Deep audit
Technical SEOEntity SEOWe audited the whole site: crawling and indexation, page structure, on-page content, entity signals, and access for AI crawlers.
The agency’s average Google position was 31.4, deep on page four, and its site had no lead tracking. We rebuilt the site around a topical map, with 19 new money pages and 90 in-depth articles, then measured every result in the agency’s own GA4, Search Console and Bing Webmaster Tools.
An Australian real estate buyer’s agency started working with Intelitune on May 9, 2026. In six months, after a full audit, a topical map, 19 new money pages and 90 in-depth articles, Google clicks rose from 591 to 1.82K, Microsoft Copilot citations grew from 1.1K to 6.8K, and search and AI assistants brought 29 tracked leads, all read in the agency’s own GA4, Search Console and Bing reports.
Average Google position 31.4, a small site with few commercial pages, and no way to count leads.
Audit, topical map, 5 old pages redirected into 19 money pages, 90 articles, full-site CRO, entity, AEO and GEO work.
Google clicks 3x, Copilot citations 6x, 53.7K Google AI impressions, 29 search and AI leads.
| Metric | Before | After | Change | Report and window |
|---|---|---|---|---|
| Google Search clicks | 591 | 1.82K | +208% | Search Console, last 6 months vs previous 6 |
| Google Search impressions | 175K | 356K | +103% | Search Console, same comparison |
| Average Google position | 31.4 | 22.2 | 9.2 places up | Search Console, same comparison |
| Google click-through rate | 0.3% | 0.5% | +0.2 pts | Search Console, same comparison |
| Google organic sessions | 748 | 2,308 | +208.6% | GA4, May 9 to Nov 9, 2026 vs Nov 8, 2025 to May 11, 2026 |
| Organic sessions, all search engines | 789 | 2,511 | +218.3% | GA4, same comparison |
| Bing clicks | 42 | 169 | +302% | Bing Webmaster Tools, May 10 to Nov 7, 2026 vs prior 6 months |
| Bing impressions | 3K | 11.1K | +270% | Bing Webmaster Tools, same comparison |
| Microsoft Copilot citations | 1.1K | 6.8K | +518% | Bing AI Performance, May 9 to Nov 7, 2026 vs prior 6 months |
| Average cited pages in Copilot | 3 | 8 | +5 pages | Bing AI Performance, same comparison |
| Google AI feature impressions | No baseline | 53.7K | n/a | Search Console Generative AI features, latest 6 months |
| ChatGPT sessions | 40 | 45 | +12.5% | GA4 source / medium, May 9 to Nov 9, 2026 |
| ChatGPT engaged sessions | 14 | 30 | +114.3% | GA4, same comparison |
| Leads from organic search | No baseline | 26 | n/a | GA4 lead acquisition, first-user channel, May 9 to Nov 9, 2026 |
| Leads from AI assistants | No baseline | 3 | n/a | GA4 lead acquisition, same window |
On page four of Google, with 591 clicks in six months. The site had 175K impressions in that time, so buyers were searching, but an average position of 31.4 meant almost none of them reached it.
Property buyers often research for weeks before they call anyone, and more of that research now happens inside AI assistants. The agency’s site had only a few pages, and no way to tell which visits turned into inquiries.
We built the structure first, then the content, then the signals AI engines use to pick a source. Every step came from the audit and the topical map.
We audited the whole site: crawling and indexation, page structure, on-page content, entity signals, and access for AI crawlers.
We mapped the questions property buyers research, grouped them into topics, and decided which page would answer each one.
Five old pages were 301-redirected into 19 new money pages built from the map, and the whole site was reworked for conversion. GA4 lead tracking was set up at the same time.
Ninety long-form articles with sourced statistics, each written to a topic in the map and structured so one section answers one buyer question in full.
Clearer entity signals, structured data and answer-first passages, so Google, Bing, Copilot and ChatGPT could tell what the agency does, who it serves and where.
Four results, each read from its own report: Google search, the other search engines, AI visibility, and leads. Every card names the report and the date range it came from.
One selected engagement. This is one of our documented cases, not an average across our clients. Results differed from case to case.
Impressions doubled from 175K to 356K and click-through rate rose from 0.3% to 0.5%. In GA4, Google organic sessions grew from 748 to 2,308, and engaged sessions from 519 to 1,387.
How to read this
Search Console compares the last six months with the six before. Percentages are worked out from the rounded values the report shows (1.82K, 356K), so treat them as close, not exact.
