Case Studies | SaaS B2B

Contract review that keeps pace with the deal

Contract Terms Review Agent

About

A SaaS B2B business working with Cloud Combinator on AWS. The client is anonymised at their request.

Challenge

The challenge had four focus areas, each of which had to be solved for the tool to earn a place in a legal team's workflow.

Speed and consistency

Manual review of client contracts is time consuming and varies from reviewer to reviewer. The system needed to apply the same standards every time and return results fast enough to keep pace with live deals.

Faithful redlining in Word

Legal teams work in Microsoft Word with tracked changes, so a summary of issues was never going to be enough. The agent had to produce a genuinely

Governance and auditability

Contract terms carry real financial exposure, so deviations had to be routed to the right approver against the client's approval matrix and thresholds, every action had to be logged, every output had to be in UK English, and every document had to carry a clear "for legal review, not legal advice" disclaimer.

Fit with existing tools

The value only lands if the tool sits where legal already works. The design had to deliver into Word and a Bullhorn-ready audit note, and support access from inside Microsoft 365 Copilot Chat as well as a direct secure interface.

Solution

We structured the build so that legal output quality was proven early, before investment in the heavier document-rendering and orchestration work.

Under 3 min

Target end-to-end review time, median (projected)

6

Specialist review sub-agents in the pipeline

4

Legal documents produced per contract

By the numbers:

  • Under 3 min - Target end-to-end review time, median (projected)
  • 6 - Specialist review sub-agents in the pipeline
  • 4 - Legal documents produced per contract
Changes

The proof of concept was designed against a clear set of acceptance criteria: a genuine redlined Word document with native Track Changes, a populated legal risk and approval summary, a client amendment letter and a Bullhorn audit note, all in UK English with a disclaimer on every page, produced end to end from a single uploaded contract.

  • Faster, consistent reviewsThe same standards are applied to every contract, with a median review target of under three minutes and a worst case within five, against acceptance targets of high clause-detection precision and recall.
  • Faithful redlinesNative Word Track Changes are injected across the body, headers, footers, footnotes and endnotes, attributed and dated, so the output drops straight into a lawyer's normal workflow.
  • Built-in governanceDeviations are routed against the client's approval matrix, every action is written to an audit trail in DynamoDB, a per-deal cost circuit breaker caps spend, and an UK English lint pass and footer disclaimer are enforced on every document.
  • Works where legal already worksThe agent is reachable through a secure private API and, optionally, from inside Microsoft 365 Copilot Chat, with contracts staying in the AWS account throughout.
  • HandoverThe POC ships with defined acceptance criteria and an evaluation harness that measures output quality against golden contracts hand-prepared by the client's legal team.

With the proof of concept proving the pattern, the natural next steps are a path to production and a set of extensions already scoped, including write-back of audit notes to Bullhorn, pulling final signed documents from DocuSign for comparison, and longer-term conversational memory so a returning reviewer picks up exactly where they left off.

AWS Stack

Amazon Bedrock Knowledge Bases

On Amazon Aurora PostgreSQL with pgvector for on-demand retrieval of the client's standard terms and approved clause playbook.

AWS Step Functions

For the auditable, multi-stage review workflow.

AWS Lambda

For the specialist sub-agents and the Word Track Changes renderer.

Amazon DynamoDB

As the system of record for deals, jobs and audit events.

Amazon S3

For contract, template and output storage, encrypted with AWS KMS.

Amazon API Gateway

With Amazon Cognito for secure, private access, alongside an optional Microsoft 365 Copilot Chat frontend.

Amazon CloudWatch

And AWS X-Ray for observability and per-deal cost monitoring.

YOU MIGHT LIKE

Related success stories

View all case studies

Case Studies | Insights

Utilising Language Recognition, Speed, and Enhanced Security to Make Social Media a Force for Good

  • Here, we take a detailed look at how the Cloud Combinator team collaborated with another cutting-edge AI service provider that provides intelligent systems to “make social media more social” for brands and users alike.
  • Arwen AI is a UK-based startup specialising in AI solutions to manage and enhance brands’ social media interactions. Founded in 2020 by Matt McGrory, Dr. David Cole, and Joel Bailey, Arwen. AI focuses on using AI to automatically detect and remove spam, toxic comments, and other unwanted content from social media platforms.
  • The team at Arwen have three core products. ‘Moderate’ is focused on identifying and removing toxic content from social media channels. ‘Engage’ helps brands identify and engage with meaningful conversations on social media, and ‘Customize’ allows brands to apply bespoke algorithms to their channels - creating an even more effective moderation and engagement.
Read more
CONTACT US

Ready to turn AI into impact?

We'll help you spot the highest-value opportunities, reduce risk around your first AI initiative, and define a clear path to results from day one.

Why talk to us:

Outcome-driven recommendations

AWS-recognised delivery expertise

Risk-aware AI adoption

Clear next step, not a sales pitch

Start with a focused 20-minute conversation about your goals — no pressure, no commitment.

This website uses cookies to enhance user experience and to analyze performance and traffic on our website.

See our Privacy Policy for details.