Data to AI™ / Gemini/Facebook Public Data
Bring competitor posts and Page profiles into the terminal or VS Code, so analysts and technical marketers build benchmark files and checks without scraping or a Graph API script.
Better answers, fewer tokens.
Gemini CLI
How does our Facebook Page compare with two competitors on audience and buzz?
Bluebird Roasters has the largest audience, but Kettle & Co gets people talking: its Talking About Count is 4.1% of its Fan Count, against 1.6% for Bluebird and 2.3% for our Page.
| Name | Fan Count | Talking About Count |
|---|---|---|
| Northwind Coffee (us) Profile | 48,200 | 1,110 |
| Bluebird Roasters Profile | 93,500 | 1,490 |
| Kettle & Co Profile | 31,800 | 1,300 |
Each Page is its own source with its own Page Identifier; Profile values are current as of the last extraction, not summed over time.
Facebook Public Data reads the public profile and posts of any Facebook Page you name by its Page Identifier, including competitors' Pages. Marketing teams use it to benchmark audience size, posting cadence and post engagement against other brands.
Read Posts for all tracked Pages from the last seven days and write reports/competitor-posts.csv with From, Created Time, Media Type, Likes Total Count, Comments Total Count, Shares Count and Permalink URL.
Every Monday an analyst runs the prompt in Gemini CLI and commits the file, so the reporting notebook and the agency both work from the same week's numbers.
Join Posts to Profile on From Id and ID, compute Likes Total Count plus Comments Total Count per 1,000 Fan Count for each post, and add the result to notebooks/engagement.ipynb.
A data analyst normalises engagement by audience size, so a small rival with loyal fans is not ranked below a large Page that posts more often.
Compare the follower figures quoted in content/compare/*.md with the latest Fan Count and Followers Count in Profile and list every claim that is out of date.
Before a comparison page goes live, a technical marketer runs the check in Gemini CLI and fixes the copy instead of publishing last year's numbers.
Data to AI™ is priced per seat, like the AI assistants your team already uses, with ~25% off on annual billing. Need tailored contracts or volume pricing? Talk to us about Enterprise.
Serve governed business data to Claude, ChatGPT and other AI tools over MCP.
$10.49/month
per seat
3 seats minimum
Flexibly deploy Dataddo in any cloud or hybrid environment.
Custom
We have a payment model that works for you
| Without Dataddo |
With Dataddo
|
Outcome for you | |
|---|---|---|---|
| Getting the data | Someone writes a Graph API script per Page, handles paging and tokens, and keeps it running on a laptop. | Dataddo extracts Posts and Profile on a schedule, and Gemini CLI queries them by entity and field name through the semantic layer. | No scraping scripts to maintain |
| Fair comparison | Raw like counts favour the biggest Page, and normalising them means joining two exports by hand. | From Id in Posts matches ID in Profile, so Gemini joins each post to its Page's Fan Count in one prompt. | Engagement normalised by audience |
| Numbers in published copy | Competitor follower figures in marketing pages are copied once and quietly go stale. | Gemini CLI checks the copy in the repo against the latest Profile extraction before each deploy. | Claims checked before release |
| Control in the terminal | A script with API access runs every request without asking anyone. | Gemini CLI asks you to confirm each Dataddo tool call, and Always allow can be set only for tools you trust. | Each lookup approved |
| Token consumption | Gemini spends tokens exploring raw Facebook Public Data columns, sampling rows and retrying until it understands the data. | Field definitions and relations are supplied up front, so the model skips the exploration and goes straight to the answer. | Fewer tokens per answer |
| Answer quality | Prone to wrong joins, invented fields and numbers that are hard to check. | Answers grounded in defined fields. In the Dataddo benchmark, 78.5% of questions were answered correctly, against 65.5% with plain CSV files. | Better answers |
The Facebook Public Data datasets marketing teams use most, with their real field names. Gemini queries them by name through the Dataddo semantic layer.
| Dataset | Metrics and attributes |
Ask Gemini
|
|---|---|---|
| Posts |
|
Which posts since 1 July have a Comments Total Count above the Page's median, with their Permalink URL? |
| Profile |
|
Which tracked Pages list a Category List Name different from ours, and what does their About text say? |
We asked 43 pre-registered questions of three real datasets - Google Search Console, HubSpot deals and Google Ads - and delivered the same rows to the same model in four ways. Only the data layer changed.
of questions answered correctly - the best of the four delivery methods tested
fewer confidently wrong answers - 14.0% of runs against 24.0% and 26.3%
more correct answers on messy CRM data - better on 8 questions, worse on none
Same model, prompts and rows in every condition; 759 graded runs, ground truth frozen before any run. On Google Search Console alone, documented CSV files did as well as Data to AI™ - the advantage comes from messier data such as HubSpot deals.
Read the benchmark →Create Facebook Public Data sources in Dataddo, one per Page Identifier, with Posts and Profile, and attach them to a Google Gemini destination or AI Model. Add a dataddo server with httpUrl https://headless.dataddo.com/mcp-data to ~/.gemini/settings.json, restart Gemini CLI, run /mcp auth dataddo and sign in.
In Gemini CLI, url falls back to SSE, which Dataddo does not support. Use httpUrl and leave out headers and any oauth block; you sign in with /mcp auth dataddo.
Yes. Agent mode in VS Code reads the same settings.json as Gemini CLI. In IntelliJ, Gemini Code Assist uses its own mcp.json, so add the Dataddo server there.
Use From Id in Posts and ID in Profile, which refer to the same Page. That lets Gemini put each post next to its Page's Fan Count, Followers Count or Category List Name.
Yes. The Page Identifier accepts the name from the Page URL, such as dataddo from facebook.com/dataddo, or a numeric page ID. Each Facebook Public Data source covers one Page.
Yes. Once signed in, Gemini reaches every flow or Context attached to an AI destination or AI Model in that Dataddo account, not only the Google Gemini ones.
Gemini CLI stores credentials in ~/.gemini/mcp-oauth-tokens.json and renews them automatically. If a session expires or is revoked, run /mcp auth dataddo again. Nothing is ever written back to Facebook.
