How should a professional services firm start an AI strategy?

Start from the business goal and choose technology afterwards. Bidwells asks each team what problem, once removed, would help it meet the firm's target of doubling revenue by 2030. Ben Lee found that AI was often not the answer to a team's problem, so the tool comes second.

Why did Bidwells move from an AI group to an innovation committee?

Because many team problems need a technology answer other than AI, and one view across the firm cuts duplication. Bidwells' tech and innovation committee covers all ten departments. Before it existed, teams ran their own due diligence, which led to duplicate tools and systems that did not deliver value.

Where does AI add most value in a property consultancy?

AI scales best across repeatable work such as running meetings, reviewing leases, writing scoping reports and carrying out site appraisals. Ben Lee called this a layer of document generation with some orchestration of data. Tailored client advice built on specific market knowledge is much harder to scale with AI.

Should every employee have an AI objective?

Bidwells thinks so, and treats it as a request to think differently. Every employee has a performance objective to identify one area of their work that AI could support. It can be as small as summarising a document with Copilot, or as large as building a new client service.

AI strategy at Bidwells: doubling revenue without doubling headcount

Downloads | AI case studies, Sales and marketing leaders

Bidwells’ AI strategy starts from a business goal: double revenue by 2030 without a proportional rise in headcount. AI is part of that plan, but the 187-year-old property consultancy does not expect AI alone to deliver it. Ben Lee, then Head of Data and AI at Bidwells, explained the approach to aibl in March 2026.

The firm runs technology decisions through one committee, manages AI projects as a build or buy portfolio and puts governance before scale. Every employee now has an AI objective. Here is how it works.

Bidwells in numbers

Double the revenue by 2030, without a proportional increase in headcount.
500 people work at the property consultancy.
10 departments are covered by its tech and innovation committee.
40 university students took on a real planning problem at its hackathon.

All figures as stated by Ben Lee in an interview with aibl.

What is Bidwells’ AI strategy?

It is a plan to help people do more, so they can grow the business. Lee expected AI to bring incremental productivity gains, short of the whole answer to doubling revenue. So Bidwells starts every project from a team’s goal and chooses technology afterwards. Every employee also has an objective to find one area where AI could help.

AI alone won’t get us there. Productivity gains are going to be incremental.

Ben Lee, then Head of Data and AI, Bidwells

How does Bidwells decide whether to build or buy AI?

It builds when it has the capacity and buys when a product fits its requirements well. Lee calls this portfolio management. Projects are sorted into buckets, from self-serve through to enterprise-wide. In-house resources then go to problems no vendor product can solve, such as tailored, client-specific advice.

What did Bidwells learn from its AI hackathon?

That understanding the problem matters more than technical flair. Forty university students tackled a real problem from Bidwells’ planning department. The winning teams asked the planners the most questions before touching any tools, and they did not need spectacular machine learning. They are now doing further work with Bidwells.

The people I really listen to in all of this are the people that aren’t talking about technology.

Ben Lee, then Head of Data and AI, Bidwells

How does Bidwells govern AI tools?

Anything deployed at scale goes through the Microsoft stack, because sensitive client and property data has to stay in the UK. The team still prototypes with third-party tools where it can move faster. Lee saw the Power Platform produce chaotic, disconnected dashboards without clear governance, and he warns AI agents could repeat it.

Key takeaways

Start from the business goal. Ask what problem, once removed, helps the team hit its target.
Use one committee across departments. It cuts duplicate purchases and shows shared problems.
Govern before you scale. Prototype fast, but deploy on a stack that meets your data rules.
Give everyone an AI objective. One area of their work that AI could support.

About the speaker

Ben Lee was Head of Data and AI at Bidwells, a 187-year-old property consultancy of 500 people. He started as a journalist, spent two decades in real estate and joined Bidwells as a comms manager. He has since co-founded Instruction. Ben Lee spoke to aibl in March 2026.

What’s in the full case study

If you need AI to support growth without a matching rise in headcount, this is a model to borrow. The full PDF goes deeper than this page, in Ben’s own words.

How the AI group became a tech and innovation committee covering all ten departments, and what it changed.
Where AI scales in professional services, and where client-specific advice still needs people.
How Bidwells balances governance and speed, from UK data rules to fast prototyping.
Five actions to take back to your own organisation.

For CEOs, COOs, and data, AI or innovation leads in professional services firms. Six pages, about a seven-minute read.