What this checkpoint covers
The questions draw on the cross-channel performance planning guide, the marginal returns and forecasting guide and the media planning framework.
Task
Write a short answer to each question, including the reason, before opening the explanations. An answer counts only if the reasoning matches. For each miss, note what you overlooked and reread the linked section. If you missed the arithmetic question, rework the language school budget exercise.
Questions
The third question uses these figures.
Question three: a florist's paid social budget steps
| Monthly spend | Orders | Average cost per order |
|---|---|---|
| £3,000 | 120 | £25 |
| £4,000 | 140 | £29 |
Target cost per order, from margin: £40. Both months had similar seasonality and no promotions, the same creative ran throughout, and orders are counted in the florist's own order system rather than the platform. The owner wants to know whether to raise spend again next month, to a higher level not yet tried.
- A finance director proposes carrying last year's channel split forward and moving a little budget from the channel with the worst reported return to the one with the best. What is the risk?
- Paid social has been judged on the same last-click cost target as search for a year and has been cut each quarter. Branded search volume is now drifting down. How do you connect the two, and what changes?
- Using the florist's figures above, should the budget rise beyond the higher level?
- A planner forecasts next quarter by multiplying a larger budget by this quarter's average return. Finance doubts the number. Who is right, and why?
- Each channel in a plan hits a target equal to the blended target, yet the blended cost per new customer from the business's own records misses. Why, and what do you change?
- A video campaign planned to build ReachUnique people who saw an ad. before a seasonal peak is booked to start in the peak week, when prices are highest. What would you change?
- A remarketing line reports the best return in the plan and the team wants to double it. What do you check first, and what limits its size?
- A team moves budget between channels every week based on the previous week's results. What is wrong with that rhythm?
Answers explained
- It encodes attribution bias. Channels that harvest demand report well and grow until they saturate, while channels that create demand report badly and shrink, and the demand they fed dries up. Start from roles and evidence instead (a plan is a set of bets).
- Demand capture draws on demand the other channels create. Judging a creation channel on last click guarantees it is cut, and the effect shows up later as falling search volume. Give social its own measure, such as cost per new customer confirmed in the business's data, and test what it adds (assigning channel roles).
- No. The last step of spend already costs more per extra order than the target allows, even though the average still looks comfortable; the working is below the answers. Hold at or below the higher level, find where the marginal cost meets the target, and move the rest elsewhere (marginal cost per acquisition).
- Finance is right. The next share of spend performs like the margin, not the average, so the forecast overstates what the increase buys. Build scenarios with diminishing returns and a stated marginal cost (forecasting outcomes from spend scenarios).
- Channels overlap. Every platform claims conversions others also claim, so channel results add up to more than the business saw. Compare channel-reported results with actual new customers and tighten channel targets by the observed overlap, so the blended CACCost to acquire one new customer. holds (targets per channel).
- Start it earlier. Demand creation needs to lead demand; reach bought in the weeks before a peak is usually cheaper and arrives in time to influence the purchase. The media planning framework treats demand seasonality and media seasonality as separate calendars.
- Incrementality and audience size. Much of what remarketing reports would have happened anyway, so it needs the strictest target and a holdout test. Its size is capped by the audience it draws on, and doubling spend into a small audience mostly raises FrequencyAverage times a person saw an ad. (allocating budget by role).
- Weekly moves are reactions, not decisions. Inter-channel changes belong to a monthly review; weekly changes belong inside channels. Weekly swings also reset automated bidding and make each channel's curve unreadable (common planning mistakes).
Working for question three
Marginal cost per order = (£4,000 − £3,000) ÷ (140 − 120) = £1,000 ÷ 20 = £50. That is above the £40 target, while the average at the higher level is only £29. The step from £3,000 to £4,000 therefore reduced profit, and a further step would almost certainly cost more again per extra order, because returns diminish as spend rises.
Where to go next
If you missed the role and overlap questions, reread the cross-channel planning guide's sections on roles and targets. If you missed the arithmetic or forecasting questions, reread the marginal returns guide. Then move on to the third module from the intermediate performance marketing level.
