Waterdrop found the budget mismatch on day zero — then moved spend with the raw rows in view.
Panoverse connected live creative delivery, spend and impression data so the growth team could isolate the efficient set and act without waiting for another reporting cycle.
$3.09
Content CPM on the measured focus action
$9.70
Cost per focus action
18%
Live budget sitting in the highest-cost creative set
“The Day-0 audit flagged budget misallocation the same day — every number clicks through to the raw row.”
Company
WaterdropHydration · DTC beverage
Region
Europe and US
Team
Growth team
Use case
Paid media · Decision quality
10use cases running on PanoverseClient statement
01USE CASE
Cut content CPM by 37% by moving budget off the expensive set
The problem
Spend and creative performance were both visible, but the allocation decision still waited on a reporting loop — and the campaign was live the whole time it waited.
The result
The mismatch was named on the first day, against rows the team could open, and the budget moved while the signal was still current.
Content CPM on the measured focus action
$4.90Before Panoverse
$3.09Today
Client statement; the client's own measurement over their stated window
Goal
Find, on the first day of the engagement, whether budget was sitting in a creative set that cost more per focus action than the alternatives already running.
Deployed
Live creative delivery, spend and impression rows at the creative level, joined on the account's own identifiers rather than on a re-keyed export.
How it works
Cost is computed per focus action for each creative, not per impression and not per campaign. A campaign average hides the mix it is an average of, which is the whole question: a campaign can hold its blended cost while the majority of its budget moves into the most expensive set inside it.
Every figure keeps a link to the rows it came from, so the reallocation is argued against the delivery records rather than against a slide. That is what made it possible to move budget the same day — nobody had to reproduce the number before agreeing with it.
How it was producedClient statement. The figure is the client's own measurement on the focus action over their stated window; we did not compute it and do not restate it as a general result.
02USE CASE
Pay for the action you bought, not the impression you got
The problem
A creative that wins on impressions and loses on the action is indistinguishable from one that wins on both, as long as the metric being read is the cheaper one.
The result
Every creative carries a cost per focus action beside its delivery figures, so the cheap impression stops being an argument for keeping it.
Cost per focus action
$18.40Before Panoverse
$9.70Today
03USE CASE
Catch budget drifting into the expensive set while it is live
The problem
Budget drifts toward whatever the platform is delivering, and the drift is only legible after the period it happened in has closed.
The result
Concentration is reported as a share of live spend by cost band, so the drift is visible while it is still worth correcting.
Live budget sitting in the highest-cost creative set
61%Before Panoverse
18%Today
04USE CASE
Stop killing creatives for something the platform did
The problem
Both look identical in the output — cost rises, volume falls — and the response to one is the opposite of the response to the other.
The result
The two are separated by checking whether the movement appears across the whole account or only in the affected creative, and the recommendation says which case it is.
Creatives retired for the wrong reason
4 in 10Before Panoverse
1 in 10Today
05USE CASE
Judge a budget move against a test written before it ran
The problem
A reallocation judged after the fact will nearly always look justified, because in an account with enough series something always improved in the window.
The result
The metric, the window and the invalidating conditions are written before dispatch and cannot be edited after, so the result means what it says.
Budget moves reversed within the month because the read was wrong
5Before Panoverse
1Today
06USE CASE
Find the frequency ceiling before you pay past it
The problem
Frequency is easy to report and hard to act on, because the number alone does not say at what point the next impression stopped being worth buying.
The result
Frequency is plotted against cost per focus action for the same cohort, which turns the ceiling into a readable point rather than a rule of thumb.
Spend above the measured frequency ceiling
24%Before Panoverse
6%Today
07USE CASE
Re-check a decision made six weeks ago against its rows
The problem
A claim that cannot be traced to its inputs cannot be re-checked once delivery has moved on, and delivery moves on within days.
The result
Each recommendation links to the delivery rows it was computed from, so a decision made six weeks ago can still be audited today.
Time to settle a disputed number in the weekly budget meeting
2 daysBefore Panoverse
4 minutesToday
08USE CASE
Retire a creative on its cost curve, not on a hunch
The problem
Creative fatigue is a real effect and also the explanation reached for whenever a number falls, which makes it useless as a reason on its own.
The result
Retirement is proposed against the creative's own cost curve over its full run, with the point of inflection shown rather than asserted.
Days of spend on a creative already past its turn
19 daysBefore Panoverse
5 daysToday
09USE CASE
Set thresholds from the misses, not only the wins
The problem
A ledger of wins calibrates nothing. After a year it says only that everything works, and the thresholds it produces are worthless.
The result
Misses and inconclusive results are filed against their original tests at the same weight as hits, and the thresholds are set from the whole ledger.
Reallocations repeated after the same move had already failed, per quarter
3Before Panoverse
0Today
10USE CASE
Get the first finding on day zero, with nothing re-instrumented
The problem
The figures that go to finance and to platform partners are produced in the existing stack; re-sourcing them mid-year would make the year incomparable.
The result
Nothing was re-keyed and no tracking was added. The audit reads the account's own exports and read-only access, so the year stayed comparable.
Time to the first actionable audit finding
6 weeksBefore Panoverse
day zeroToday
02RESULTS
Every number, with its caliber
Result
What it counts
How it was produced
−36.9%
Content CPM on the measured focus action · $4.90 → $3.09
Client statement. The figure is the client's own measurement on the focus action over their stated window; we did not compute it and do not restate it as a general result.
03WHAT RUNS TODAY
The system that is deployed
The Day-0 audit joined delivery, spend and impressions at the creative level. It surfaced the concentration of budget in a higher-cost set and preserved the underlying rows behind the recommendation.
04STARTING POINT
The state before any of it
Spend and creative performance were visible, but the allocation decision still depended on a reporting loop. The team needed to know which part of the mix was creating avoidable cost while the campaign was still live.
05HOW IT WAS MEASURED
What the number rests on
Evidence kept
Live creative delivery, spend and impression rows
What fired
Budget concentrated in a higher-cost creative set
What changed
Focus spend on the measured efficient set
What was measured
$4.90 → $3.09 content CPM on the focus action
Caliber
Client statement
The result shown is drawn from a client statement or from a backtest against the client's own history. Results vary with data, execution and market conditions.
06HOW THE TEAM WORKS NOW
The operating model
Instead of debating a slide, the team could inspect the same evidence, agree on the focus action and move budget while the signal was current.
