The Ad Ops Handover Doc That Actually Works: What to Write Down Before You Go
An ad ops handover doc that actually works answers five questions: what needs a decision daily, what the thresholds are, who to contact for what, what's currently weird, and what to leave alone. Most handover docs answer none of them - they're a list of areas to watch ("keep an eye on revenue") rather than decisions to make, which quietly transfers a full-time job to someone who already has one.
The failure isn't laziness; it's that the person writing the doc knows too much. Years of context compress into "you know, the usual" because to them, it is usual. The covering person inherits the labels without the baselines - they know to "watch pacing" but not what behind-for-this-advertiser looks like, to "check revenue" but not which dips are Tuesday-normal. This guide gives you the five-section structure that transfers the decisions instead of the anxiety, and makes the honest argument most handover advice skips: half of what goes in a typical handover doc shouldn't be written down at all - it should be automated, because a document can't watch anything.
Why Most Handover Docs Fail
Read a typical ad ops handover doc and you'll find a list of surveillance duties: check pacing, watch revenue, keep an eye on the video fill, monitor the ad units. Every item is an area, not a decision. The covering person is being asked to replicate the author's daily watching - without the years of baselines that make the watching efficient.
That's the core failure: a handover doc can transfer decisions, but it can't transfer watching. Watching depends on knowing what normal looks like - which campaigns drift and self-correct, which revenue dips are day-of-week noise, which ad units are historically fragile. Written down, that context becomes a novel nobody reads. Left out, the covering person re-derives it from scratch, badly, while doing their own job too.
The fix is to stop writing surveillance instructions and write decision support instead - and to hand the surveillance itself to something that doesn't take vacations.
The Five Sections That Actually Transfer Coverage
A few notes on getting each right:
Daily decisions (not daily checks). "Review the flagged alerts by 10 AM and action anything red" is a decision instruction. "Check pacing" is not. If your team runs automated monitoring, this section can be three lines; if it doesn't, this section becomes the impossible part - see the next section.
Thresholds are the context-transplant. This is where the senior person's head gets written down, but only the parts that drive action: "under 90% pace with less than 10 days left = rebalance now," "eCPM down more than 15% versus the 7-day average = investigate the demand source before EOD." Numbers, not vibes.
Escalation deserves brutal honesty. The covering person's real question is "what am I allowed to break?" Answer it: what they can action alone, what needs the manager, and the short list of things that justify interrupting a beach ("anything touching these two advertisers"). Also the reverse - what looks alarming but is known-weird and should be left alone. The "currently weird" section prevents the classic coverage error: heroically fixing something that wasn't broken.
What to Automate Instead of Writing Down
Here's the part most handover advice misses: the biggest section of a traditional handover doc - the daily watching - shouldn't be in the doc at all. A document can't watch anything. It can only instruct a stretched human to watch, which is exactly the arrangement that fails every summer.
The daily sweep - is delivery on pace, did revenue move against baseline, are the ad units healthy - is systematic, repetitive, baseline-driven work. That's automation's home turf. With a monitoring layer doing the sweep and flagging deviations worst-first, the handover doc shrinks from "replicate my mornings" to "action the flagged list using the thresholds below." That's a coverage plan a person with another full-time job can actually execute.
This is precisely what ProOps Ads Tracker does for GAM publishers: the daily checks across campaigns, revenue, and inventory run automatically every morning, seven days a week, and whoever's in the seat gets the short list by 8 AM. During vacation season it's the difference between a handover doc that says "watch everything" and one that says "here's what to do when something's flagged" - the second one works. (For the fuller seasonal coverage picture, see the companion piece: vacation season ad ops.)
Keeping the Doc Alive After Vacation Season
The best handover docs aren't vacation documents - they're the team's operating manual, exercised annually. Written properly once, the five sections double as onboarding material for the next hire, sick-day coverage, and the continuity plan for the departure you don't see coming. The thresholds section in particular tends to become the team's de facto alerting spec: the numbers you'd tell a covering human are the numbers your monitoring should be flagging on year-round.
Two maintenance habits keep it honest: update "currently weird" the day something new becomes weird (not the day before the next vacation), and re-validate the thresholds each quarter against what actually triggered action. A handover doc that's six months stale is a confidence trick - it reads authoritative and misleads precisely because of it.
If writing the thresholds section surfaces that your team doesn't actually have agreed numbers - that pacing risk and revenue dips are judged by feel - that's worth fixing beyond vacation season. It's one of the four areas our free ad ops audit maps, and it's usually the highest-value one.
FAQ - Ad Ops Handover Docs
What should an ad ops handover document include?
Five sections: daily decisions (what needs a human call each day and by when), thresholds (the numbers at which watching becomes acting), an escalation map (what the covering person can action alone versus escalate), a "currently weird" list (known issues that look broken but should be left alone), and contacts with relationship context. Each section answers a question the covering person will actually face, rather than assigning surveillance duties.
Why do most ad ops handover docs fail?
Because they transfer areas to watch instead of decisions to make. "Keep an eye on revenue" asks the covering person to replicate the author's daily watching without the baselines that make it efficient - which campaigns drift normally, which dips are day-of-week noise. A document can transfer decisions and thresholds; it cannot transfer watching.
What parts of ad ops coverage should be automated rather than documented?
The daily sweep: whether delivery is on pace, whether revenue moved against baseline, whether ad units are healthy. It's systematic, repetitive, baseline-driven work - automation's home turf. With monitoring doing the sweep and flagging deviations worst-first, the handover doc shrinks from "replicate my mornings" to "action the flagged list using these thresholds," which a person with another full-time job can actually execute.
How do you write thresholds into a handover doc?
As numbers that drive action, not vibes: "under 90% pace with fewer than 10 days remaining = rebalance now," "eCPM down more than 15% versus the 7-day average = investigate before end of day." Thresholds are the transplantable part of a senior person's context - and once written, they double as the team's alerting spec year-round.
Is a handover doc only for vacations?
No - written properly, it's the team's operating manual exercised annually. The same five sections cover sick days, onboarding a new hire, and continuity through unexpected departures. Keep it honest with two habits: update the "currently weird" section the day something becomes weird, and re-validate thresholds quarterly against what actually triggered action.