How to compare your brand's AI share of voice vs competitors.

How people do this without eCommerce Insights

The manual version: you open six browser tabs (ChatGPT, Perplexity, Gemini, Claude, Copilot, Google AI Mode), type the same prompt into each, screenshot the answers, paste them into a Google Doc, and highlight every brand mention. You do this for twenty prompts. You count. You tally. You move to a spreadsheet: prompts in rows, brands in columns, 1 if the brand was cited and 0 if not.

A day later you have a matrix. Your brand was cited in 11 of 20 prompts on ChatGPT, 7 of 20 on Perplexity, 3 of 20 on Gemini. Is that good? You have no idea, because you don't have a running baseline. Next week you run it again to get trend data — same twenty prompts, six engines — and realize you don't have time to keep doing this. The spreadsheet ages out by the next board meeting.

The real problem with manual share-of-voice tracking is not the first scan, it is the recurring scan. Nobody has twelve hours a month to screenshot and tally. The board gets numbers once; the trend line never materializes.

How to do this in eCommerce Insights

1. Add your brand and top three competitors. Enter your brand name and Shopify URL. Add the names and URLs of the three competitors you care most about. eCommerce Insights cross-references each competitor's Shopify catalog (if public) to know which product citations belong to them.
2. Select a prompt set. Pick 30-60 category-typical buyer prompts. eCommerce Insights suggests a default set based on your category and your hero SKUs. Edit, add, or save your own. The prompt set matters a lot — narrow prompts give you precision; broad prompts give you category-level reach. Most brands run both.
3. Run the share-of-voice scan. eCommerce Insights queries ChatGPT, Perplexity, Google AI Overviews, Gemini, Claude, and Copilot with every prompt. It captures each answer, identifies every brand mention, and logs every product citation. First scan for 40 prompts across six engines takes about 25 minutes.
4. Review the share-of-voice dashboard. You see your share as a percentage, per engine and overall, alongside each competitor. The weekly trend line shows movement. Drill into any engine to see the specific prompts where you won, lost, or tied. Filter by product category for granular views.
5. Export a two-slide board-ready summary. Slide one: share-of-voice breakdown per engine with QoQ trend. Slide two: competitor movement commentary ("Brand X gained 8 points on Perplexity, likely driven by their March PDP rewrite"). PDF or PNG. Drop into the deck. Done.

For the ROI framing that pairs with this data, see prove AI search ROI to leadership.

What "good" looks like

• Brand share of voice above 25% across six engines for your category prompt set. Below 15% means entity-recognition gaps; above 40% is strong leadership.
• Product citation share at least half the brand share. If the ratio is worse — lots of brand mentions, few product citations — the gap is PDP-level.
• Stable QoQ movement within a 5-10 percentage-point band. Larger swings usually point to a specific event (competitor launch, rebrand, engine retrain).
• Per-engine coherence. A brand that dominates Perplexity but disappears on ChatGPT usually has an engine-specific indexing issue worth diagnosing.

Related jobs

Related tools

• Shopify SKU Visibility Grader — the free preview focused on per-SKU visibility.
• ChatGPT product visibility checker — single-engine, single-product competitor view.