Timeless International

Website Development / SEO / Content Marketing / 2026

Gosheni Safaris: 24% more search clicks after the move, and the 57 tours that went missing

Client: Gosheni Safaris

Moving goshenisafaris.com off WordPress in October 2026 meant handing Google 487 changed addresses at once. Four weeks on, Search Console showed 24% more clicks, and five times as many from people who had not typed the company’s name. It also showed ten of the phrases that sell safaris dropping, some from the first page of results to the seventh. This is what we measured, what we got wrong and what we did about it.

The challenge

A safari company earns its place in Google one page at a time, over years. Before the move, Gosheni’s WordPress site held an average position of 8 for “tanzania safari packages”, 9 for “big five safari” and 3 for “mount kilimanjaro”. Replacing a whole site puts all of that at risk at once: every address changes, every title and description is written again, and Google reads the lot from the start.

So the job had two halves. Move without losing what had been earned. Then check, in Search Console, whether that was true.

The approach

Before the switch, a command read the old site’s Rank Math sitemaps and wrote a redirect for each of the 487 addresses in them. Go-live had one pass mark: a second command had to find all 487 answering, and the switch waited until it did. Redirects keep their query strings and count their hits, an old address nobody planned for is logged the first time it is asked for, and renaming a page today writes its own 301.

The new pages are rendered on the server, so Google and AI assistants get the whole page in the first response. Each page carries one linked block of structured data naming the company, the trip or place, and the breadcrumb trail. One sitemap is built from the same rule that decides whether a page may be indexed, so the two cannot disagree.

Then we measured. On 6 November we set four weeks of Search Console data against the four weeks before, phrase by phrase. On 7 November we fetched all 789 pages in the live sitemap the way a search engine does, without running JavaScript, and checked each one against a list.

The results

Clicks from Google rose from 1,270 to 1,580, up 24%, between the four weeks to 7 October and the four weeks to 4 November. Clicks from searches without the company’s name went from 14 to 74, and Google showed the site for those searches 26,300 times against 10,354.

The same report held the bad news. “Tanzania safari packages” fell from an average position of 8 to 61, “big five safari” from 9 to 68 and “maasai mara safari” from 12 to 70. Ten sales phrases fell this way.

The main cause was ours. Fifty-seven tours had no price on the old site, so our import brought them in as hidden drafts and sent their old addresses to general country pages. One of them had been second in Google for “5 day tanzania safari”. After the move its address opened a page about Tanzania. On 6 November all 57 went back up as “Price on request”, their 120 old address versions were pointed at the right tours and checked one by one, and the tours in the sitemap went from 108 to 164.

More was missing than the tours. Over the next five days we brought back 64 sections of long copy that the move had cut from country and park pages, turned 158 headings that had become plain paragraphs back into headings across 40 posts, and sent the old addresses of 69 lodges to the right park page.

It is too early to call it a recovery. An outside tracker following 200 phrases put the site’s average position at 28.4 on 2 October, 37.0 at its worst on 30 October and 27.8 on 13 November. That is one tool’s estimate from US desktop searches, not Search Console. The next four-week comparison will say more than we can now.

All case studies · Start a project