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

E-commerce Marketing › Advanced

Checkpoint: Scaling a Catalogue Across Channels

Scenario questions set in an invented online toy retailer test whether you can structure a large catalogue across paid and organic channels. They cover why one campaign favours a few best sellers, splitting by margin, role and stock, separating prospecting from dynamic retargeting, keeping brand and existing customers from flattering results, choosing indexable filter pages and reading product reports, with each answer explained.
CheckpointAdvanced levelModule 2: Scaling the catalogue across channels4 minUpdated

What this checkpoint covers

Linnet Toys is an invented online toy retailer with a large catalogue, one Performance Max campaign, one Meta catalogue ad set and a filtered category navigation. The questions draw on the Performance Max structure guide, the Meta catalogue ads guide and the faceted navigation guide.

Illustrative example — not a real result

Linnet Toys: the account and its product report

  • 3,000 products in one Performance Max campaign at a target return of 4.0. Over the last 30 days, 40 products took 70% of spend and 1,900 had no impressions.
  • Search term insights put Linnet brand terms at 25% of Performance Max conversion value; returning customers placed 55% of its orders.
ProductMarginSpendReturnStock
Wooden train set45%£2,4003.2Deep
Licensed character figure15%£3,1005.9Deep
Craft kit50%£900No salesDeep
Outdoor play tent38%£1,2004.440 units, six weeks before the gift peak
Jigsaw puzzles, 120 products42%£0NoneDeep

Task

  • Answer every question in a sentence or two, with the reason, before reading any explanation.
  • Mark each answer right, partly right or wrong, and reread the linked guide for anything you missed.
  • Where an answer changes your design, revise the sports nutrition catalogue exercise.

Questions

  1. Why does one campaign over the whole catalogue concentrate spend on a few best sellers, and when is that a problem?
  2. The trading manager proposes one campaign per toy category. What would you do instead?
  3. The Meta ad set shows the full catalogue to past visitors and new people together, and its return looks excellent. What would you change?
  4. Since Performance Max launched, the brand search campaign has shrunk while Performance Max reports a high return. How do you find out what it really adds?
  5. The category can be filtered by age, material, price, colour and character. Which filter views would you build as indexable pages?
  6. Using the product report, which products would you move, and where?
  7. The play tent is a strong seller with little stock left before the peak. What happens to it in each channel?

Answers explained

Illustrative example — not a real result

Linnet Toys: break-even by product

Break-even is 1 ÷ margin. Wooden train set 1 ÷ 0.45 ≈ 2.2, so 3.2 clears it. Character figure 1 ÷ 0.15 ≈ 6.7, so 5.9 loses money despite the best return in the report. Play tent 1 ÷ 0.38 ≈ 2.6. Jigsaws ≈ 2.4.

The figure's £3,100 bought about £18,300 of revenue and £2,740 of contribution, about £360 short of its ad cost. The train set's £2,400 bought about £7,700 of revenue and £3,500 of contribution.

  1. Bidding follows predicted conversions, and best sellers have the history. The campaign keeps buying what it knows converts, and the tail never earns impressions. That is a problem when the favoured products carry thin margins or brand demand, because one target hides both. See how many campaigns.
  2. Split by target and role, not category. Categories share margins and goals, so splitting them only starves campaigns. Group by margin band, role and stock with custom labels, and use listing groups and product reports for visibility. See split by target, not taxonomy.
  3. Separate the jobs. Prospecting gets a product set of proven, high-margin, well-stocked toys and excludes customers and recent visitors; a Remarketing ladder keeps the full catalogue by intent level. Judge prospecting on new-customer cost and margin, since past visitors inflate the blend. See product sets.
  4. Exclude the brand and read the account. Attach a brand exclusion, return brand queries to search, set the customer acquisition goal and report new and returning customers separately. Judge the change on account totals, because ROAS moves between campaigns when brand traffic does. See brand exclusions.
  5. Age and material, plus their pairs where demand and range allow. Parents search by age and by wooden toys; nobody searches by colour, and price bands rarely deserve pages. Character pages may qualify if the range is deep. Colour, sort and deep combinations take a Canonical Tag to the nearest indexable page or stay out of the crawl. See the decision framework.
  6. Move the figure, check the kit, find the jigsaws. The figure needs a target above its break-even or the add-on group. The craft kit's spend with no sales is usually a feed problem first. The jigsaws earn nothing because nothing reaches them: check feed attributes and listing groups before giving them a target. The train set has room for more reach. See reading product-level reports.
  7. It leaves acquisition before it sells out. Remove it from prospecting sets and the listing group that buys new customers, and keep it in retargeting while stock lasts. Promoting a toy that sells out before the click wastes the visit. See stock depth in product sets.

Where to go next

If the structure questions caught you out, redo the sports nutrition catalogue exercise. The next module starts with the house plant cohort exercise, which asks what the customers these campaigns win are worth. For feed and shopping foundations, return to the intermediate e-commerce level.

Pages this lesson applies

Terms used on this page

Also related

Related