. Bonus delivery is extra volume a publisher chooses to give, which the same terms cap at 10 percent above the order when the agency uses its own ad server, unless the agency agrees in writing. A makegood is owed delivery that replaces something missing, so report the two apart.
When does a publisher owe a makegood?
The standard terms connect makegoods to six different situations. The table lists each clause, what triggers it and what the template says should happen. The binding terms of any campaign are the ones in the order the parties sign, which can change any of these.
| Clause | Trigger | What the template provides |
|---|---|---|
| VI(b) Makegood procedure | Delivery falls below guaranteed levels, or an ad (placement or creative unit) is omitted | Both sides try to agree a makegood flight; if none is agreed, the agency may take a credit for the under delivered portion |
| II(d) Editorial adjacencies | Ads appear next to content the order's adjacency guidelines exclude | On written request the ads are removed and makegoods given, or a credit issued, or the ads not billed; correction efforts within 24 hours of notice |
| IV(c) Reporting failure | The publisher misses the agreed time for an accurate and complete report | The agency may start makegood discussions under section VI |
| XIII(d) Discrepant measurement | The controlling count is more than 10 percent lower than the other count over the invoice period and reconciliation fails | The agency may treat the gap as under delivery, with any makegood measured by the agency's ad server |
| XIII(g) Ad server fixed | The publisher takes longer than 72 hours to resume after the agency's ad server is working again, without reasonable explanation | The publisher owes the agency a makegood |
| VIII(a) Force majeure | A delay or default beyond either side's control | The publisher recommends a substitute within five business days; without an acceptable substitute or makegood, a pro rata reduction applies |
How are makegoods handled in programmatic guaranteed deals?
Programmatic guaranteed moves the order into the platforms, so the makegood moves there too. In OpenRTB 2.6 (opens in a new tab) a guaranteed deal is marked guar 1, meaning the bidder must bid, yet the buyer's platform still has to do the bidding. Google Ad Manager's page on buyers for programmatic guaranteed (opens in a new tab) says it is the buyer's responsibility to ensure all active guaranteed deals fully deliver, and that an under delivery the publisher caused does not count against the buyer.
That makes the cause of a shortfall the first question. If the buyer's line item paused, ran out of budget or carried creatives the deal could not accept, impressions were offered and not taken, and the publisher has delivered its side. If the publisher's forecast was wrong, a site lost traffic or ads were omitted, the shortfall is the publisher's and a makegood is the expected remedy.
Both sides of one Google pairing document the mechanics. Ad Manager's makegood help (opens in a new tab) describes a replacement proposal line item the publisher creates from a sold, reopened proposal with a Display & Video 360 buyer: targeting, delivery and pacing carry over and can be updated, the CPM rate cannot be changed, the quantity cannot exceed the original line item's impressions, and only one makegood is allowed per line item. The buyer accepts it as in any negotiation, and Display & Video 360's help (opens in a new tab) says buyers can also request one from the deal with a reason.
| Aspect | What Google's help page says |
|---|---|
| Who starts it | Ad Manager: the publisher controls the process; Display & Video 360: either side can request or create one |
| Where it appears | Below the original deal in Display & Video 360, labeled Makegood |
| Media cost | No additional media cost for the makegood, only for the original deal; programmatic fees still apply, based on the original CPM |
| Volume | No more than the original line item's impressions; potential guaranteed volume shows original plus makegood |
| Eligibility | Standard line items only, end date in the future or within the last 30 days; not Sponsorship, YouTube or CPD deals |
| Limits | One makegood per line item, and no makegood of a makegood |
How should a makegood be agreed and documented?
A makegood changes the delivery a client paid for, so it needs the same paper trail as the original buy. The standard terms require changes to an order in writing, and section VI(b) bars running a makegood beyond the order's period without the agency's prior written consent. The steps below keep the remedy traceable.
Agree a makegood in five steps
- Name the failure and its cause. State which clause or deal term was missed, with the delivery report, screenshot or log that shows it, and whether the cause sits with the publisher or the buyer.
- Agree the volume against the shortfall. Use the same count that governs the order, so the makegood repairs the measured gap rather than a different number.
