What a website rebuild costs in search traffic: one client, the year to August
Over the Year to August, This Client's Search Visits Rose 35%. This August, They Were Down 64%.
The year to August against the same months a year earlier, then the months since the rebuild. The good half, the bad half, and what we'd check first now.
Get This Report for Your SiteSee What It Costs
Rather write it out? Send a note
A Pest Control Company With a Wildlife Division
The subject of this page is a pest control company with a wildlife division. Not a pure wildlife company. Pest is the larger half of the business, and the wildlife work runs beside it. That's true of every client in these studies.
We're keeping the name off the page. Part of what's below is ours to publish and theirs to live with. Every figure stays exactly as we measured it.
Two things happened at this company in May 2026, and you need both before you see any traffic number.
They moved to a new website. We built it. A rebuild like this one moves pages to new addresses, and the old addresses are the ones search engines have been sending people to for years.
Then the owner changed direction. Wildlife had been earning a large share of the site's traffic. The company decided to lead with general pest work instead, and the wildlife and rodent content stopped being the priority.
How We Measured It
A wildlife year isn't flat. What drives the call changes by season, and it changes differently in every market. So we compare the year to August against the same months a year earlier: October 2025 through August 2026, against October 2024 through August 2025. Eleven whole months on each side. The reporting tool we pull from reaches back only to late September 2024, so the earlier year has no whole September. Starting both years in October keeps every month in the table a whole month.
The full-year number is search visits from Google Analytics, the visits it files under organic search. That counts every search engine, not only Google. We use it because Google Search Console only keeps 16 months, and two years side by side take 23.
Where Search Console still held both years when we pulled it on Sept. 25, 2026, May 13 to Sept. 22, we show clicks too. Before the rebuild, the two measures tracked each other closely. From June 2025 through May 2026, monthly search visits ran within 17% of monthly Google clicks in every month. Since then, search visits have run above clicks, by 19% in July and 28% in August.
The Pest Sites We Can Compare
Before the numbers for this site, the context for them. These are the pest control clients whose reporting holds search visits for both years, October through August. One more pest client's reporting doesn't break out search visits, so it isn't here.
| Site | Year before | Latest year | Change |
|---|---|---|---|
| A | 4,231 | 6,619 | +56% |
| B, this page | 22,896 | 30,944 | +35% |
| C | 868 | 998 | +15% |
| D | 646 | 738 | +14% |
| E | 1,196 | 1,050 | -12% |
| F, our longest-running client | 45,834 | 14,634 | -68% |
Four of the six grew, this site among them, on the strength of the months before the rebuild. The four smaller sites live on their home, city and service pages. Three of them grew, and E fell 12%. The biggest loss is our longest-running client, where articles earned 82% of Google clicks from May 13 to Sept. 22, 2025.
The Full Year, Month by Month
The year splits cleanly at the rebuild.
| Month | Year before | Latest year |
|---|---|---|
| October | 1,143 | 2,688 |
| November | 724 | 2,674 |
| December | 470 | 2,278 |
| January | 760 | 2,902 |
| February | 1,329 | 3,851 |
| March | 2,146 | 4,996 |
| April | 2,193 | 4,359 |
| May, rebuild May 19 | 2,647 | 2,881 |
| June | 3,537 | 1,692 |
| July | 4,144 | 1,410 |
| August | 3,422 | 1,238 |
Google Analytics doesn't add its months up to its year total exactly, so the months come to within 2% of the year totals.
October through April, search visits went from 8,765 to 23,748. About 2.7 times the year before. March 2026 was the best month in two years of data.
June through August, they went from 11,103 to 4,340. Down 61%. August alone went from 3,422 to 1,238, down 64%.
Count the whole year to August and it still comes out ahead, 22,896 to 30,944, up 35%. We're not going to lead with that number and stop there. The half that matters for next year is the second one.
What We Found
We reviewed the old site before any work started. It had more pages than the rebuild, so the before counts run higher than the 291 pages the launch crawl found.
Three problems, counted page by page.
| What the audit found | Pages |
|---|---|
| Running the homepage meta description word for word | 187 |
| Meta descriptions too long to show in a search result | 310 |
| Titles too vague to tell Google what the page was about | 221 |
What We Fixed
The technical pass ran May 19 to May 20, 2026.
Every count dropped. Two didn't reach zero. The 40 redirects left all point to outside sites, which is where they should point. One page still carries two H1 headings.
| Fix | Before | After |
|---|---|---|
| Internal 4xx errors | 3 | 0 |
| Redirects the crawl found | 123 | 40, all to outside sites |
| Pages with two H1 headings | 6 | 1 |
Then the verified after-state. The crawl on May 20, 2026 covered 291 pages and returned zero duplicate title groups and zero duplicate meta description groups.
One more didn't reach zero. A single meta description still runs past 160 characters, on the contact page, at 174. We're naming it, because a report that rounds 1 down to 0 is the kind of report you have already paid for once.
