Data to AI™ / Gemini/TikTok Ads
Pull TikTok Ads campaign, creative and landing page data into the terminal or your IDE, so notebooks, scripts and pre-release checks run on current paid data.
Better answers, fewer tokens.
Gemini CLI
Which TikTok campaigns converted cheapest last month, and where did the budget go?
Spring Retargeting converted cheapest at $11.40 per conversion on the smallest budget. Creator Launch took the most Spend at $14,200 but cost $19.70 per conversion.
| Campaign name | Spend | Conversion | Cost per conversion |
|---|---|---|---|
| Spring Retargeting TikTok Insights | $4,560 | 400 | $11.40 |
| Creator Launch TikTok Insights | $14,200 | 721 | $19.70 |
| Prospecting Broad TikTok Insights | $9,850 | 612 | $16.09 |
Spend and Conversion come from a Basic report at campaign level summed by Stat Time Day; Cost per conversion is the monthly Spend divided by Conversion, not an average of daily values.
TikTok Ads extracts paid advertising data from TikTok Ads Manager advertiser accounts: spend, impressions, clicks and conversions by campaign, ad group or ad, audience breakdowns by age, gender, country and placement, campaign and ad group settings, ad creatives, Smart+ creative materials and Instant Form leads.
Query Spend, Impressions, Clicks and Conversion by Campaign name and Stat Time Day for the last 90 days and save it as data/tiktok_daily.csv.
A marketing analyst feeds the CSV into a forecasting notebook in the same repository. Gemini CLI asks before each Dataddo call, and the prompt sits in the commit message so the pull can be repeated.
List every Landing Page URL from active TikTok Ads and check each path against the routes in src/pages, flagging any that the release removes.
Before a web deploy, a developer runs the check from the terminal and catches ads that would land on a deleted page, then tells the paid team which ads to update before the release goes out.
From Smart+ Creatives Metrics by Campaign, compute Clicks per 1,000 Impressions for each Main Material Name and write the top 20 to reports/smart-plus.md.
In Gemini Code Assist agent mode in VS Code, a technical marketer turns the question into a small script, so the Smart+ creative ranking runs the same way every week.
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 | |
|---|---|---|---|
| Data for scripts | An analyst downloads a TikTok Ads Manager CSV, renames columns by hand and commits a file nobody can refresh. | Gemini CLI queries TikTok Insights through the dataddo server, so the script can be rerun on fresh data with the same field names. | Repeatable data pulls |
| Release safety | Developers ship site changes without knowing which TikTok ads still point at the pages they move or delete. | The Ads dataset lists Landing Page URL and Operation Status, so a pre-deploy check compares live ads with the site routes. | Fewer broken ad clicks |
| Access across tools | Each script stores its own TikTok API token, and nobody knows which one is still valid. | Gemini stores Dataddo credentials in ~/.gemini/mcp-oauth-tokens.json and renews them, while TikTok sign-in stays with the Dataddo authorizer. | No API tokens in repos |
| Token consumption | Gemini spends tokens exploring raw TikTok Ads 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 TikTok Ads 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
|
|---|---|---|
| TikTok Insights: Basic report by campaign |
|
Which Campaign name had a falling CTR over the last 14 days while Impressions kept rising? |
| TikTok Insights: Basic report by ad |
|
Which ads had CTR above the account average last week, and how much Spend did they take? |
| TikTok Insights: Audience report |
|
Which Age group produced the most Clicks per unit of Spend last month? |
| Campaigns |
|
How is the total Budget split by Objective Type across our TikTok campaigns? |
| Ad Groups |
|
Which Ad Group Name entries run on more than one of the Placements available? |
| Ads |
|
Which Ad Name entries share the same Ad Text but send traffic to a different Landing Page URL? |
| Smart+ Creatives Metrics by Campaign |
|
How is Smart+ Spend spread across creative materials for each Campaign ID? |
| Lead Generation Leads |
|
On which weekdays are most leads created, based on Created Time? |
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 a TikTok Ads source in Dataddo with TikTok Insights, Ads or Smart+ datasets and attach it to a Google Gemini destination or AI Model. Add Dataddo to ~/.gemini/settings.json with httpUrl https://headless.dataddo.com/mcp-data, restart Gemini CLI, run /mcp auth dataddo and sign in.
The documented setup covers Gemini CLI in the terminal and Gemini Code Assist in the IDE. VS Code agent mode reads the same settings.json; in IntelliJ, add the server to the mcp.json of Gemini Code Assist.
Check that settings.json uses the httpUrl key. The url key falls back to SSE, which Dataddo does not support. Leave out headers and any oauth block, restart Gemini CLI and run /mcp auth dataddo.
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. Attach only the TikTok Ads data you want queried.
No. Gemini queries extracted TikTok Ads data through the semantic layer by entity and field name. Dataddo never accepts SQL from the model, and nothing is written back to TikTok.
TikTok Insights, Ads and the Smart+ datasets read a relative window that moves forward on each run. Run a full data re-sync with a wider date range once, and Gemini can query the older months.
Tokens are renewed automatically. If a session expires or is revoked, run /mcp auth dataddo again and sign in with your Dataddo email and password, with MFA if prompted.
