AI answers

AI answers

The same 30 questions put to six AI engines, latest check 1 October 2026 and previous 1 September 2026. The audit is re-run monthly.

67%was 50%

Questions where at least one engine named or cited Harbour & Key, 20 of 30.

27%8 of 30
named in the answer
63%19 of 30
cited as a source
33%10 of 30
neither
Gemini
57%+24 pts
17 of 30 answered · was 33%
Perplexity
37%+14 pts
11 of 30 answered · was 23%
Google AI Mode
30%+13 pts
9 of 30 answered · was 17%
Claude
23%+6 pts
7 of 30 answered · was 17%
Google AI Overview
15%-17 pts
4 of 26 answered · was 32%
ChatGPT
3%same
1 of 30 answered · was 3%
The tracked questions30 frozen questions
30 of 30 questions
namedcited onlynothingno AI answer
One row per question, one dot per engine, in the order shown at the top. Open a row to see how each engine answered and which sites it read. The arrow compares the question with 1 September on the same test: named or cited by at least one engine.
CompetitorsHarbour & Key 2nd
1ashfordvaleproperty.example
26.7%
2harbourandkey.example
21.4%
3zoopla.co.uk
16.1%
4reddit.com
12.5%
5getagent.co.uk
11.7%
6alliedsurveyors.example
11.1%
7rightmove.co.uk
10.0%
8brookfieldestates.example
10.0%
Share of the sources AI engines cite across the whole question set, normalised because engines cite very different numbers of sites and a raw count would simply reward the chattiest one.
Sources the engines read14 buying-decision questions
reddit.comForum · 12 of 14
getagent.co.ukComparison site · 7 of 14
carrickandlowe.exampleCompetitor site · 7 of 14
zoopla.co.ukProperty portal · 7 of 14
ashfordvaleproperty.exampleProperty portal · 6 of 14
rightmove.co.ukProperty portal · 6 of 14
northlighthomes.exampleCompetitor site · 5 of 14
reallymoving.comComparison site · 4 of 14
top10propertyagents.co.ukRound-up directory · 4 of 14
Review sites only surfaced on questions phrased around best or reviews, and on which agent should I use Google's AI often builds its shortlist from rival firms' own sites rather than from comparison sites.
How this is measured, and the engine-by-engine counts
Engine by engine1 Sep → 1 Oct
EngineNamedCitedChecked
Gemini5 → 68 → 16uptwice
Perplexity1 → 2up7 → 10uptwice
Google AI Mode2 → 54 → 7once
Claude1 → 3up5 → 5twice
Google AI Overview3 → 18 → 4once
ChatGPT0 → 01 → 1twice

Counts of questions, out of 30. Only a change larger than an engine's own run-to-run disagreement is marked, so unmarked figures have not moved meaningfully.

Named means the answer text itself names Harbour & Key. Cited means a Harbour & Key page appears in the sources behind the answer, whether or not the answer names the firm. An engine can do either without the other, so the two are always shown apart.

ChatGPT, Claude, Gemini and Perplexity answer every question twice, so a single flip in one run does not read as a change. Google AI Overview and AI Mode answer once. A question that flips on one of those two is still shown in the list above, but one answer can swing it, so treat those arrows as a prompt to look rather than as proof of a move.

Google AI Overview does not appear on every search. Questions where it showed nothing are excluded from its own percentage rather than counted against it, and marked with a faint dash in the list.

Being named in AI answers is tracked and aimed at, never promised.

Real client work. Names and locations changed.