Listing time cut 94%.
Capacity up 18×.
An AI tool that reads product data in any format — PDF, CSV, text, image or URL — and turns it into brand-voice-aligned, SEO-ready listings ready for the website, in minutes rather than hours.
time per listing
listing capacity
costs per year
on investment
Client
A large kitchenware retailer selling across an extensive online product catalogue.
Goal
To fix the inefficient process of turning manufacturer product descriptions into brand-voice listings ready for website deployment.
Ninety minutes to list a single product.
Every product listing was built by hand. Editors spent up to 90 minutes per item reading manufacturer content, writing copy in the brand’s tone, formatting technical data, and preparing assets for upload.
The source material never arrived in one shape. Data came in as PDFs, CSVs, plain text, images and URLs, each needing its own manual extraction and normalisation before an editor could even start writing. On top of that, every listing had to be checked for brand voice, SEO standards and compliance — work that limited how fast the catalogue could grow and made seasonal peaks and new vendor onboarding a bottleneck.
Three constraints were non-negotiable:
- Up to 90 minutes per product listing
- Data arriving as PDFs, CSVs, text, images and URLs
- Brand voice, SEO and compliance checked by hand
Any format in, brand-voice listings out.
The team built E.V.A. — an Enhanced Virtual Assistant that ingests manufacturer content in any medium and converts it straight into the retailer’s brand voice, outputting web-ready listings in CSV format for direct upload.
Any format in, one brand voice out — twenty finished listings in under two minutes.
Behind that sits a full pipeline: automated OCR and structured-data extraction, taxonomy mapping, a brand-voice model trained on the retailer’s own style guide, and SEO enrichment through keyword insertion and meta description generation. The description generator produces 20 web-ready variants per product in under two minutes, enabling A/B testing and category-specific messaging.
Rollout was phased by vendor family and product category, starting with high-volume kitchen tools to validate quality before scaling. Rule-based checks and human-in-the-loop review covered edge cases during rollout, with monitoring dashboards feeding continuous improvement of the voice model and SEO rules.
Full return on investment in month one.
By automating the repetitive work of listing creation, the retailer reclaimed editorial time for merchandising strategy and creative optimisation, while every listing stayed brand-aligned, SEO-ready and compliant at scale.
* Case studies reflect work undertaken by our Heads of AI either during their tenure with Head of AI or in prior roles before they were part of the Head of AI network; they are provided for illustrative purposes only and are based on conversations with our Heads of AI.
More case studies.
Your biggest pain point.
Fixed in 14 days. 50% off.
This started with one conversation. Book a 30 minute brainstorm call — we’ll plan your first AI project together and issue your 50% discount code. No payment today.
*Case studies reflect work undertaken by our Heads of AI either during their tenure with Head of AI or in prior roles before they were part of the Head of AI network; they are provided for illustrative purposes only and are based on conversations with our Heads of AI.