- Agree where and when it runs. List the sites, placements, formats and dates, and keep them inside the flight unless the agency agrees in writing to run later.
- Check the replacement inventory. Apply the client's suitability rules and site list to any placement the original order did not include.
- Report it apart. Tag makegood delivery separately from the original line in both systems, so reports show the repair instead of blending it in.
A makegood record holds
- The original insertion order number or deal ID and the line it repairs.
- The failure, the evidence and the date the publisher was told or gave notice.
- The agreed makegood volume, measured by the same controlling count as the order.
- The sites, placements, formats and dates of the makegood.
- The names of the people on each side who approved it, with the written approval attached.
- The final delivery of the makegood and the date the issue was closed.
Can a makegood run on different sites or placements?
It can, if both sides agree. The standard terms leave the conditions of a makegood flight to the parties, and section II(a) says any exception to delivering the ad to the sites specified on the order must be approved by the agency in writing. Nothing in the template forces the makegood onto the original placement.
The risk is that a convenient replacement is not equivalent. A publisher with spare inventory on another section, a network site or a smaller format may offer it first. The test is whether the client would have approved that placement in the original plan. Ask for the domains and sections by name, review them like the original list, and refuse network properties the client never saw. In Ad Manager, makegood targeting carries over from the original line item and can be updated, so check the targeting on the makegood itself before accepting.
What does a makegood look like on a real campaign?
Fictional example
A theme park's season pass campaign in Pennsylvania and New Jersey
An unnamed amusement park sells summer season passes and books a programmatic guaranteed deal with a regional news and family events publisher, covering readers in Pennsylvania and New Jersey, in English, for a six week spring flight. The agency's buying platform targets the deal with display and video creatives.
Halfway through, the publisher's delivery report and the agency's report both show the deal behind its daily pace. The agency checks its side first: the line item is active, has budget and bids on every request. The publisher finds that a site redesign removed one of the agreed placements for nine days, so the shortfall is its responsibility.
The two sides agree a makegood for the missing volume on the same sites and the restored placement, ending on the original end date. The agency accepts it in its platform, tags it as a makegood in reporting and files the email approvals, the delivery reports and the publisher's notice in the campaign record.
How does PlacementGraph help when a makegood needs new inventory?
Frequently asked questions
Is a makegood the same as a credit or a refund?
No. A makegood repays the shortfall in advertising. A credit reduces what the agency owes for the under delivered portion, and under the standard terms it applies when no makegood can be agreed. A refund is possible there only for a cash prepayment made for that specific order, with the agency current on other amounts owed.
Does a makegood apply to open auction or private auction buying?
Not in the usual sense. A makegood repairs a guarantee, and open auction and private auction deals guarantee no volume. A seller might still offer something to fix an error, such as an adjacency problem, but that is a goodwill arrangement agreed case by case, not a remedy the deal type promises.
Who decides whether a shortfall is the publisher's fault?
The parties, using the order's terms and the delivery evidence. Compare the controlling count with the guarantee, then check the buyer's side for paused line items, spent budgets or rejected creatives. In Google Ad Manager's programmatic guaranteed rules, a shortfall caused by the publisher does not count against the buyer's service level agreement.
Can a makegood extend past the campaign end date?
Only with the agency's prior written consent under section VI(b) of the standard terms. For a seasonal or dated message, delivery after the end date may be worth little, so agree whether a later makegood is useful before accepting one, or take the credit the terms allow instead.
Sources
- Standard Terms and Conditions for Internet Advertising for Media Buys One Year or Less, Version 3.0 (opens in a new tab)4A's and IAB
- Makegoods for Programmatic Guaranteed deals (opens in a new tab)Google, Display & Video 360 Help
- Give buyers Makegoods (opens in a new tab)Google, Ad Manager Help
- DSP partners for Programmatic Guaranteed (opens in a new tab)Google, Ad Manager Help
- OpenRTB 2.6 specification (opens in a new tab)IAB Tech Lab, OpenRTB repository
Researched and written by PlacementGraph Research. Read how we research, source and update these pages. Corrections: diana@placementgraph.com.