Content optimization case | Robert James | Published October 8, 2026
An AI SaaS business had ten published pages that needed closer alignment with search intent. I revised their titles, meta descriptions, H1s, structure, and full-page content before the next content batch. This case focuses on the delivered optimization work and the method used to follow up on each page.
Client: AI SaaS business
Scope: Ten existing pages
Implementation: Seven pages: August 5; three pages: August 6, 2026
Status: Ten pages optimized; historical seven-day comparison available
The starting opportunity
I prioritized published pages rather than treating a larger content library as the immediate solution. The task was to make each result snippet and page body better reflect what the searcher wanted. That meant checking the page's purpose against the search results before rewriting individual elements.
A concrete implementation decision
One affiliate-content page was revised to explain the timeline to a first sale, break-even, and profit. The recorded title change made those three milestones explicit, and the tracker records a full-body revision alongside the title, meta description, and H1 changes. I applied the same intent-led review across the ten-page set.
How I framed the comparison
The page-level tracker covers all ten pages, their publication dates, implementation notes, and GSC/GA4 metrics. I kept the comparison windows attached to each publication group. That preserves the timing of the work rather than assigning one date range to every page.
Matched comparison windows
Seven pages went live on August 5. Their comparison is July 28-August 3 versus August 6-12. Three went live on August 6. Their comparison is July 30-August 5 versus August 7-13. Each pair covers seven days and excludes its implementation day.
Historical seven-day results
Across the ten tracked page rows and their respective seven-day comparison windows, GSC impressions rose from 8,330 to 9,006 (+8.1%), and search clicks rose from 17 to 33. Summed per-page query counts increased from 433 to 549 (+26.8%); the same query can appear on several pages, so this is not a deduplicated query total. Impression-weighted average position moved from 10.40 to 9.54.
Measurement limits
The same rows record organic sessions moving from 42 to 40 and CTA-click events from 22 to 8, so the outcome was mixed. I calculated these historical totals from the ten individual page rows using their recorded dates; inconsistent summary and introductory date fields were excluded. The client engagement has ended, and longer-term results are unavailable. This seven-day observation does not establish sustained growth or causation.
Discuss a similar project
Send me a page with search exposure but weak content fit. I can assess its intent, prioritize the changes, and improve the existing page before recommending more content.