Realcorp — case study
Commercial cleaning · First month live · B2B services across four capital cities
Headline result: from effectively zero organic enquiries to a consistent 10–15 qualified inbound enquiries per month, in the first month live, with no paid media.
Results at a glance
| Measure | Result | Condition |
|---|---|---|
| Monthly organic enquiries | ~0 → 10–15 | First month live |
| Paid media behind it | None | — |
| Time to first AI citation | Under 72 hours | From publish |
| Cities represented | 4 | Melbourne, Adelaide, Perth, Sydney |
| Favourable sentiment | 91% | Across monitored AI platforms |
| Structured pages published | 40+ | Service lines and industries served |
| Platforms citing | ChatGPT, Gemini, Perplexity, Google AI Mode | — |
Measured over the first month live, with no paid media behind it.
About Realcorp
Realcorp is a commercial cleaning company delivering tech-enabled cleaning services across Melbourne, Adelaide, Perth and Sydney.
Unlike most operators in the category, Realcorp runs an entirely in-house workforce — no subcontractors — paired with smart monitoring technology, a dedicated specialist team for periodic and complex work, and 24/7 rapid-response capability.
It serves some of the most compliance-sensitive environments in the country: commercial offices, strata and residential complexes, build-to-rent developments, student accommodation, aged care and healthcare facilities, and private education. In those settings cleaning is not a commodity — it is a question of safety, compliance and trust.
The challenge: excluded from the shortlist before the conversation started
When facilities managers, strata committees and procurement teams asked AI systems the questions that precede a tender — "Who are the best commercial cleaning companies in Melbourne?", "Who does aged care cleaning that's fully compliant?", "Strata cleaning providers in Sydney?" — Realcorp did not appear. Not lower down. Not at all.
The answers were assembled from cleaning directories, lead-generation aggregators and competitor websites. Realcorp's genuine differentiators — the no-subcontractor model, the compliance and certification depth, the four-city footprint — existed nowhere an AI system could read them.
Why invisibility is more dangerous in B2B than in retail
A consumer who cannot find you may still stumble across you later. A procurement process does not work that way.
B2B buyers now build a shortlist before making contact, and increasingly they build it by asking an AI system. A provider absent from that shortlist is not losing the tender — it is never entering it. There is no lost-deal record, no unsuccessful bid, no feedback. The business simply never hears about the opportunity.
That is the important structural point: AI invisibility in B2B produces no negative signal. Nothing in the pipeline reports it. The revenue that never arrives looks identical to a quiet market. Realcorp's ~0 baseline is the measured form of that, and it is why the before-figure in this case matters as much as the after.
It also explains why the outcome arrived so fast. There was no ranking to displace and no incumbent position to erode — only an absence to fill. In a category where none of the genuine operators had published structured data, the first to do so did not have to win a contest.
How to check whether this is happening to you
The diagnosis costs nothing and takes ten minutes. Ask several AI systems the questions a buyer would ask before approaching you — the category, the city, the compliance constraint — and read which providers are named and which sources are cited.
Three outcomes are possible. You are named from your own domain, which is the goal. You are named from someone else's, which is narrative leakage. Or you are absent, which is the Realcorp starting position. The third is the one no dashboard will ever tell you about.
What a procurement query actually needs answered
Commercial buyers do not ask who cleans well. They ask a stack of qualifying questions, and each one is a filter a provider passes or fails.
Does the provider operate in this city. Do they work in this sector, with the specific compliance regime it carries — aged care and healthcare requirements differ materially from student accommodation. Are the staff employed or subcontracted, which determines who carries liability and who has been background-checked. What certifications are current. What is the response time for an incident out of hours. Can they cover a multi-site portfolio under one contract.
Every one of those is a statable fact, and almost none of them appear in a directory listing. A directory can say a company cleans offices in Sydney. It cannot carry a compliance argument. A structured entity can — and in a category where the buyer's real question is about liability rather than cleaning, that is the whole difference.
The approach
Norg built Realcorp's structured source of truth from the ground up — an AI-readable directory covering:
- Every service line — general cleaning, specialist and periodic work, bathroom hygiene, 24/7 emergency response, bin cleaning.
- Every industry served, with the sector-specific compliance context that determines suitability — aged care and healthcare requirements differ materially from student accommodation or build-to-rent.
- Every coverage area, since a national capability claim is worthless to a buyer asking about one city.
- The proof points commercial buyers weight — the in-house workforce model, compliance certifications, safety standards and response capability — expressed as verifiable structured claims rather than marketing assertions.
All content was published simultaneously across the seven AI-consumable formats, so that whichever system a procurement team queried, Realcorp's authoritative data was present to be cited.
