Case Studies | Retail

Turning an inbox of email orders into automatic system entries

AI Automated Ordering

About

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

Challenge

The challenge had three focus areas, each one standing between the team and less manual work.

High volume of manual re-keying

With 150 to 250 orders a day entered by hand, order processing consumes a large share of the team's time. At peak, that manual load becomes a bottleneck exactly when responsiveness matters most.

Unstructured, inconsistent order formats

Orders arrive as free text or as PDF attachments, with no fixed layout. Any automation has to read messy, varied inputs and still pull out the right products, quantities and details to match the OrderWise schema.

Accuracy and trust

To replace a human step, the system has to be dependably accurate, targeting 95% or better, tick the confirmation-email flag in OrderWise, and fail safely by notifying the sales team when it is unsure, so nothing is silently dropped.

Solution

We scoped the work as a proof of concept on the email channel, deliberately kept narrow so it could be validated quickly and be ready before the Christmas peak, with a clear path to production afterwards.

150-250

Orders received by email per day

95%+

Target extraction accuracy for the PoC

100+

Target test orders with no human intervention

By the numbers:

  • 150-250 - Orders received by email per day
  • 95%+ - Target extraction accuracy for the PoC
  • 100+ - Target test orders with no human intervention
Changes

The design delivers an automated path from an emailed order to a created OrderWise entry with no manual re-keying, built to fail safely and to keep a full record of every order it handles. The figures below are the success criteria set for the proof of concept rather than measured production results.

  • Hands-off ingestionAmazon SES and S3 capture every order email automatically, with a S3 event kicking off processing the moment an order lands.
  • AI-powered extractionClaude 3.5 Sonnet on Amazon Bedrock reads free-text and PDF orders and outputs clean JSON mapped to the OrderWise schema, with an optional self-verification step.
  • Straight-through to OrderWiseExtracted orders are submitted via the OrderWise API with the confirmation-email flag set, removing the manual entry step entirely.
  • Safe by defaultEvery interaction is logged in DynamoDB, and failures raise an alert to the sales team so no order is silently lost.
  • HandoverDelivered as a proof of concept with a defined path to production, so the client can move from validation to daily use with confidence.

With the concept proven on the email channel, the intended next steps are hardening the pipeline for production ahead of the Christmas peak and quantifying the time saved per order to establish return on investment. These are forward-looking directions from the proposal rather than delivered outcomes.

AWS Stack

Amazon SES

For receiving inbound order emails to the orders@ mailbox.

Amazon S3

For storing raw emails and extracted content and for event-based triggering.

AWS Lambda

For the serverless email-parsing and LLM-processing steps.

Amazon Bedrock

(Claude 3.5 Sonnet) for extracting order details into structured JSON.

Amazon DynamoDB

For recording every order interaction end to end.

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.