Bing clicks rose from 42 to 169 and impressions from 3K to 11.1K. Across all search engines, GA4 counted 2,511 organic sessions against 789 in the six months before, including DuckDuckGo, Ecosia and Yahoo.
How to read this
Bing clicks come from Bing Webmaster Tools (May 10 to Nov 7, 2026). Session counts come from GA4 (May 9 to Nov 9, 2026). The two tools count differently, so their numbers don’t match exactly.
Microsoft’s AI Performance report counted 6.8K citations across Copilot and partner experiences, up from 1.1K, with the average number of cited pages rising from 3 to 8. ChatGPT and Gemini visits are still small but growing, and ChatGPT’s engaged sessions more than doubled.
How to read this
Google’s Generative AI features report shows zero for the earlier six months. That reflects the report, not proof of zero visibility, so 53.7K stands as current-period evidence only.
Microsoft Copilot6.8KCitations, up from 1.1K. Average cited pages 3 → 8.
Impressions in the Generative AI features report, latest 6 months.
Sessions, up from 40. Engaged sessions 14 → 30, with 4 key events.
Sessions, up from 0. A new source this period.
GA4 credits 26 new leads to organic search and 3 to AI assistants, by the channel that first brought each person to the site. Visits from ChatGPT turned into key events at 8.89% per session, against 2.34% for Google organic.
How to read this
We set up lead tracking when the engagement started, so there is no earlier lead count to compare with. ChatGPT’s rate comes from 45 sessions and 4 key events, a small sample, so read it as an early signal.
These are the eight reports behind every figure above, from the agency’s own GA4, Search Console and Bing Webmaster Tools. Click any report to see it full size.
What’s hidden and why. None of these reports shows the agency’s domain or account details. The lead report shows only the organic search and AI assistant rows; the rows for other channels were removed. The figures in the charts above match these reports.
Ask to see a live account on a call →Publishing the gaps is part of the method. Here is what these figures can’t tell you.
Google organic engagement rate fell from 69.39% to 60.1%, while average engagement time held at 1m 14s. Three times the visits came with a slightly lower share of engaged ones.
Lead tracking started with the engagement, so the 29 search and AI leads have nothing earlier to compare with. We don’t claim a lead growth rate.
The results period runs from May to November, and spring is usually a busy season in Australian property. Part of the lift may be seasonal; this comparison doesn’t adjust for it.
ChatGPT sent 45 sessions and Gemini 5 in six months. The citation and impression numbers are large; the visits they produce so far are not.
The 19 money pages came first, so the 90 articles had commercial pages to support from day one.
The biggest multiples in this case came from Microsoft: Bing clicks 4x and Copilot citations 6x.
Without tracking there is no baseline. Set it up before the work starts, not after the results arrive.
ChatGPT visits converted at 8.89% per session against 2.34% from Google. Small numbers, worth watching.
See where your real estate site stands on Google and in AI answers, measured the same way.
Request a diagnostic →GA4 recorded 29 new leads between May 9 and November 9, 2026 whose first visit came from search or an AI assistant: 26 from organic search and 3 from AI assistants. Lead tracking was set up at the start of the engagement, so there is no earlier count to compare against, and we don’t claim a growth rate.
They come from the AI Performance report in Bing Webmaster Tools, which counts how often a site’s pages are cited as sources in Microsoft Copilot and partner experiences. From May 9 to November 7, 2026 the agency had 6.8K citations, up from 1.1K in the six months before, and its average number of cited pages rose from 3 to 8.
The method doesn’t depend on the country. Site structure, content, entity signals and tracking work the same way in the US, the UK and Australia, and buyers use the same AI assistants. What changes is the competition and the words buyers search with, which is why every engagement starts with a diagnostic of your own site and market.
A deep audit, a topical map of buyer questions, 5 old pages redirected into 19 new money pages, conversion work across the whole site, GA4 lead tracking, and 90 in-depth articles with sourced statistics. On top of that came technical, entity, AEO and GEO work so Google, Bing and AI assistants could understand and cite the agency.
We’ll read your Google and AI visibility the same way: your own data, a named date range, nothing modeled.
Every figure on this page is read from the source report named next to it. In the exports, the domain, brand searches and page paths are hidden; the numbers are unchanged, and the source exports are on file.
This is one of our documented engagements, not a typical or average result. Outcomes depend on the starting point, the market, the competition and how much of the work gets implemented. Figures are reported as recorded and are not annualized or projected.
The engagement is ongoing. Other marketing the business ran during this period isn’t separated out of these figures.
Figures last checked against the source reports on Nov 10, 2026.