Data to AI™ / Gemini/Instagram
Feed reel, story and follower numbers from your Instagram Business account into notebooks, scheduling configs and chart scripts, straight from the terminal or VS Code.
Better answers, fewer tokens.
Gemini CLI
Which content format reached the most accounts per post last month?
Reels reached about 2.4x more accounts per post than single images and drew most of the shares. Carousels sat in between on both.
| Format | Posts | Avg Reach | Avg Shares |
|---|---|---|---|
| Reels Reel Lifetime Insights | 14 | 18,240 | 412 |
| Carousels Carousel Post Lifetime Insights | 9 | 9,860 | 133 |
| Images Post Lifetime Insights (Image) | 11 | 7,450 | 61 |
Reach is averaged per post from lifetime insights, not added up across posts or days.
Instagram Organic extracts non-paid insights from Instagram Business accounts: posts, stories, reels, audience demographics and engagement.
Read Online Followers by Hour of Day and update config/posting-slots.yaml with our three busiest hours.
The social team's scheduler reads its posting times from a file in the repository. Gemini CLI proposes the change from real follower activity, and someone reviews the diff before it ships.
Save Caption, Permalink, Views, Avg Watch Time and Shares Count for every reel in Reel Lifetime Insights to data/reels.csv, then add a cell ranking them by Avg Watch Time.
An analyst working in VS Code with Gemini Code Assist gets the file and the ranking cell in one step. Permalinks make it easy to open the reels that held attention and study their first seconds.
Pull Value by Follow Type from Business Follows Unfollows for this quarter and write a matplotlib script that saves charts/net-follows.png.
The chart is regenerated from the scheduled flow each quarter instead of being redrawn by hand, so the slide always uses the same definition of net follows.
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 | |
|---|---|---|---|
| Reel metrics for code | Someone copies reel numbers from the Instagram app into a spreadsheet, one post at a time, before a notebook can use them. | Gemini CLI reads Reel Lifetime Insights through the dataddo server, with Caption and Permalink next to Avg Watch Time and Total Watch Time. | Every reel in one file |
| Where the connection lives | Exported files spread across laptops and repositories, and nobody knows which project uses which copy. | Add the server to .gemini/settings.json for one project, or to ~/.gemini/settings.json in your home folder for all projects. | Set up once per scope |
| Demographic snapshots | A script filters follower demographics by date and returns nothing, because the export never had a date column. | Demographic datasets have no date range; each run stores current values with a Timeframe field that Gemini can inspect before writing the filter. | Filters that match the data |
| Approving calls | An automation token with full Instagram access sits in a dotfile and runs without anyone checking it. | Gemini CLI prompts before each Dataddo tool call, and you pick Always allow only for the tools you trust. | You see each lookup |
| Token consumption | Gemini spends tokens exploring raw Instagram 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 Instagram 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
|
|---|---|---|
| Business Interactions |
|
Save daily Views, Reach and Accounts Engaged for the last 90 days to a CSV for our social notebook. |
| Reel Lifetime Insights |
|
Export Caption, Permalink and Avg Watch Time for every reel to a CSV in the content repo. |
| Story Lifetime Insights |
|
Which stories had many more Profile Visits than Follows, as a markdown table? |
| Business Follows Unfollows |
|
Write Value by Follow Type for each Period Start this quarter to a CSV so a script can chart net follows. |
| Online Followers |
|
Return Online Followers by Hour of Day as JSON for our post scheduling config. |
| Business Follower Demo Country / Age / Gender |
|
Which Age and Gender combinations each hold more than ten percent of followers by Value? |
| Carousel Post Lifetime Insights |
|
List carousel posts whose Shares exceed their Saved count, with the Caption, as a markdown table. |
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 →Sync your Instagram Business account with a Dataddo Instagram Organic source and attach it to a Google Gemini destination or AI Model. Put the dataddo server with httpUrl https://headless.dataddo.com/mcp-data under mcpServers in your Gemini settings.json, restart Gemini CLI, then run /mcp auth dataddo and sign in.
Either works. ~/.gemini/settings.json makes Instagram data available in every project, while .gemini/settings.json inside a repository limits it to that project. Gemini Code Assist in VS Code reads the same file.
No. Gemini only reads Instagram data Dataddo has already extracted, through the semantic layer, and nothing is written back to Instagram. Publishing and replies stay in the Instagram app.
Daily Reach counts unique accounts per day, so a script that adds up seven days counts repeat viewers several times. Extract a weekly or monthly breakdown in a separate source and flow for totals that match.
Yes, for sign-in. Gemini CLI opens the Dataddo login page in a browser on the same machine, where you enter your Dataddo email and password, with MFA if prompted. Google and Microsoft SSO are not offered there.
Instagram Organic covers the Insights of your Business account. Paid results come from the Facebook Ads connector, where Publisher Platform separates Instagram placements; attach that flow to an AI destination and Gemini can query it too.
Up to 2 years, the limit of the Instagram API. A new source starts from its relative date range, so load older months once with a full data re-sync.
