We kicked off in March, paused through autumn while you needed us to, and restarted in June with a sharper brief: Dunedin and Otago decision-stage searchers rather than broad global terms. The strategy layer you approved on 26 June came from triangulating 116 keywords across four markets and three data sources, landing on 96 approved keywords (57 focus), a cluster strategy, an AI prompt library, the full page plan and on-page spec, and the 50-question FAQ bank. All of it lives in your Master Sheet.
Where you started is worth remembering, and the July baseline snapshot keeps it on record: an AI-readiness score of 30/100, domain authority of 18 against Elm's 30 and The Opera's 46, your site cited in just 2 of the 49 AI answers we measured while Elm appeared in 9, and search clicks that had slid from 333 in February to 159 by July. The old homepage had no H1 and no structured data at all.
The calibration pieces, the new answer-focused pages and the rebuilt site were the response to that baseline, and August is the first month you can see it turning: the click slide reversed (198, the best month since May), and the new species pages started earning impressions within days of going live (the hoiho page 148, albatross 132, sea lion 108). The rest of this page is the migration verdict and where that leaves us.
Your new website went live in late August. A migration is the single riskiest event in search, because if the old addresses do not point cleanly at the new ones, years of Google history can disappear overnight.
Yours came through clean. We tested every mapped redirect ourselves rather than taking a checklist's word for it: 14 of 14 pass, every one a single hop to a live page. The domain settles correctly, genuine missing pages return the right code, and the catch-all for the old tour URLs works.
Comparing the four days after launch against the matching four days before: clicks per day up 13.5% and impressions per day up 55.1%. The new site is being shown more, not less.
This was the real risk. All eight of the new answer-focused pages shipped with their planned titles, the wildlife hub is correctly set up so the individual species pages carry the weight, and 17 of the 18 pages in the plan are live. The structure we built the strategy around came through the rebuild intact.
Searches for your own brand name show an average position drop after launch. Our read is that this is transitional noise from the cutover, and a live check still puts you at number 1 on the related terms, but we are flagging it early. We will confirm either way in mid September.
Google now shows an AI answer on 25 of your 55 tracked searches. Nearly half. For a tourism business where people research heavily before booking, that is the shift that matters most, and your ChatGPT referrals more than doubling during the migration month suggests the new content is already being read that way.
The biggest content prizes are also now clear: the three highest-volume searches in your set are the species terms ("yellow eyed penguin" at position 35, "hoiho" at 28, "nz sea lion" outside the top 50), and the new species pages that just started earning impressions are exactly the vehicles to climb them. That is where the blog programme aims next.