Data to AI™ / Claude/Facebook Ads
Ask Claude which campaigns, placements and creatives earn their budget, using Meta ad data that refreshes on schedule instead of last week's Ads Manager export.
Better answers, fewer tokens.
Claude
Which placements returned the most purchase value per dollar spent last month?
Instagram Reels returned 4.1 in purchase value per unit of spend, the best of the three, on about a third of the budget. Facebook Feed took the largest share at 3.2.
| Placement | Spend | Purchase Roas | Ctr |
|---|---|---|---|
| Instagram Reels Platform Position Device Ad Insights | $8,400 | 4.1 | 1.32% |
| Facebook Feed Platform Position Device Ad Insights | $12,600 | 3.2 | 1.05% |
| Instagram Stories Platform Position Device Ad Insights | $5,200 | 2.4 | 0.71% |
Spend and Purchase Roas come from one placement breakdown the Facebook Ads API supports, not from separate exports matched by hand.
Facebook Ads extracts paid performance from Meta ad accounts, from account down to ad level: spend, delivery, conversions and ROAS, broken down by placement, device, demographics or creative asset.
Join Ad Creatives (by Ads) with Facebook Ads Insights. Which Creative Title and Body patterns share the best CTR and lowest Cost per purchase?
Claude reads the headline and body copy next to delivery results, so the next creative brief starts from what already worked instead of from the loudest opinion in the room.
Query Spend and Purchase (value) by Campaign name for last week through the dataddo server and write the ROAS table to reports/meta-weekly.md.
In Claude Code, the agent writes the same report into the marketing repository every Monday. The query logic sits in version control, so numbers do not drift between weeks or analysts.
In Platform Position Device Ad Insights, which Platform Position and Impression Device pairs spent the most per purchase last month?
Claude ranks placement and device pairs by cost, so the media buyer can exclude weak positions in the next campaign setup and show the table that justified the change.
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.
per seat, billed monthly per seat, billed annually
Flexibly deploy Dataddo in any cloud or hybrid environment.
We have a payment model that works for you
| Without Dataddo |
With Dataddo
|
Outcome for you | |
|---|---|---|---|
| Breakdown reports | One Ads Manager download per breakdown, then placement, age and creative files are matched by hand before Claude can compare them. | Ready datasets such as Platform Position Device Ad Insights hold only breakdown combinations the Facebook Ads API supports, and Claude queries them by name. | No stitched exports |
| Long comparisons | Year-over-year questions need many exports, one period at a time, because each download covers a short window. | A full data re-sync loads up to 37 months in 90-day batches, with automatic pagination, so Claude can compare seasons. | Up to 37 months of history |
| Analysis in code | An analyst copies Ads Manager numbers into a notebook and rewrites the cleanup steps every time the export columns change. | Claude Code queries the same ad datasets through the dataddo server while you write scripts, so field names come from the semantic layer. | Reusable report scripts |
| Token consumption | Claude spends tokens exploring raw Facebook 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 Facebook Ads datasets marketing teams use most, with their real field names. Claude queries them by name through the Dataddo semantic layer.
List of curated datasets
Need a dataset, metric, or attribute you don't see?
Tell us what's missing and we'll add it to the connector.
Runs on your schedule with a relative date range that moves forward on every run. The Facebook Ads API returns up to 37 months of history, loaded in batches of up to 90 days.
Facebook Ads connector →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 Facebook Ads source in Dataddo with datasets such as Facebook Ads Insights and Platform Position Device Ad Insights, then attach it to a Claude destination or AI Model. In Claude, add the custom connector Dataddo Data Access with the URL https://headless.dataddo.com/mcp-data. In Claude Code, run claude mcp add and sign in.
No. Claude only reads Facebook Ads data Dataddo has already extracted. Nothing is sent back to Meta, so campaigns, budgets and ads stay untouched.
Reach counts daily unique people, so daily rows cannot be summed into a week. Create a separate source and flow with a weekly breakdown and ask Claude about that one for matching totals.
Up to 37 months, the Facebook Ads API limit. A new source starts from its date range, so run a full data re-sync to load older months, which Dataddo fetches in batches of 90 days.
Yes. Run claude mcp add with the Dataddo URL, sign in with /mcp, and Claude Code queries the same ad datasets while you build reports or scripts in a repository.
Only combinations the Facebook Ads API supports. Mixing demographic and geographic breakdowns can distort absolute values through sampling, so ask Claude about one breakdown at a time.
All of them. The Free plan allows one custom connector, Pro and Max users add it themselves, and on Team or Enterprise an Owner adds Dataddo Data Access once for everyone.