A second crawl on July 18, 2026 covered 318 pages and found zero broken internal links.
What It Cost
Since the rebuild, Google clicks tell the same story as search visits. August went from 3,256 clicks to 965, down 70%, and July from 3,820 to 1,186. From May 13 to Sept. 22, the longest stretch Search Console held for both years when we pulled it on Sept. 25, 2026, clicks fell from 14,159 to 6,325. Down 55%.
The timing points at the rebuild. The site was ahead of the year before for seven straight months, right up to launch. In the two weeks before launch, clicks ran 23% ahead of the same weekdays a year earlier. Then it launched, and by late June it was running at about a third of the year before.
The pages that lost the most were articles, not service pages. Rat and mouse questions, and one post about keeping skunks away. Every page with an animal or a wildlife word in its address, old and new together, went from 8,118 clicks to 1,644 on those dates. Down 80%.
What Went Up Anyway
We built a wildlife removal hub, one page per animal, plus an exclusion and repair section. They replaced older animal pages at shorter addresses. So the fair comparison is the new pages against every old page that now redirects into them, same dates a year earlier.
The old pages earned 276 clicks from May 13 to Sept. 22, 2025. The new ones earned 317 on the same dates in 2026. A gain of 41. Small. We're not going to dress it up as anything else. It went up while the site around it fell by more than half.
The homepage went from 259 clicks to 934 on the same dates. The new blog folder went from 492 to 4,416, mostly because the rebuild moved the old articles into it.
Why It Fell
The data points at three things, and it can't split them, because all three landed in the same weeks.
The first is the launch itself. The site was running ahead of the year before right up to launch and fell after it, so the rebuild is part of the cause.
The second is the change of direction. The owner took the focus off the wildlife and rodent content that had been earning a large share of the traffic, and moved it to general pest work. Traffic follows attention, and the attention moved.
The third is happening to every site that lives on articles. The answer at the top of Google has gotten heavier. Ahrefs measured a 34.5% lower click rate for the top result when an AI Overview sits above it, on March 2025 data, and 58% lower on December 2025 data. Question-style articles, like the rat and skunk posts that fell hardest here, are the searches that answer lands on. This site leaned on them harder than most. Before the rebuild, articles earned about nine of every ten of its Google clicks.
Google alone doesn't explain it. Three of the four smaller sites in the table, the ones that live on local and service pages, grew over the same months. The fourth fell 12%. They don't lean on articles the way this one did, so they can't tell us how much of the article loss was Google.
We're also not going to tell you the decline is fixed, or give you a month when it comes back. We don't know, and a date would be a guess with a number attached.
What We Would Measure First Now
Before launch, every old address that earns clicks, and the new address it points to. At launch, every one of those old addresses opened once, to confirm it lands on a live page. After launch, those same pages week by week against the same weeks a year earlier, so a drop shows up in the first report, not the ninth month.
That doesn't make a rebuild free. It means the cost gets seen while it can still be fixed.
Why We Are Telling You This
A rebuild that moves addresses puts search traffic at risk, and so does a change of direction. This site did both in the same month, in a year when Google started answering more questions itself.
So the question was never whether it would cost something. The question is whether the agency measured the cost and said the number out loud, the good half and the bad half both.
That's the whole argument for this page. You're going to hire somebody who reports on their own work. Some quarter, your numbers will move the wrong way, and the report that lands on your desk will either say so or find a chart that hides it. We would rather you judge us by our worst report than by how we describe ourselves.
If you're weighing a rebuild and nobody has told you what it costs in search traffic, book a call or send a note. We'll tell you what we would measure before you start, so the number isn't a surprise nine months later.
Book a CallFAQ
Why the Year to August, and Why Search Visits?
Because what drives a wildlife call shifts through the year, and it shifts differently in every market. A few months against a few months measures the calendar, not the website. October through August against the same months a year earlier puts nearly every season on both sides, in whole months. Search Console can't reach back two years, so the year number is Google Analytics search visits. Where Search Console holds both years, we show Google clicks too.
If the Year Went Up, Why Is This Page About a Loss?
Because the year has two halves, and the second one is the one heading into next year. The months before launch grew strongly. The months since launch are down by more than half, and August fell 64% on the August before. Leading with the 35% and stopping there would be the chart that hides it.
Was It the Rebuild or Was It Google?
Both, plus the change of direction. The site was running ahead of the year before right up to launch, then fell, so the timing points at what happened that month. Google's AI answers most likely took clicks from question searches over the same year. Four of the six pest sites in the table grew over the year. Three of them live on home, city and service pages. The fourth is this site, which got about nine of every ten Google clicks from articles before the rebuild. So the table can't settle how much was Google.
Would You Publish My Numbers Like This?
We keep every client's name off these pages, gains and losses alike. The numbers stay exactly as measured.