Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Performance Marketing › Intermediate

Checkpoint: Performance Targets and Bidding

Scenario questions on choosing a target type, deciding when to hand bidding to automation, reading conversion volume and setting a return floor from margin. Each answer gives the reasoning and links back to the page that teaches it, so a miss leads straight to the right reading.
CheckpointIntermediate levelModule 1: Targets and bidding4 minUpdated

What this checkpoint covers

The questions draw on the target CPA versus target ROAS comparison, the manual versus automated bidding comparison and the break-even ROAS definition. Each describes a situation and asks what you would do and why.

Task

Write a short answer to each question before opening the explanations. It counts as right only if your reason matches. For each miss, note the signal you overlooked and reread the linked section. If several went astray, redo the eyewear target-setting exercise.

Questions

The fifth question uses these figures.

Illustrative example — not a real result

Question five: a garden retailer's two lines

LineContribution marginReported ROAS
Seeds and bulbs50%3.0
Garden furniture20%4.0

The account reports a blended ROAS of 3.6, and the owner reads that as a healthy result. Both lines share one target, and margins are after product cost, delivery, fees and returns, as the finance lead calculated them.

  1. A tutoring company collects enquiry forms. Every enquiry goes to the same adviser and contract values are only known weeks later. The marketer wants target ROAS. What do you advise?
  2. An online homeware shop is about to move a campaign from target CPA to target ROAS. Its conversion values include tax, delivery charges and orders later refunded. What should happen first?
  3. A brand-new campaign for a niche industrial supplier records only a handful of conversions a month. The account manager wants automated bidding from day one. Is that sensible?
  4. A campaign on target ROAS spends well and hits its target, but the owner wants more sales. A colleague suggests raising the target. What happens, and what would you change instead?
  5. Using the garden retailer's figures above, what is the break-even ROAS for each line, and is the blended result good news?
  6. A lead generation account has used manual bids for a year. Tracking was rebuilt last quarter to count only qualified leads, and volume is now steady. What would you do and how?
  7. A team sets its new target CPA to last quarter's average because "it is what the campaign can do". When is that a mistake?
  8. A week after switching to target ROAS, revenue has dipped. The owner wants to change the target again and move budget at the same time. What do you say?

Answers explained

  1. Stay on target CPA. Every enquiry is worth the same when it converts and no value is known, so value bidding has nothing to learn from. Set the target CPA from enquiry-to-customer rate and margin until the CRM passes back deal values (when to choose target CPA).
  2. Fix the values first. Tax, delivery and refunded orders teach the system that some conversions are worth more than the business keeps. Until values reflect what the business retains, target CPA is safer (common mistakes with target types).
  3. No. With almost no conversions to learn from, the algorithm spends cautiously or explores expensively. Start on manual bids and revisit once volume is steady. The manual versus automated bidding comparison sets out the conditions.
  4. Volume falls. A higher ROAS target tells the system to withdraw from auctions it judges less profitable, which protects margin and costs volume. If the campaign sits well above break-even, lowering the target in a small step is the way to buy more sales, checking total profit after each move (why a higher target reduces volume).
  5. Seeds sit comfortably above their floor; furniture is losing money. The floor is one divided by margin, so furniture's higher reported return still sits below its floor while seeds clear theirs, and the blend hides the loss; the working is below the answers. Set separate targets per line, as the break-even ROAS definition recommends.
  6. Test automation against manual bids. Reliable qualified-lead tracking with steady volume is the point at which automation tends to win. Test it on part of the traffic, start the target near recent actual cost and leave it alone through learning (how to switch safely).
  7. When the average is unprofitable or very profitable. If the average already exceeds what a customer is worth, copying it locks in losses; if it sits far below, a looser target may buy profitable volume. The target belongs to margin and customer value (setting targets from margin).
  8. Wait, then change one thing. A dip in the first week is expected while the strategy calibrates, and conversion lag makes recent days look worse. Changing target and budget together resets learning and leaves two causes for one effect. Agree a review date that respects the lag (learning periods).
Illustrative example — not a real result

Working for question five

Seeds and bulbs: break-even ROAS = 1 ÷ 0.50 = 2.0, against a reported 3.0, so comfortably above the floor. Garden furniture: break-even ROAS = 1 ÷ 0.20 = 5.0, against a reported 4.0, so below the floor and losing money on every attributed sale. The blended 3.6 averages a profitable line with a loss-making one and hides the loss.

Where to go next

Reread the decision table in the target type comparison or the switching section of the bidding control comparison for any misses, then continue to the second module from the intermediate performance marketing level.

Pages this lesson applies

Terms used on this page

Also related

Related