Skip to content

Type a term, e.g. API, Shopware or SEO.

    ← All projects

    04 · GEO · E-commerce · Automation

    Automated GEO optimization for an online store

    A large online retailer was barely visible in AI-powered search systems — despite good Google rankings. As the share of AI search grew, it was missing out on qualified traffic.

    Client project · client anonymized (confidentiality) · project from 2025

    • GEO
    • OpenAI API
    • JSON-LD
    • n8n
    • Google Search Console API
    800 pages
    optimized automatically in 3 days
    100%
    of prioritized pages with structured data
    More
    mentions and visits from AI search after around 8 weeks
    according to the client

    01 · Problem

    An online retailer with more than 15,000 product pages ranked well in classic Google searches — but was all but invisible in ChatGPT, Perplexity and Google AI Overviews. As AI searches made up a growing share of total traffic, the retailer was increasingly losing qualified visitors to competitors that had bet on GEO earlier. Manually optimizing thousands of pages was feasible neither economically nor in terms of time.

    02 · Solution

    vona built a fully automated GEO pipeline: the 800 highest-traffic product pages were identified and prioritized via the Google Search Console API. The OpenAI API then generated page-specific, optimized JSON-LD structured data (Product, BreadcrumbList, FAQPage), E-E-A-T-compliant meta descriptions and semantically precise content snippets that AI search systems recognize as citable. An n8n pipeline automated the entire process — from analysis through generation to pushing the results straight into the store's CMS.

    03 · Result

    800 pages were fully optimized in 3 days — by our estimate, this would have taken several months by hand. Structured data is in place on all prioritized pages. After around 8 weeks, the client reported noticeably more mentions in Google AI Overviews and more visits from AI-powered searches. We deliberately give no exact figure here, because the measurement method is not documented. The pipeline also optimizes newly listed products automatically.

    04 · Workflow

    1. Store: 15,000+ product pages
    2. Prioritize: Search Console API: 800 strongest pages
    3. OpenAI API: generates per page, Structured data (JSON-LD), Meta descriptions, Content snippets
    4. Store CMS: n8n pushes it in directly
    5. AI search: more mentions in AI Overviews
    6. NEW PRODUCTS ARE OPTIMIZED AUTOMATICALLY
    An n8n pipeline handles the whole path: out of more than 15,000 product pages, the Search Console API picks the 800 strongest. The OpenAI API generates structured data, meta descriptions and citable content snippets for each page, and n8n pushes them straight into the store CMS. New products are optimized automatically.

    Next project

    Your project here?

    Tell us about your problem. The first conversation is of course free of charge and without obligation.

    Request a project →

    Newsletter

    News from the vona workshop.

    AI tools we actually use, our weekly AI review & insights from our projects – short, practical and without spam.