Case Studies | SaaS B2B

Automating the conveyancing contract pack

Contract Pack Agent

About

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

Challenge

The challenge had three focus areas, each mapping to a distinct part of the conveyancing workflow.

Reliable, auditable document retrieval

Ordering official copies from a third-party provider is a fixed sequence of clicks, but it branches on whether a title is registered, freehold, or leasehold, and it must exclude Charge documents and match the correct property. Getting this wrong means the wrong pack, so the retrieval step needed to be deterministic and evidenced, not best-effort.

Accurate drafting from source documents

The contract pack, the Replies to the platform, the Contract Document, and the platform, has to be generated from the retrieved title documents and a Memorandum of Sale. This is where judgement about field mapping matters and where a language model earns its place, provided its output is grounded strictly in the source documents.

Clean integration with the client's portal

To be useful, the workflow had to start from the portal and push its results back into it, filing retrieved and generated documents under Deducing Title and Property Information without manual intervention, so the automation disappears into the way the team already works.

Solution

Cloud Combinator split the build into two phases so the reliable, rule-based work was proven before any generative step was layered on top.

2

Delivery phases, retrieval then generation

3

Conveyancing documents generated per matter

Eu-west-2

London region, in the client's own account

By the numbers:

  • 2 - Delivery phases, retrieval then generation
  • 3 - Conveyancing documents generated per matter
  • Eu-west-2 - London region, in the client's own account
Changes

The engagement was accepted against the success criteria set out in the Statement of Work: automated retrieval of the correct official copies, correct address matching with Charge documents excluded, portal push under Deducing Title and Property Information, and Bedrock-

  • Deterministic where it counts — Rule-based retrieval runs on Playwright and Fargate rather than unconstrained AI clicking, giving reliable, repeatable orders with screenshot and trace evidence for every step.
  • AI reserved for drafting — Amazon Bedrock is applied only in Phase 2, grounded in the retrieved documents and an agreed field mapping, so generated output stays tied to source material.
  • Portal-native — Retrieved and generated documents are pushed back into the client's portal automatically, so the automation fits the team's existing way of working.
  • Secure by design — A third-party provider credentials are held in AWS Secrets Manager, resources sit inside a VPC, and run state is tracked in DynamoDB for a complete audit trail.
  • Handover — A handover session, Loom walkthrough, and documentation were delivered so the client can operate and extend the workflow.

Generated contract documents, all deployed on AWS with IAM controls, audit logging, and a documented handover to the client.

With retrieval and drafting proven on AWS, the client has a foundation that is no longer just a single workflow but a platform suited to the continued development of further conveyancing automations.

AWS Stack

Amazon ECS Fargate

For running the Playwright browser-automation container for stateful, longer-running retrieval sessions.

AWS Step Functions

For orchestrating the routing logic, task sequencing, retries, and failure handling across both phases.

Amazon Bedrock

For AI-assisted generation of the contract documents from retrieved source material in Phase 2.

Amazon S3

For durable storage of retrieved documents, generated outputs, screenshots, and traces.

Amazon DynamoDB

For per-run workflow state and results, queryable by reference.

AWS Secrets Manager

For secure storage of a third-party provider portal credentials.

Amazon CloudWatch

For logs and metrics across every workflow step and container run.

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.

We use cookies

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

See our Privacy Policy for details.