Tutorial
From install to your first AI Visibility Score
PositionBird automatically measures how often ChatGPT, Perplexity, Gemini, Claude, and Google Search results recommend your store, then turns the results into a ranked fix list. Here's exactly what happens, step by step: no code, no setup beyond onboarding, nothing to maintain.
- 01
Install from the Shopify App Store, or create an account with your website2 minutes
One click from the App Store listing. PositionBird only requests the read products permission. It never touches orders, customers, or checkout, and billing runs entirely through Shopify. Not on Shopify? Create an account with your website and PositionBird crawls your public site instead; billing is by card.
- 02
Confirm your brand
Tell us the brand name shoppers actually use, plus any aliases and spellings. This is what we look for inside AI answers. The more complete it is, the more accurate your mention detection.
- 03
Pick the prompts that matter
PositionBird reads your catalog (or crawls your site) and suggests buyer-intent questions your customers ask AI, like "best soy candles for gifts", and you can add your own. These prompts are your tracked battleground.
- 04
Add competitors to benchmarkGrowth+
Name the stores you compete with. Every scan then measures share of voice: when a shopper asks, who does the AI actually recommend, you or them?
- 05
Scans run automatically
On your plan's cadence (monthly to daily), PositionBird queries the AI engines your plan includes (ChatGPT, Perplexity, Gemini, and Claude) through their official APIs, checks Google's classic results (organic top ten, featured snippet, People Also Ask) through a search-results data API, samples each prompt more than once to smooth out AI randomness, and detects every mention and citation of your store and competitors. API answers are a proxy for what shoppers see in the consumer chat apps, not a screenshot of them. The engines, models, sampling and score formula are published on the methodology page.
- 06
Read your AI Visibility Score
Results roll up into a single 0–100 score with a tier (Invisible, Emerging, Competitive, or Leading) plus per-engine and per-prompt breakdowns, a citation ledger of the sources each engine trusted, and a trend line so you can see whether you're gaining ground scan over scan.
- 07
Act on ranked recommendations
The app turns your results into a prioritized fix-list: FAQ content that answers the prompts you're losing, structured data, llms.txt, and coverage gaps, ordered by impact and effort so the next move is always obvious. When a competitor takes a prompt you were winning, you get an alert after the scan that caught it.
See your score in one scan
Free plan available, with no trial clock, no credit card. Install from the App Store or sign up with your website.
Get it on the App Store