Case Studies | SaaS B2B

A profit-first AI assistant for advertising locations

AI Search Assistant

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 about turning location data into recommendations users can trust and act on.

Profit-first recommendations

The platform needed to do more than surface data. It had to reliably produce best-guess, profit-first estimates for prime advertising real estate, ranking a shortlist of the strongest options against the metrics that matter.

Explainable results

A ranking on its own is hard to trust. Every recommendation had to arrive with a clear rationale generated by the model, explaining the reasoning behind its choices and staying accurate to both the metrics and the user's context.

Keeping the data current

Recommendations are only as good as the underlying metrics. The system needed to keep its vector store up to date as Location Live's location metrics changed, so answers always reflected the latest picture without manual intervention.

Solution

An end-user submits a query describing what they need from their advertising real estate, such as location, product and target audience. The query reaches a retrieval-and-generation Lambda, which pulls the relevant metrics from an Aurora PostgreSQL pgvector store, feeds the query and those metrics into an Amazon Bedrock model, and returns a ranked list of ideal locations together with a rationale as a single answer that Location Live can display however it likes.

15 days

Engagement to a deployed serverless platform

2

Purpose-built Lambdas: one to ingest metrics, one to retrieve and generate

100%

Serverless, deployed as code via CloudFormation

By the numbers:

  • 15 days - Engagement to a deployed serverless platform
  • 2 - Purpose-built Lambdas: one to ingest metrics, one to retrieve and generate
  • 100% - Serverless, deployed as code via CloudFormation
Changes

The platform was delivered against its success criteria: a serverless assistant that produces profit-first, ranked recommendations for advertising locations, each paired with a model-generated rationale, running on infrastructure that keeps itself current and deploys as code.

  • Profit-first shortlists — The assistant ranks advertising locations against Location Live's key metrics and returns a best-guess, profit-first list of the strongest options.
  • Built-in rationale — Amazon Bedrock generates a plain-language explanation alongside every shortlist, so users can see why each location was chosen.
  • Retrieval-augmented answers — Aurora PostgreSQL with pgvector grounds each response in Location Live's own metrics rather than the model's assumptions.
  • Self-updating data — A scheduled ingestion Lambda keeps the vector store aligned with the latest metrics, with filterable metadata for accurate retrieval.
  • Handover — Location Live received a working demonstration, documentation and Loom walkthroughs, with the stack deployable into their own AWS account via CloudFormation.

With a serverless, AWS-native foundation in place, Location Live has a recommendation engine that scales with demand and stays current on its own, ready to extend with more metrics, richer rationale and new use cases as the product grows.

AWS Stack

Amazon Bedrock

For generating ranked, profit-first recommendations and their supporting rationale.

Amazon Aurora PostgreSQL

With pgvector for storing location metrics as vectors and grounding answers through retrieval.

AWS Lambda

For the serverless ingestion and retrieval-and-generation functions at the core of the platform.

AWS CloudFormation

For provisioning the whole platform as repeatable infrastructure as code.

AWS IAM

For least-privilege roles around each function and service.

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.