What this checkpoint covers
Klaviran Rooflights, an invented maker of rooflights for home extensions, sells flat and pitched rooflights online and quotes for bespoke lanterns, walk-on rooflights, skylights and replacement glazing. The questions apply the guide to bidding with few conversions, the guide to bidding through sales and tracking outages and the guide to scaling Google Ads spend.
Klaviran Rooflights: the account, the order records and a budget step
Quotation campaigns last quarter, each on its own target CPA and limited by low conversion data: lanterns 6 quote requests, walk-on rooflights 4 and skylights 5, all at £150; replacement glazing 9, at £40.
| Online campaign | Target / actual ROAS | Status | Lost share: budget / rank |
|---|---|---|---|
| Flat rooflights | 600% / 640% | Limited by budget | 31% / 9% |
| Pitched rooflights | 800% / 910% | Underspending | 2% / 46% |
| Standard range (Shopping) | 600% / 620% | Eligible | 4% / 22% |
One workshop builds the bespoke products, at about £2,900 margin an order, from one monthly budget split between them; a glazing order earns about £260. Online orders break even at a ROAS of 400%. Over two quarters, 440 specification sheet downloads led to 4 orders, 56 plan uploads to 14, at a similar rate for each product, and 28 quote requests to 12.
For question seven, six weeks either side of a flat rooflights budget rise, read once delayed conversions arrived: its spend rose from £5,000 to £7,000 and its value from £32,000 to £39,200, while Shopping's value fell from £50,000 to £48,600 at an unchanged £8,100.
Task
- Write a decision and a reason for every question before reading on; an answer counts only if the reason matches.
- Reread the guide section linked from each miss.
Questions
- The bespoke campaigns all show limited status. The agency proposes tightening each target so the strategies "try harder". What comes first?
- What should the bespoke strategy then bid towards? The agency suggests specification sheet downloads, the most frequent action.
- The agency proposes one target CPA portfolio across the bespoke work and replacement glazing, to pool data. Does it help?
- A trade weekend discount lifted conversion rate sharply last year and ran out of budget; spring lifts it gradually. The proposal: a seasonality adjustment for all of spring, a lower target ROAS for the weekend. What changes?
- A checkout release makes the purchase tag fire twice on some orders. The agency calls it harmless; the owner wants to merge the bespoke campaigns that week. What happens?
- The owner proposes raising every budget by the same proportion and launching Performance Max. Where is the headroom, in order?
- The agency called the flat rooflights budget step a success a week after it, on the campaign's own ROAS. Keep it?
Answers explained
- Consolidate. The three share a goal, a workshop, a margin and one budget, so only theme separates them: one campaign, an ad group per product. A tighter target adds no data and shrinks the auctions entered. See consolidating before you automate.
- Plan uploads beside quote requests, with orders imported. Downloads rarely become orders, so bidding on them buys useless conversions. Uploads sit close to the sale and convert alike across products; orders stay a secondary action so drift shows. See a more frequent bidding signal.
- No. A portfolio pools campaigns that share a goal and a margin, at the price of one target. Glazing earns a fraction of a bespoke order, so one target suits neither. It would fit bespoke campaigns kept apart only for team budgets. See pooling in a portfolio.
- Adjust the weekend, not spring. The short, dated weekend suits an adjustment estimated from last year's rate change, plus a separate budget rise, since adjustments move bids only. The target stays put. Smart BiddingGoogle Ads' automated bid strategies that set each auction's bid from conversion signals. learns a gradual season itself, so spring needs budget, not an adjustment. See seasonality adjustments.
- Treat it as an outage. A false rise in conversion rate makes the model overbid. Fix and verify the tag, exclude the window where Google documents it for the strategy, hold budgets, targets and the merge, and record the window's start and end, as the tracking outage response SOP does. See changes to hold back.
- Flat, then pitched, then new campaign types. Flat rooflights lose Impression ShareShare of eligible impressions you received. to budget above break-even, so budget rises in steps. Pitched underspends under a strict target, so the target loosens in steps, never past break-even. Wider queries follow, Performance Max last, one lever at a time. See where headroom comes from.
- No. A week is too soon, because Conversion LagThe time between an ad interaction and the conversion it leads to, which leaves the most recent reporting days incomplete. fills recent days in, and the campaign's average flatters the step. Judged by Marginal ROASExtra revenue returned by the last increment of ad spend. at account level, it loses money. See judging each step at the margin.
Answer seven: the budget step at the margin
Campaign: extra value £7,200 for extra spend £2,000, a marginal ROAS of 360%, below the 400% break-even, while the campaign's average after the step reads 560%. Account: Shopping lost £1,400 of value at unchanged spend, so the account gained £5,800 for £2,000, a marginal ROAS of 290%. Return the budget to its earlier level, or test a smaller step from there.
Where to go next
Misses on the first five send you back to the garden building maker exercise; misses on the last two, to the budget tests of the advanced Google Ads level, before the piano retailer reconciliation exercise.