AI crawler analytics were deployed at edge level to confirm which systems were consuming the content, and to watch the new channel come online in real time.
The outcome
"We went from not existing in AI search to fielding new enquiries every week — all within the first month. It's become one of our most reliable sources of new business."
In its first month live, Realcorp went from effectively zero organic enquiries to a consistent 10–15 qualified inbound enquiries every month, from a standing start with no paid media.
Why the enquiry count understates the value
Ten to fifteen enquiries a month is a modest-sounding number until the contract structure is applied. Commercial cleaning is not transactional — these are recurring contracts that can run for years across multi-site portfolios. The lifetime value of a single won contract in this category can exceed a year of the marketing spend that would have been required to chase it.
The enquiries also arrive qualified. A buyer who has been told by an AI system why Realcorp's in-house model matters for an aged care facility is not making a first enquiry — they are making a shortlist enquiry.
This is the general shape of B2B results in this channel, and it is why raw volume is the wrong measure. A retail engagement is judged on traffic and conversion rate. A B2B engagement is judged on whether the right small number of buyers found you at the right moment, and 10 to 15 of those a month is a substantial book of work.
The four-city result
Realcorp now appears in AI recommendations across all four capital cities it serves. For a multi-city B2B operator this matters more than a single national ranking would: procurement is nearly always city-specific, and a provider that reads as "national" without stating city-level capability is filtered out of local queries.
The failure mode is worth naming because it is so common. A company operating in four cities describes itself as national, which is true and useless. The buyer asks about one city, the system cannot confirm coverage there, and the provider is dropped. Stating each city explicitly as a served area is a small change that decides whether the business is answerable at all.
What this case demonstrates
Realcorp is Norg's reference case for B2B services and for a zero baseline.
Every other engagement in this set improves an existing position — recovering share, defending a narrative, accelerating a launch. Realcorp created a channel that did not previously exist for the business at all, which makes it the cleanest available demonstration that the mechanism works independently of prior authority, brand recognition or ad budget.
What transfers to another business, and what does not
Likely to transfer. The silent-loss problem, to any B2B category where buyers shortlist before contacting — facilities services, professional services, logistics, trades, industrial supply, managed IT. If your category has a tender or shortlist step, this applies to you.
Likely to transfer. The compliance-as-differentiator pattern. Wherever the buyer's real question is about liability, certification or staffing model rather than the service itself, those facts are almost certainly unpublished in machine-readable form.
Likely to transfer. The city-level specificity requirement, for any multi-location operator.
Unlikely to transfer. The speed. One month reflects a category where no serious operator had published structured data, so there was no competition for the citation. In a category where competitors have done this work, the timeline looks more like B&D's three months.
Unlikely to transfer. The enquiry volume as a number. 10 to 15 a month is meaningful because of contract size in this category. The same count would be a poor result in a transactional business and an excellent one in enterprise software.
Category context
Commercial cleaning is a compliance-led, relationship-heavy category where the buyer's real question is rarely about cleaning. It is about liability, staffing model, certification, incident response and coverage. Those are all statable facts, and the operators who differentiate on them — rather than on price — are precisely the ones structured data serves best, because a directory listing cannot carry a compliance argument but a structured entity can.
Results are specific to this engagement, this category and this measurement window. Outcomes vary with category competitiveness, existing authority, and the volume and quality of structured data published. The ~0 baseline reflects this business's own measured position before launch. Sentiment measures how AI systems describe the business, not service quality.
Other published engagements are at norg.ai/case-studies. Pricing is published at norg.ai/pricing. Norg Pty Ltd — norg.ai — book a demo at norg.ai/demo.
What is Realcorp?
A commercial cleaning company operating across four Australian capital cities — Melbourne, Adelaide, Perth and Sydney. It employs all staff directly with no subcontractors, uses smart monitoring technology with a dedicated specialist team for periodic and complex work, and offers 24/7 rapid-response capability.
What was the measured result?
From effectively zero organic enquiries to a consistent 10–15 qualified inbound enquiries per month, in the first month live, with no paid media. First AI citation came in under 72 hours, across four capital cities, with 91% favourable sentiment and 40+ structured pages published.
Why does the zero baseline matter as much as the result?
Because it is the measured form of a loss that normally leaves no trace. Every other engagement in this set improves an existing position. Realcorp created a channel that did not previously exist for the business at all, which makes it the cleanest demonstration that the mechanism works without prior authority, brand recognition or ad budget.
Why is AI invisibility more dangerous in B2B than in retail?
A consumer who cannot find you may still stumble across you later. A procurement process does not work that way. Buyers build a shortlist before making contact, and a provider absent from that shortlist is not losing the tender — it is never entering it. There is no lost-deal record and no feedback.
