Ashton Jones-Obikpo

Before AI: Data Quality & Context in Commercial Real Estate

Let’s have a conversation about why data quality and context matter in our world of commercial real estate.

GTM Playbook 01

Question: How can commercial real estate firms turn fragmented information into reliable commercial intelligence?

In commercial real estate, our questions shouldn't be reductive. With the pace of AI adoption accelerating, it's tempting to build solutions simply to keep up.

But a solution that works for one organisation may not work for yours. A solution adopted by a competitor may not be right for you.

I would argue that racing to adopt AI is largely redundant if the underlying foundations have not been established.

Intelligent adoption begins with understanding how we organise, validate and connect our data. It requires repeatable, accurate and scalable processes that allow information, when appropriately enriched, to produce reliable commercial intelligence.

PLEASE – Not all data is equal!

The value of our data is dependent on the questions being asked, the importance of the decision being made, and the specific problems we are trying to solve.

This is especially important when attempting to understand and anticipate occupier behaviours that may indicate future property requirements.

We know that commercial real estate information is often fragmented. Company records, lease events, building specifications and workplace evaluations may be held across different sources. We can chart market activity in considerable detail, but these records may help us establish what has happened without necessarily explaining why it happened or what might happen next.

That distinction matters!

A lease expiry might tell us when an occupier faces a contractual decision, but it does not independently explain what that occupier intends to do.

The challenge, therefore, is not simply collecting more information. It is establishing a consistent method for organising, validating and connecting what is already available.

The process should be repeatable, efficient, scalable and fast.

For this to be possible, information must be structured within systems capable of accommodating increasing volumes of data without compromising its integrity or usefulness.

AI can accelerate the interpretation of information, but it cannot automatically establish whether that information is accurate, complete or commercially relevant.

Before we ask what AI can tell us, we should understand the quality and context of the information we are asking it to interpret.

The Parable of the Blind Men and the Elephant

The parable of the Blind Men and the Elephant illustrates a challenge familiar to commercial real estate.

Each person encounters a different part of the elephant and develops an understanding based on their limited perspective.

Each account reflects part of the reality, but none describes the whole.

Consider the information associated with a single commercial office building.

The physical building: Floor plans, mechanical and electrical specifications, energy performance certificates, maintenance records, warranties and fire-safety documentation describe the asset's physical condition and operational requirements.

The commercial building: Lease agreements, rent reviews, break clauses, licences and transaction records describe its contractual obligations and commercial position.

The occupier: Company information, operational locations, workforce requirements, organisational changes and decision makers provide another perspective.

The relationships: Landlords, managing agents, surveyors, contractors, consultants and solicitors each hold information generated through their respective responsibilities.

These records are rarely maintained within a single system.

A contractor may hold the latest fit-out drawings, a managing agent may maintain operational records, while a solicitor retains transaction documentation.

Another platform may contain information about the occupier's business. Each source serves a legitimate purpose, but the collective picture can remain incomplete.

An Approaching Office Lease Expiry — Does It Constitute a Potential Requirement Signal?

A lease expiry represents a verifiable contractual event, provided the information is accurate.

However, it does not independently establish whether the occupier intends to relocate.

Understanding that decision may require consideration of the building's suitability, lease provisions, the occupier's future workplace requirements, relocation costs and supply-and-demand dynamics in the market.

Some of this information may be publicly accessible; much of it will not. A verified data point is not necessarily a verified commercial opportunity.

Connecting information provides valuable context, but the resulting intelligence is only as reliable as the evidence supporting its interpretation.

The Importance of Data Governance Frameworks

Data is extracted from different formats, sources and platforms. Without consistent structures and standards, connecting datasets can introduce errors rather than improve intelligence.

The domino effect is a risk that must be acknowledged.

Poorly structured or inconsistent data can undermine Ideal Customer Profile (ICP) analysis, opportunity pipelines and, most importantly, commercial decision-making. Established data management frameworks provide guidance for reducing these risks.

DAMA - Establishing the Structure

The Data Management Body of Knowledge (DAMA-DMBOK), published by DAMA International, provides guidance across 11 data management knowledge areas, including governance, modelling, integration, metadata and data quality.

It helps organisations establish accountability, manage information consistently and align data practices with business objectives.

DAMA is not a prescribed technology solution. It is guidance that should be adapted to an organisation's requirements.

