Media Planning
Part of Media budgets and allocation
Allocating budget between learning and established channels
Set a learning allocation that can answer a useful question while protecting justified established activity.
Keep funding established channels while current evidence supports their job. Allocate money to a new channel only when the amount can answer a defined question and its possible loss fits within the campaign limit. There is no useful universal percentage split.
Establish the amount that can move
Begin with the approved campaign ceiling and subtract booked commitments and essential supporting work. Identify the established activity needed to serve demand the business can fulfil. The remaining flexible amount is the upper boundary for learning; it is not automatically the amount to spend.
An established channel needs evidence relevant to the current offer, audience, outcome and period. Last year’s cost per lead may be a starting point, but changes in creative, inventory or lead qualification can weaken the comparison. Record which assumptions still hold before protecting its allocation.
Price the learning question
State the decision the trial should inform. For example: could a new explanatory format produce accepted enquiries in the service area at a cost the business could support? Specify the placement, creative, destination, outcome record and review date. Estimate the media and supporting costs using a current quote or a clearly bounded forecast.
Set two limits: the least spend that could produce evidence useful for that decision, and the most the business can afford to lose without disrupting necessary activity. Neither number is universal. If the useful trial would cost more than the acceptable loss, narrow the question, seek a cheaper preliminary check or defer it. A small allocation that merely establishes a presence is not a meaningful trial.
| Allocation | Approval question | Review point |
|---|---|---|
| Established activity | What current evidence and business need justify this amount? | When delivery, cost or outcomes change enough to affect the case |
| Learning activity | What question can this amount answer? | When the agreed evidence or stop condition is reached |
| Uncommitted amount | What justified change may need funding? | When an opportunity or delivery problem is documented |
Keep booked money separate from the uncommitted amount. These are planning categories, not prescribed shares.
Established vs Learning Budget Allocation: Key Differences
- Purpose
- Sustain proven performance; meet current business demand
- Decision Driver
- Current evidence and business need
- Review Trigger
- When delivery, cost or outcomes change enough to affect the case
- Purpose
- Answer a specific, defined question about a new channel or format
- Decision Driver
- What question can this amount answer?
- Review Trigger
- When agreed evidence or stop condition is reached
How to Allocate Budget for Learning Trials
- Establish the flexible budgetSubtract booked commitments and essential work from the campaign ceiling
- Define the learning questionSpecify placement, creative, destination, outcome and review date
- Set spending limitsDetermine minimum useful spend and maximum acceptable loss
- Execute with clear measurementUse Google Ads experiments for valid comparison designs
- Review and decideIncrease, repeat, maintain or stop based on evidence
Judge the trial on its question
Check early whether the activity served the intended market, carried the approved message and collected the agreed outcome. Sparse early results should be shown with their spend and raw counts, not compared uncritically with a mature channel’s long-run average. Align outcome definitions, cost coverage and periods before making a cost comparison.
A question about whether a change caused additional outcomes needs an appropriate comparison design. Google Ads Help documents experiment setup; review the current documentation to understand what a proposed experiment measures before budgeting for one.
Key Metrics for Evaluating Learning Trials
- Spend to Date
- Not applicable – use raw counts with early results
- Market Reach
- Check if activity served intended audience
- Message Delivery
- Confirm approved creative was used
- Outcome Collection
- Ensure agreed outcome was recorded correctly
Release the next amount deliberately
Approve an initial amount and decision date. At review, increase the allocation, repeat with a corrected setup, maintain a bounded amount or stop. Record the evidence and the spend governed by that choice. A channel becomes established only when its evidence remains relevant to the conditions in which it will be used; a material offer or measurement change may require a fresh learning decision.
