Budget reallocation in paid media, without breaking the plan

When a reallocation is actually being asked for

The request rarely arrives as “please reallocate”. It arrives as “move some budget to the thing that’s working”, “can we trim the awareness line”, or “finance wants 20% back”. Each of those is a reallocation, and each one has the same three questions underneath it.

What does each line need to run at all? Every account has a floor: the daily spend below which a line stops exiting learning, stops delivering, or delivers numbers too thin to read. The floor comes from the account’s own history, never from a benchmark.
What does the move do to the targets? If the plan promised a number per market or per stage, the move changes that number. The report in three weeks will show it whether or not you said it now.
Whose decision is it? Moving money inside a line you own is yours. Moving money between things the client approved is theirs, and the honest way to ask is with the arithmetic already done.

A worked example

Fictional. Illustrative figures that add up; not benchmarks. One market, two platforms, eight weeks, $40,000 net working media. Week three. The client’s finance lead asks you to move 20% of what is left “from Meta to Search, since Search is the one converting”.

Line	Weekly budget	Daily	Floor (own history)	Cost per enquiry, weeks 1 to 2
Search, brand and category	$2,000	$286	$150	$38
Meta, prospecting	$3,000	$429	$400	$91
Total	$5,000	$714

Five weeks remain: $25,000. Twenty percent of that is $5,000, or $1,000 a week moving from Meta to Search.

What the move does to each line

Line	After the move	Daily	Floor	Verdict
Meta	$2,000 a week	$286	$400	Below the floor
Search	$3,000 a week	$429	$150	Above it, on paper

The Meta line does not become a smaller Meta line. At $286 a day against a $400 floor it becomes a line that spends money and produces nothing you can read, which is worse than either keeping it or switching it off. So the real choice is not “20% to Search”. It is Meta whole, or Meta off.

What the move does to Search

Search converts at $38 because it catches people already looking. There are only so many of them. The account’s own impression share says the brand and category terms already show for 78% of the searches available; the remaining 22% is mostly the expensive end of the auction. Put $1,000 more a week into that and the cost per enquiry does not stay at $38. The honest forecast, from the account’s own history of what happened the last time Search was pushed, is $55 to $70 on the increment.

The three options that are real

Option	Search	Meta	Forecast enquiries, five weeks	What you are promising
A. As asked: move $1,000 a week	$3,000	$2,000, below floor	Search about 300, Meta unreadable	A Meta line that cannot be judged
B. Keep the plan	$2,000	$3,000	Search about 265, Meta about 165, total about 430	The plan as approved
C. Meta off, Search up to its useful limit, the rest returned	$2,600	$0	Search about 310, the rest of the money back	Fewer enquiries, lower cost, $2,400 a week unspent

Which of B and C is right depends on something the table cannot tell you: whether the client needs volume or needs efficiency this quarter. That is why the reply below asks, instead of choosing quietly.

The reply to send

“Moving $1,000 a week from Meta to Search takes Meta from $429 a day to $286 against a floor of $400 on this account, so it would stop producing readable numbers rather than fewer of them. Search is already showing for 78% of the searches available, and the increment would cost roughly $55 to $70 an enquiry against $38 today. Two options hold up: keep the plan as approved, forecast about 430 enquiries over the five weeks; or switch Meta off, take Search to about $2,600 a week where it still earns, and return about $2,400 a week. Fewer enquiries, better cost. Which matters more this quarter? I’ll move the money the same day you tell me.”

One rule under all of it: a reallocation that pushes a line under its floor is a cut, so call it one. And a reallocation that changes what the plan can deliver is the client’s decision, made with the numbers in front of them.

What would change the answer

If Meta’s floor were $200, the move as asked would work, and the only question left is Search’s limit.
If Search’s impression share were 40%, there is headroom, and the increment might hold near $38. Check the account before you assume either way.
If the client’s objective were awareness rather than enquiries, Meta is not the line that is underperforming; it is the line doing the job the plan gave it.

Now make the call yourself

The free case puts you in a different reallocation: a 30% budget cut two days before launch, four markets, lead targets per market, and the plan’s own minimums on the table. You commit to what you change first, then read the reasoning against your call. It is not this example again, so nothing above spoils it.

Try the free case

About four minutes. No signup.

A media plan example, every calculation shown: where the floors and the forecast above come from in the first place.
Client reporting for paid media: what to send the morning after a reallocation shows up in the numbers.
Media, 2027 Edition: 38 cases, including a budget that tripled while the return fell, and a forecast that went stale after a cut. You commit before the reasoning opens.

Budget reallocation, without breaking the plan.