An online garden retailer gave its category pages a reason to rank

Representative examplesEvery case study in this library is an illustrative composite of the kind of engagement we deliver — written to show our method and standards, not to name clients.

Where they started

Garden retail runs on seasons, and this online retailer curates its range by hand: a merchandiser who knows what actually grows, sixty categories from bedding plants to tools, and grids that change with the weather. Each category page carried one thin supplier paragraph — identical tone across all sixty, matching no query anyone types. The real expertise lived in the team's daily reply emails, where shoppers ask about clay soil, north-facing walls, and what survives a shady balcony. Long-tail searches went to competitors who answered them.

What it was costing

Category pages carried one thin supplier paragraph each — identical tone across 60 categories, matching no real query. Long-tail terms ('shade plants for clay soil') went to competitors who answered them.

What they could see

• Long-tail searches like 'shade plants for clay soil' ranked competitors who answered the mission; the retailer's pages matched none.
• All sixty category pages carried the same single supplier paragraph in the same supplier voice.
• Staff answered the same growing questions by email every day — answers that existed nowhere on the site.
• Non-brand organic traffic had been flat for two consecutive seasons while the range itself sold well.

The constraints we worked inside

• The merchandiser curated products by hand; copy changes could not disturb the grids.
• Seasonality ruled the range; copy had to survive rotation.
• The retailer's voice is plain-spoken expertise — corporate SEO tone would have broken the brand.

What had been tried before

A previous SEO vendor supplied AI-generated category blurbs for all sixty categories at a flat rate.

The tone read like a catalogue nobody wrote; the merchandiser refused to publish past the first three, and two blurbs misstated growing conditions the staff knew were wrong.

The team wrote blog posts answering common growing questions, separate from the category pages.

The blog ranked occasionally but the categories — where the grids and the money live — never gained; shoppers read a post and left without meeting the plants.

What we proposed

We proposed writing from the buying mission rather than the keyword: each category gains copy answering the mission behind the search — conditions, care level, what actually grows — drafted from the buyer emails the team already answers daily, so the words start in the team's own register. Mission-based sub-sections (by light, by problem, by skill level) would turn grids into guides, with product links doing the internal-linking work naturally. Copy keyed to conditions, not SKUs, survives the seasonal rotation. Success would be measured as entrances on mission terms, per category.

Just as important is what we ruled out, and why:

• Generating all sixty categories' copy in one batch — Unreviewed generation already failed once for tone and twice for accuracy; the merchandiser's sign-off per category is the quality bar, and seasons make some categories pointless off-cycle.
• Hiring an outside garden writer — The answers already exist in the team's reply emails; a writer without that history would guess at conditions and reproduce the supplier-paragraph problem at greater cost.
• Using customer reviews as category content — Reviews live on product pages and prove individual purchases; category guidance needs horticultural accuracy a named human signs off, not star-rating collage.

How the work ran

Delivered by the growth pod — SEO strategist + content editor over 6 weeks, with working increments reviewed with the client every week.

The stack, and the reasoning

Search-intent mapping

'Shade plants for clay soil' is a mission, not a keyword; mapping the query language per category defined the sub-sections each guide needed.

Content briefs

Briefs separated horticultural claims from prose, so the merchandiser approved facts in minutes and the editor kept the plain-spoken voice without a battle.

Internal linking

Mission sections link to matching products and across categories — clay soil guides to drainage tools — so the grid stays untouched while the linking work happens naturally.

Analytics

The win condition was entrances on mission terms, tracked per category; an aggregate traffic line would have hidden which missions actually pulled.

What went wrong

How we worked together

Cadence

A Monday 30-minute working call with the merchandiser and the owner; a batch of two or three drafted categories moved through review each week.

Client side

The merchandiser owned horticultural accuracy and the grids; the owner owned priorities and budget; a part-time assistant pasted approved copy into the CMS.

Decisions

Horticultural claims were approved line by line by the merchandiser; category order was re-cut monthly against season and revenue.

They provided

The buyer-email archive, CMS access, the merchandiser's review time, and condition notes per category from whoever knew the stock.

What changed

The headline: non-brand organic entrances to category pages, 12 weeks post-rollout versus prior period — +61%, read from Analytics, same-period comparison. A second check: category pages with entrances on mission terms at 4 → 23.

The pages now answer what customers ask before they email it, and the daily replies have measurably shortened — staff link the guide section instead of retyping it. The merchandiser sends customers to her own category pages, which she had never done. Mission-term traffic arrives already knowing what it wants, and the assistant who pastes copy can now brief new categories without us in the room.

The result was read from Analytics, same-period comparison against the pre-engagement baseline over the stated window, with a guardrail check on category pages with entrances on mission terms. Where platform-reported numbers and business outcomes differ, this record says which layer it is quoting.

What they own now

• The category-copy pattern and page template with contents block
• The brief template separating horticultural claims from prose
• The season-and-revenue priority calendar for the remaining categories
• The mission-term analytics view, wired per category
• A one-hour walkthrough for the assistant covering new category rollouts