Shoosmiths uses artificial intelligence to capture legal expertise and improve contract reviews

Shoosmiths built an AI contract review tool that saves 3 to 5 hours per document. The firm also deployed a no-code agent to enforce procedural consistency for retail clients.

Categorized in: AI News Legal
Published on: Jun 25, 2026
Shoosmiths uses artificial intelligence to capture legal expertise and improve contract reviews

A senior innovation advisor at Shoosmiths won a car for building an AI agent that slashed document-search times and enforced procedural consistency for a high-street retailer's property portfolio. The agent, built with Microsoft Copilot Studio's no-code tools, reflects a wider push at the law firm to embed AI across its work-including a custom contract review system named Apollo that codifies decades of legal expertise.

An agent built without writing code

Joanne Bevan joined Shoosmiths 28 years ago as a secretary in the real estate team. Today she's a senior innovation advisor who has never written code. When the firm began piloting Microsoft 365 Copilot, she was an early adopter, using it for transcripts and drafting before moving to Copilot Studio. "I found it really, really easy to build agents," she said.

Her winning entry tackled a concrete problem. A long-standing retail client had hundreds of leased properties, each requiring detailed contracts and very specific procedures. After staff turnover, newer lawyers were spending too long finding the right documents and guidance. Bevan helped the team build a structured SharePoint repository, then created an agent that lets people ask questions in plain English and retrieve exactly what they need.

"It's just a simple retrieval agent," she said. "But it can't be underestimated how powerful this can be." The value isn't only speed-the client gained confidence that lawyers were following the correct process every time. Shoosmiths' use of Copilot Studio underscores the relevance of Microsoft AI Courses for legal teams aiming to build similar retrieval agents.

Codifying legal playbooks with Apollo

Tony Randle, director of client technology and service improvement at Shoosmiths, says the firm reviews hundreds of contracts every year. Before generative AI, it built a natural language processing system for one contract type-a project that took two years. Large language models offered a faster path, but instead of training a generic tool to think like a lawyer, Shoosmiths decided to encode its own expertise.

The result is Apollo, a contract review system running on Microsoft Azure. Apollo draws on playbooks that Randle describes as "the aggregate of scores of lawyers' experience … distilled down to a gold standard." Each playbook sets out what a contract should contain, why it matters, and how to fix issues using tried-and-tested drafting. Apollo compares draft contracts against these rules, flags deviations, and explains the reasoning-citing the precise section of the playbook. "It doesn't need to be a lawyer," Randle said, referring to the AI's ability to spot rule breaches. The transparency is essential for a profession built on judgement and trust.

Sarah Hartley, a trainee solicitor at the firm, found Apollo especially useful as a learning tool. "It doesn't just assist with marking up a lease, it identifies issues that I might not otherwise have spotted, suggesting practical drafting solutions and providing a clear explanation of why it has flagged those provisions for change," she said. This hands-on application of AI for Legal shows how firms are moving beyond generic products to systems that capture deep domain knowledge.

Efficiency gains and a learning culture

Shoosmiths CEO David Jackson says the firm has "gone all in" with Microsoft technology, giving staff access to Copilot and Copilot Studio and encouraging experimentation. "We're not building these tools to cut jobs, we're doing it to empower our people, help them learn, and be more innovative. And it means we offer our clients something they can't get elsewhere," he said.

The efficiency case is strong. Randle estimates that this type of AI-assisted review can save three to five hours per document. The team piloting Apollo handles roughly 3,000 documents a year. During intense M&A periods, Shoosmiths has advised on more than 500 corporate transactions in a single year. The firm plans to extend the approach to more document types and teams-63% of partners in an internal poll said they wanted to be early adopters.

Why this matters for legal professionals

Shoosmiths' experience shows that AI agents and contract review systems don't replace lawyers; they make their expertise more accessible and consistent. For legal professionals, the key takeaway is that codifying a firm's playbooks-the hard-won drafting rules and fixes-turns scattered knowledge into a reusable asset. When that asset is combined with no-code agents and transparent AI review, junior lawyers learn faster, documents are turned around more quickly, and clients see tighter adherence to agreed processes. The technology rewards firms that invest in capturing their own standards, not just adopting off-the-shelf tools.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)