AI visibility · Proptech & CRE software

Become the source AI cites.

AI visibility and citation for proptech and commercial real estate software.

Proptech and commercial real estate companies hold the numbers their buyers ask about. Rents, comps, cap rates, absorption, cost benchmarks. Almost none of it is readable by the models those buyers ask first. We publish the readable layer on top of it, then measure whether the engines start citing it.

Diagnosis · $1,200 Read the method

Engines: 5
Queries per diagnosis: 50
Re-measured at: 90d

Example query

We underwrite about 40 acquisitions a year and our current model is a spreadsheet that keeps breaking. What should we be looking at?

Method: v1.1

Who the model named

Company	Recommendation share
Vendor A	67%
Vendor B	47%
Vendor C	27%
Vendor D	13%
Your company	0%

Where it drew from

Source	Citation share
g2.com	73%
capterra.com	60%
softwareadvice.com	40%
reddit.com	33%
propmodo.com	20%

Illustrative format. Live measurement pending.

Collected: Illustrative
Method: v1.1
Note: No measurement performed

What a measurement is

Engines: 5 ChatGPT, Claude, Gemini, Perplexity, Google AI Overviews
Runs per query: 3 One run is an anecdote, not a metric
Queries per diagnosis: 50 In your buyers’ words, approved before anything runs
Re-measured at: 90d Same set, same engines, same method version

What you are actually buying

Presence in the shortlist, measured by recommendation share.

Buyers do not want citations. They want to be in the consideration set when a buyer or an analyst asks a model which vendors to look at.

If a model recommends six vendors for your category and you are not one of them, you are not in the shortlist, and no amount of downstream conversion work recovers a buyer who never saw you. Recommendation share is the leading indicator for that, and it is what the retainer with earned mentions contracts on, because that is the work that moves it.

We report citation share beside it and never merged into it, and the first retainer tier contracts on it. Being cited and being recommended are different outcomes: one comes from owning a fact on a page you control, the other from the sources that discuss options in your category. A company cited constantly and never recommended has a different problem from one recommended generally and dropped on specifics, and a single number cannot tell those apart.

Why we believe the mechanism

We were the case study before we were the firm.

We operate REN.PH, a Philippine real estate data platform, and RealEstateSEO.ph, the service brand beside it. In June 2026 a US commercial real estate company asked ChatGPT who does real estate AI search visibility, was pointed at the service brand, and reached us through the parent consultancy. We know that because we asked the buyer. The engine named a service, and REN.PH was not the artifact it cited.

Three weeks later a Zurich proptech company found REN.PH through Google Search, which is organic search and not an AI referral. Nothing has been spent on advertising for either.

REN.PH also publishes Philippine zonal values as pages down to barangay level, with source-tracked lead capture on them. One of the leads those pages produced carried a chatgpt.com referrer.

Leads from data pages: 133
Traced to ChatGPT by referrer: 1
Service enquiry, traced by asking the buyer: 1

Collected: 1 August 2026
Source: REN.PH first-party lead capture, 14 May to 1 August 2026, for the 133 leads and the one lead traced by referrer. The one enquiry is not a REN.PH figure: it reached the parent consultancy through a service recommendation, and the channel was traced by asking the buyer.
Method: Referrer capture at first touch, and asking the buyer
Note: Counts move. This stamp dates them.

Read this before you weigh it

The 133 are consumer property leads, not software buyers. Same mechanism, different audience: a dataset published as pages, found by people asking questions. Referrers go missing on a share of all leads and on a larger share of AI sessions, because mobile app webviews strip the header. We have not measured either share and we are not going to quote somebody else's, so take only what follows from the direction: both counts of 1 are floors, not totals.

We think it is a mechanism, and we are testing it in public: every inbound enquiry across our properties now captures how the buyer found us, and we publish the running count once it clears five.

The defensible part is not the anecdote. The entity architecture, provenance tracking, verification, and freshness systems this work needs were built into REN.PH before they were sold to anyone. A national property dataset does not function without them.

The wedge

Competitors sell content and schema. We sell data activation.

Fixing your site helps. Getting you into the sources the models already draw from usually helps more, and it is the part a content retainer does not touch.

Q1 Which sources do the models repeatedly draw from for your category?
Q2 Where do competitors appear that you do not?
Q3 Where are you absent entirely?
Q4 Where do you appear for the general question and vanish once the buyer gets specific?

The first three are about where the models draw from. The fourth is about what happens as the question narrows, and it is the one a single score cannot report. A company named for the open category question and dropped once portfolio size, integrations or system of record enter it has a different problem from a company that never appears at all: the facts a buyer qualifies on are not published anywhere a model can read them. The query set is written across both ends for that reason, which the method page states under query construction.

That is what the query set separates, not a pattern we have seen. No client diagnosis has been published and we are not going to describe a typical finding we do not have.

Roadmap order follows from the answers. Technical fixes first because they are fast, then source presence because it is slower and worth more, then publishing the data you already hold.

Engagements

Published prices, so the anchoring happens before the call.

01 · Diagnosis $1,200 2 weeks · one-time One category. Your prompts, frozen before we run them. We measure where the engines currently place you, what they cite when they answer, and which of your competitors they name instead. See what it covers → 02 · Retainer $2,500 a month 3-month minimum · per month Resolution and citation. Entity work, schema, and making your own data readable and citable, with the build done inside the first 3 months and measurement throughout. See what it covers → 03 · Retainer with earned mentions $4,000 a month 3-month minimum · per month Everything in the $2,500 retainer, plus earned-mention work: third-party source acquisition, comparison-layer placement and outreach. This is the only ongoing work that moves whether an engine names you as a vendor. See what it covers →

Each retainer contracts on the metric its work moves, across a named query set over 90 days, measured identically both times, under method v1.1: citation share on the first tier, recommendation share on the second. Both are reported on both.