Carolina Pintos Therapy · Organic Search

SEO Status

Prepared by Argentle
Reporting date September 17, 2026
Search Console through September 12, 2026
Clicks per day
56

Down 33% from 84 a day in September 2025.

Money keyword position
2.6 → 5.1

lymphatic drainage massage houston. Clicks fell 68%.

Impressions, one query
−98%

Cost about 1 click a day. This is noise, not the problem.

Reviews unanswered
191

Of 1,021. Google counts owner replies. Cheapest fix here.

The traffic fell because rankings fell

Year on year, same 21 days. These terms held steady impressions and lost position, so fewer people clicked. That is a real ranking loss, not a reporting artefact.

Search termPosition, 2025 → 2026Clicks
lymphatic drainage massage houston
2.65.1
93 → 30
lymphatic massage houston
2.114.7
20 → 11
lymphatic drainage massage
4.233.5
98 → 66

Across the whole export

  • 74 search terms lost more than one position. Together they went from 356 clicks to 156.
  • 74 terms gained more than one position, and together gained 41 clicks. The exchange was not even.
  • One blog post lost 234 clicks, half the year on year total. A sibling post on a near identical subject gained 127 and rose from position 28.8 to 5.0. The site is competing against itself.

Clicks per day, by month

Average organic clicks per day by month, June 2025 through September 2026
Per day, not per month, so that a 28 day February and the 12 days of September so far sit on the same scale. Hover any bar.

Four changes in six weeks, and they overlap

The decline starts on 9 August. Three site changes landed after it began, close together. That makes attribution hard for anyone, which is why we are adding a way to record changes as they ship: from the next one on, a checkpoint will be able to say which change moved the numbers.

Aug 9 to 22
Impressions slide from 16,414 a day to 3,849. Smooth, no step. Begins nine days before the Google update, so the update is not the cause.
Aug 21, 22:28 & 22:57
Listing submitted to Semrush, then 44 blog pages change timestamp in 35 seconds. Worth knowing what ran that evening.
Sep 2
SEO plugin removed. No step change in the data before or after.
Sep 10 to 12
Titles and headings rewritten on 34 of 37 pages. Redirects and internal links done correctly.

Audit

Employee portal indexed in Google, linked from all 114 pagesOpen
/contact/ redirects to a salary increase formOpen
No services hub, and /services/?anything returns 404Open
WordPress login linked from every pageOpen
191 reviews never answered, 111 of them from 2025Open
Website address says Suite 260 South, 72 directories say Suite 260Open
The Woodlands is not in Listing Management at allDecision
What ran on the evening of August 21?Your answer
Structured data incomplete and partly invalidLower priority
Listings: 66 of 72 connected. The August "27 to fix" is doneResolved
Duplicate booking pages consolidatedResolved
Internal linking survived the September move intactVerified
Staging site locked downResolved

The order we would recommend

The growth plan already drafted is structurally right: those six procedure pages are real gaps and we would build them. Our view is about sequence. The three biggest recoverable items are on pages and terms that already exist, and they are cheaper and faster than anything new.

Repair first · 01
~5 clicks/day
Two blog posts competing for the same searches
One lost 234 clicks while the other gained 127. Both already written. No new content.
Repair first · 02
~3 clicks/day
The money keyword, 2.6 to 5.1
High intent local traffic, on impressions the site already earns.
Repair first · 03
191
Unanswered reviews
A factor Google names in writing. No developer, no content, no risk.
Then build.

The six procedure pages, in the plan's own order. For scale: its strongest target is sized at 140 searches a month, roughly 1 click a day if won outright, after writing and after months of waiting to rank.

Spanish last.

Most real content work, least evidence behind its sizing. Worth the pilot, after the rest is moving.

Before any of it.

We still cannot say why the money keyword fell 2.5 positions. Answering that costs nothing and needs no changes to the site. Getting it wrong sends the whole budget the wrong way.

What we have done

  • Crawled the whole site. 115 addresses, every page's title, headings, links and markup recorded.
  • Froze a baseline. 16 months of Search Console, plus a snapshot of every page. Nothing can be proven to have worked without it.
  • Built change detection. We can now list which pages changed and when, to the minute.
  • Checked the client-side analysis against the data. Its numbers reproduce; its cause does not.
  • Audited the Semrush account through the API: 72 listings and all 1,021 reviews.
  • Ran the comparison that found the ranking loss. Two date ranges, page and query level.

Questions only you can answer

  • What was running on the evening of 21 August, roughly 5:30pm to 6:00pm Houston time?
  • Which form of the Memorial address is correct, Suite 260 South or Suite 260?
  • You mentioned Houston Lymphatic twice. They built out their service pages in May and June, right as the slide accelerated. Is there anything from the clinic's side that makes that more than a coincidence of dates?
  • Should The Woodlands be added to Listing Management, and who owns that call?

Where the numbers come from. Google Search Console for carolinapintos.com: a 16 month baseline taken 14 September, and two date range comparisons taken 17 September. Semrush Local API, 16 September. Direct checks of the live site on 16 and 17 September. Positions are averages weighted across a period, not a ranking on a given day.

The full write up, with every figure, every caveat and everything we could not verify, is in the accompanying document. It is written to be read by a person or handed to an AI assistant.