Case Studies | SaaS B2B

Generative AI for smarter sales outreach

AI Sales Email Generation PoC

About

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

Challenge

The proof of concept concentrated on three focus areas.

Generating relevant emails at volume

The core test was whether a fine-tuned language model could produce targeted emails reliably, with the success bar set at around 1,000 emails per day. Throughput and relevance together, not just fluent text, were what mattered.

Choosing the right AWS platform

The client needed evidence, not opinion, on whether AWS Bedrock or Amazon SageMaker suited their model and budget. The plan was to start on Bedrock and, if results fell short, deploy the model suggested by prompt research on SageMaker.

Keeping the model fresh

An one-off model is not enough for a living sales database. The proof of concept needed a pipeline that could retrain the model on an infrequent but repeatable basis, so quality would hold as the client's data grew.

Solution

Cloud Combinator ran the engagement as a six-stage accelerator, structured so the client gained the platform, the skills and the evidence to continue building in-house.

1,000

Emails per day the model targets (projected)

6 stages

Cloud Accelerator Program, discovery to WAFR

2 models

Suggested plus fine-tuned, on Bedrock

By the numbers:

  • 1,000 - Emails per day the model targets (projected)
  • 6 stages - Cloud Accelerator Program, discovery to WAFR
  • 2 models - Suggested plus fine-tuned, on Bedrock
Changes

Acceptance for the engagement was defined around the Cloud Accelerator Program: sign-off on the delivery scope, confidence in the client's internal technical lead, and a proof of concept that met the original scope while demonstrating Bedrock or SageMaker as a suitable environment to continue developing the client model. The figures below are the proof-of-concept success targets set in scope.

  • A working email generator — A fine-tuned language model designed to draft targeted prospecting emails from the client's filtered database.
  • A platform decision, evidenced — A structured way to prove whether Bedrock or SageMaker fits, starting on Bedrock with SageMaker as the fallback path.
  • A repeatable retraining path — An infrequent retraining pipeline so model quality keeps pace with the client's growing data.
  • Skills left behind — An immersion day and best-practice baseline that upskilled the client's team to continue development themselves.
  • Handover — A Well-Architected Framework Review and internal tech-lead sign-off, so the client owns and manages the new AWS stack with confidence.

With a validated platform, a fine-tuned model and an upskilled team, the client left the accelerator positioned to take the proof of concept forward, scaling AI-drafted outreach as their prospect database and customer base grow.

AWS Stack

Amazon Bedrock

For deploying and fine-tuning the language model behind email generation.

Amazon SageMaker

As the alternative training and deployment environment for the model suggested by prompt research.

Amazon S3

For storing the training data, model artefacts and generated output.

AWS IAM

For the permissions and security controls set up around the proof of concept.

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.