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

Content Marketing › Intermediate

Diagnosing Underperforming Articles for a Toy Retailer

Practice reading search and analytics data together to find out where each article fails. You take an invented toy retailer's blog extract, separate discovery, engagement and contribution problems, and give every article one action with a reason before any rewriting starts.
ExerciseIntermediate levelModule 3: Measuring and improving content5 minUpdated

Why this matters

"The blog is not working" is not a diagnosis. The measuring content performance guide separates three questions: was the piece found, was it engaged with and did it contribute to an outcome. Each failure has a different fix, and rewriting a page that was never found wastes hours. The Google Search Console tutorial explains how to read the discovery side honestly: filter, read click-through against position and compare like periods. This exercise asks you to apply both before deciding anything.

The performance extract

Marbleway Toys is an invented toy retailer with an online shop and a blog of gift and buying guides. The owner says the blog brings "hardly any sales". The extract covers July to September, non-brand queries, one country.

Illustrative example — not a real result

Marbleway Toys: search and analytics extract, July to September

ArticleImpressionsClicksCTRAvg positionMain query
A. /blog/best-gifts-for-5-year-olds-202324,0002901.2%5.8gifts for 5 year olds
B. /blog/gifts-for-6-year-olds9,8001501.5%11.2gifts for 6 year olds
C. /blog/birthday-presents-six-year-old7,100600.8%16.4gifts for 6 year olds
D. /blog/wooden-toys-guide15,5001,2408.0%2.9are wooden toys better
E. /blog/marble-run-buying-guide6,2002103.4%9.6best marble run
F. /blog/christmas-stocking-fillers1,10090.8%23.0stocking fillers for kids
G. /blog/brickworks-range-guide3,9001203.1%7.0brickworks sets
H. /blog/toy-fair-stand-20224000%48.0none of note
ArticleOrganic sessionsEngagement ratePurchases in sessionNotes
A27052%3Title and heading say 2023; 8 of the 20 products listed are out of stock
B14061%2In Search Console, the page ranking for its query alternates with C
C5547%0Covers the same ages and many of the same products as B
D1,18078%2No links to products; ends with a newsletter sign-up only
E20070%14700 words; competing pages carry comparison tables
F1040%0Last November and December: 3,400 clicks and 41 purchases
G11535%0Marbleway stopped stocking the Brickworks brand in spring; 6 sites link to the page
H425%0A news post; no external links

Diagnosing each article

Read each row as a journey. Low impressions mean a discovery problem. Impressions with a weak CTR for the position mean the result is shown but not chosen, which usually points at the title, the snippet or a stale promise; the CTR calculator helps when you want to check the arithmetic. Clicks with weak engagement mean the page does not deliver what the result promised. Engagement without purchases means the page is read but leads nowhere. Two pages sharing one query in GSC is a structural problem.

Two cautions. Clicks in Search Console and sessions in GA4 are counted differently and will not match; compare each source with itself over time rather than with the other. And seasonal pages must be judged against their season, not against whichever months the extract happens to cover. The KPI tree framework is a useful frame for the outcome side: purchases from an article depend on sessions, the share of readers who reach a product and the share of those who buy, and the action should target the weakest of those inputs.

Task

  1. For each article, name the stage where it fails: discovery, engagement, contribution or structure.
  2. Give each article one action: keep, update, merge, redirect or retire.
  3. Write the reason for each action in one sentence, quoting the evidence from the extract.
  4. For every update, say what changes on the page.
  5. Rank the actions by expected effect against effort, and say which the owner should do first.

Worked answer

  • A, update. Found and shown near the top, but rarely chosen: the click-through is weak for that position and the title promises last year's list. Remove the year from the title and heading, replace out-of-stock products, and keep the URL, since moving it would throw away its history.
  • B and C, merge. Two pages compete for one query and both rank poorly as a result. Keep B, which has the stronger history, move anything useful from C into it, redirect C to B and update internal links.
  • D, update. Found, clicked and read closely, yet it contributes almost nothing because there is nowhere to go. Add in-context links to the wooden toy range and a clear next step. A contribution failure, not a writing one.
  • E, update. The best converter on the blog, just off page one. Add the comparison table the results page rewards and link to it from the marble run category.
  • F, keep. The summer figures say nothing about a Christmas page. Judged against last December it performed well; schedule a refresh for early autumn so it is current before the season.
  • G, redirect. The page describes a range Marbleway no longer sells, so it cannot be made useful, but other sites link to it. Redirect it to the construction toys category, the closest replacement.
  • H, retire. Nobody finds it and nothing links to it. Remove it and record the decision in the inventory.

Order. E and D first: least work, clearest outcome gain. Then A before the gift season, then the merge, then the quick G and H. F waits for its refresh.

Illustrative example — not a real result

Why E comes first

E turned 200 sessions into 14 purchases, a rate of 7%. D turned 1,180 sessions into 2, about 0.2%. If D's new product links bring it even to half of E's rate, it would produce around 40 purchases a quarter from traffic it already has. E, moved from position 9.6 onto page one, would gain clicks on top of a rate already proven.

Common mistakes

  • Rewriting A's body when the problem is the title and the stock list.
  • Keeping both B and C because each brings some traffic. Together they hold each other back.
  • Retiring F on summer data.
  • Deleting G and losing what its links bring; redirect it instead.
  • Comparing Search Console clicks with analytics sessions and treating the gap as an error.

Check your diagnoses against the worked answer, then return to the intermediate content marketing level and start the content strategy capstone.

Pages this lesson applies

Terms used on this page

Also related

Related