FAIR - Making Data Usable

The FAIR Guiding Principles provide complementary guidance for making data:

FAIR does not independently guarantee accuracy or enforce governance. It provides principles that make information easier to discover, exchange and reuse.

How DAMA and FAIR Complement Each Other

In her January 2026 article, FAIR and DAMA: Complementary Approaches to Operational Data Governance, Ajithavalli Chellappan proposes a useful interpretation of their relationship.

She describes DAMA as providing strategic direction and a governance framework, while FAIR helps translate that direction into measurable operational practices. Her central observation is particularly relevant: frameworks alone are insufficient. Organisations must translate their principles into practical responsibilities, processes and measurable standards.

Within CRE, information should be structured around the commercial questions an organisation needs to answer.

A design-and-build contractor evaluating potential fit-out opportunities may interpret an occupier's lease event differently from an architectural practice identifying potential design commissions. Both may examine the same underlying information, but the relevance of individual signals and the conclusions they can reasonably draw will differ.

The underlying data must remain verifiable; its commercial interpretation must remain specific to the decision being made.

From Data Quality to Decision Quality - Establishing Measurable Standards

Adopting data management frameworks does not, by itself, demonstrate that commercial intelligence has become more reliable.

The effectiveness of a system must be established through measurable outcomes, not simply the volume of information it contains.

The UK Government Data Quality Framework identifies six core dimensions for evaluating data quality, drawing on the work of DAMA UK.

Their application should reflect the decisions an organisation intends to make.

Measuring What Matters

Consider a CRE organisation maintaining information about office occupiers, buildings and upcoming lease events.

Relevant measures might include:

They do not, however, establish whether a commercial interpretation is correct.

Reliable Records to Reliable Signals

An organisation might accurately identify 100 approaching lease expiries. That does not mean it has identified 100 prospective relocation or fit-out opportunities.

Consider an occupier approaching a lease event.

An independently verified increase in workforce requirements may strengthen the case for a future workplace decision. Evidence of an active workplace evaluation could provide further context. However, neither necessarily confirms an intention to relocate. The occupier may instead be considering a refurbishment, consolidation, lease renewal or changes to its existing workplace strategy.

The commercial assessment must consider what each signal establishes, how the signals relate to one another and what remains unknown.

A Repeatable Process

Improvement begins with establishing a baseline, implementing defined validation and governance procedures, and continuously evaluating performance against that baseline.

It is essential to have clear responsibilities, documented sources, and a structured methodology for recording commercial signals.

Individual signals can be assessed according to their:

A documented scoring methodology can help distinguish preliminary indicators, emerging requirements, and commercially qualified opportunities.

Importantly, confidence should not increase simply because several data points support the same narrative. Those signals may originate from the same underlying source or reflect outdated information.

The objective is not simply to generate intelligence faster. It is to establish whether that intelligence is sufficiently reliable to inform a decision.

A reliable commercial intelligence system should not construct a narrative around an isolated data point. It should allow hypotheses to be evaluated against multiple, independently verified signals, within a relevant commercial timeframe, with conclusions traceable to their supporting evidence.

The value lies not in the quantity of signals collected, but in the strength of the evidence connecting them to a commercial decision.

One distinction remains important: corroboration is not the same as predictive accuracy.

Multiple independent signals may strengthen the evidence supporting a potential requirement, but they do not establish the probability that it will materialise. Predictive accuracy requires separate evaluation against observed outcomes over time.

The application of predictive systems to commercial real estate intelligence, including their limitations, will be explored in the next GTM Playbook – Building Blocks article.

If you have insights to share or would like to discuss the ideas explored here, feel free to connect with me on LinkedIn.

AJ

References & Further Reading

DAMA International - Data Management Body of Knowledge(DAMA-DMBOK) https://dama.org/learning-resources/dama-data-management-body-of-knowledge-dmbok/

GO FAIR Foundation - FAIR Guiding Principle

https://www.gofair.foundation/fair-principles

Ajithavalli Chellappan (2026) - FAIR and DAMA: Complementary Approaches to Operational Data Governance:

https://medium.com/@ajithavalli/fair-and-dama-complementary-approaches-to-operational-governance-4068e7eb86ec

UK Government (2020) - The Government Data Quality Framework

https://www.gov.uk/government/publications/the-government-data-quality-framework/the-government-data-quality-framework

#CRE intelligence #GTM Playbook #GTM engineering #commercial real estate #data governance