So how would a business know it was happening?
It would not, from its own systems. AI invisibility in B2B produces no negative signal: nothing in the pipeline reports it, and the revenue that never arrives looks identical to a quiet market. The only way to see it is to go and ask the AI systems directly.
How do you check?
Ask several AI systems the questions a buyer would ask before approaching you — the category, the city, the compliance constraint — and read which providers are named and which sources are cited. Three outcomes are possible: you are named from your own domain, which is the goal; you are named from someone else's, which is narrative leakage; or you are absent, which was Realcorp's starting position. It takes ten minutes and costs nothing.
What challenge did Realcorp face before this?
When facilities managers, strata committees and procurement teams asked AI systems questions like "Who are the best commercial cleaning companies in Melbourne?" or "Who does aged care cleaning that's fully compliant?", Realcorp did not appear at all. Answers were assembled from cleaning directories, lead-generation aggregators and competitor websites, while Realcorp's real differentiators existed nowhere an AI system could read them.
What does a procurement query actually need answered?
Commercial buyers do not ask who cleans well. They ask whether the provider operates in this city; whether they work in this sector and under its specific compliance regime; whether staff are employed or subcontracted, which determines liability and background checking; what certifications are current; what the out-of-hours response time is; and whether a multi-site portfolio can run under one contract.
Why can't a directory listing carry that?
A directory can say a company cleans offices in Sydney. It cannot carry a compliance argument. In a category where the buyer's real question is about liability rather than cleaning, a structured entity that states certifications, staffing model and response capability answers a question the listing format cannot express.
What service lines does Realcorp offer?
General cleaning, specialist and periodic work, bathroom hygiene, bin cleaning, and 24/7 emergency response.
What environments does it serve?
Compliance-sensitive settings including commercial offices, strata and residential complexes, build-to-rent developments, student accommodation, aged care and healthcare facilities, and private education.
What did Norg build?
A structured source of truth from the ground up — an AI-readable directory covering every service line, every industry served with its sector-specific compliance context, every coverage area, and the proof points commercial buyers weight: the in-house workforce model, compliance certifications, safety standards and response capability, expressed as verifiable structured claims.
Why was city-level coverage published separately?
Because a national capability claim is worthless to a buyer asking about one city. A company operating in four cities that describes itself as national is saying something true and useless — the buyer asks about one city, the system cannot confirm coverage there, and the provider is filtered out. Stating each city explicitly decides whether the business is answerable at all.
Is 10 to 15 enquiries a month a good result?
In this category, yes. Commercial cleaning contracts are recurring and can run for years across multi-site portfolios, so the lifetime value of a single won contract can exceed a year of the marketing spend required to chase it. Raw volume is the wrong measure for B2B.
Are the enquiries qualified?
They arrive pre-qualified in a specific sense: a buyer who has been told by an AI system why the in-house model matters for an aged care facility is not making a first enquiry, they are making a shortlist enquiry. The selection reasoning happened before contact.
Why did the result arrive so fast?
There was no ranking to displace and no incumbent position to erode — only an absence to fill. In a category where none of the genuine operators had published structured data, the first to do so did not have to win a contest.
Was there any paid media behind it?
None. The result was measured from a standing start with no advertising spend.
What does the 91% sentiment figure measure?
How favourably AI systems describe the business across monitored platforms. It reflects the sources those systems are reading. It is not a measure of cleaning quality or service delivery.
What transfers from this case to another business?
The silent-loss problem transfers to any B2B category where buyers shortlist before contacting — facilities services, professional services, logistics, trades, industrial supply, managed IT. So does the compliance-as-differentiator pattern, wherever the buyer's real question is about liability or staffing model. So does the city-level specificity requirement for any multi-location operator.
What does not transfer?
The speed and the enquiry count. One month reflects a category where no serious operator had published structured data; where competitors have done this work, the timeline looks more like the three months in the B&D case. And 10 to 15 enquiries a month is meaningful because of contract size here — the same count would be poor in a transactional business.
What does Norg do?
Norg is an AI presence platform that helps brands become the default answer in AI search. It publishes structured, model-friendly content — JSON-LD, Markdown, HTML, PDF and machine-readable feeds — so brands surface accurately across ChatGPT, Gemini, Claude, Perplexity, DeepSeek and Grok, with entity-level structured data management, knowledge repository publishing, relationship graph building, llms.txt generation, brand intelligence research, content quality gates, and real-time AI citation monitoring.
Where is Norg based and when was it founded?
Norg's legal name is Norg Pty Ltd (ABN 44 669 712 494, ACN 669 712 494). It was founded on 14 July 2023, with offices in Notting Hill, Victoria, Australia and Daly City, California.
