inemaSEO 7 — Systems: build your own open Semrush|INEMA.CLUB-PRO

Lesson 5 · 15 min

bing-ia: Microsoft's official AI citations cross-referenced with Google

Since February 2026, Bing Webmaster has had the AI Performance report: how many times Copilot and Bing's AI summaries cited each of your pages, and the phrases the AI used to search. Today it's the only official, free "I was cited by AI" dashboard. bing-ia stores these exports and cross-references them with Google.

Start the lesson
A person compares two stacks of documents side by side with a highlighter.

In this lesson you will

  1. 1What AI Performance shows (and what it doesn't)
  2. 2Export the CSV and map columns that Microsoft doesn't document
  3. 3The 4 cross-reference lists with gsc-historico
  4. 4Why Bing matters twice

Goal: cross-reference the Bing AI Performance export with Google and pull content ideas from the grounding queries.

01From the Bing dashboard to the four lists

There is no API: the way in is to export the CSV from the dashboard. Tap each step.

Click each block in the diagram to see what it does.

A citation is not ranking or authority; it is the closest official signal to "the AI used my site".

02What to do with each list

Tap each one.

1. Cited + click

The AI uses it as a source and Google sends people. Keep it up to date: date, numbers, authorship.

This is the page most worth protecting.

03Import and cross-check

Run it from the ferramentas/bing-ia folder. The repository example uses FICTITIOUS CSVs (made-up numbers, real inema.pro URLs) cross-checked against the real database from lesson 4.

Terminal · import the exports
cd ferramentas/bing-ia
python3 bing_ia.py importar ../../exemplos/bing-ia/entrada/*.csv --db dados/bing.sqlite
Terminal · cross-check with gsc-historico
python3 bing_ia.py relatorio --db dados/bing.sqlite --gsc-db ../gsc-historico/dados/gsc.sqlite --site inema.pro --saida saida/inema.pro
Terminal · when the tool doesn't recognize the columns
python3 bing_ia.py importar my-pages-export.csv --tipo paginas --colunas "url=<column with the URL>,citacoes=<column with the citations>,data=<column with the date>" --db dados/bing.sqlite
python3 bing_ia.py importar my-queries-export.csv --tipo consultas --colunas "consulta=<column with the query>,citacoes=<column with the citations>" --db dados/bing.sqlite
Example from the repository

Real command, fictitious data (repository example):

FICTICIO_ai-performance-grounding-queries.csv: 6 linha(s) de consultas (colunas {'citacoes': 'Citations', 'consulta': 'Grounding Query', 'data': 'Date'}) → dados/bing.sqlite
FICTICIO_ai-performance-paginas.csv: 6 linha(s) de paginas (colunas {'url': 'Page', 'citacoes': 'Citations', 'data': 'Date'}) → dados/bing.sqlite
38 citações em 5 página(s), 6 grounding queries, 4 ideia(s), citadas sem clique no Google 5, clique sem citação 1 → saida/inema.pro/relatorio.html

04Practice: your citations

No access to the dashboard yet? Do E1 with the sample, and E2 once your site is in Bing Webmaster (course 1, lesson 3).

E1 · Run the sample

10 min

Import the sample CSVs and cross them with the gsc-historico database from lesson 4.

Done when: the terminal shows "38 citações em 5 página(s)" and relatorio.html opens with the four lists.

E2 · Your export

25 min

Export the AI Performance pages CSV for your site and import it. If the tool asks, build the --colunas option with the names from your file.

Done when: your export is in the database (the "N linha(s) de paginas" line) and you've written down the real column names.

05Test what you learned

The explanation shows up after your answer.

Question 1 of 4

Today, what is the only official, free dashboard for AI citations?skill

06Takeaways

OfficialBing AI Performance: citations and grounding queries, no clicks, CSV only.
Cross-checkCited × click on Google in 4 lists; ideas from the queries Google didn't see.
2026 yardstickBing matters twice: Copilot and, according to a Seer study (not OpenAI), ChatGPT.

Going further

Previous lessonNext lesson