Data to AI™ / Claude/YouTube
Ask Claude which videos earn watch time, where viewers find them and who watches, using your channel data instead of a pile of report exports.
Less tokens, better answers.
Claude
Which traffic sources brought the most watch time to our September uploads?
Suggested videos drove 63% of watch time across these three sources, about twice as much as YouTube search, and held viewers longest at 46% average view duration.
| Traffic Source Type | Views | Watch Time (Minutes) | Average View Duration (%) |
|---|---|---|---|
| Suggested videos Owned Channel Videos Traffic Sources | 48,200 | 212,100 | 46% |
| YouTube search Owned Channel Videos Traffic Sources | 31,500 | 101,000 | 38% |
| External Owned Channel Videos Traffic Sources | 9,800 | 21,600 | 27% |
Views and Watch Time (Minutes) are summed per traffic source; Average View Duration (%) is weighted by Views, not a plain average of videos.
YouTube Organic extracts your channel's non-paid performance: views, watch time, subscribers, traffic sources, audience, comments and playlists, per video and per day.
Which Traffic Source Type and Traffic Source Detail values bring viewers to each upload from last quarter, and which Tags do the winners share?
Claude compares discovery paths with Titles and Tags from Videos Metadata, so the content team spots topics that already attract viewers but have only a few uploads so far.
Summarize the questions viewers ask in Comments and Replies this month, grouped by Video ID, with Engaged Views for each video.
The video team picks follow-up topics viewers have already asked for, because Claude pairs recurring questions with the engagement of the upload they were posted under.
Compare Videos Metadata with Videos by Playlists through the dataddo server and add uploads that sit in no playlist to content/backlog.md.
A marketing analyst keeps the content backlog in the repository, so Claude Code adds orphaned uploads as a reviewable change instead of a spreadsheet nobody opens again.
Data to AI™ is free while in private pre-release. We onboard teams in waves and shape the product with their feedback. At launch, it will be priced like the AI assistants your team already uses. Need tailored contracts or volume pricing? Talk to us about Enterprise.
Serve governed business data to Claude, ChatGPT and other AI tools over MCP.
Free during pre-release
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 | |
|---|---|---|---|
| Joining report types | Traffic, device, demographic and playlist reports are exported one by one and joined on Video ID before Claude ever sees them. | The datasets share Date, Channel ID and Video ID, so Claude combines them through the semantic layer within one conversation. | One question, several reports |
| Look-alike metrics | Views and Engaged Views, or Average View Duration in seconds and in percent, sit side by side in a pasted table with no definitions. | Each field reaches Claude with its business description, so it knows which duration or which view count it is working with. | The right metric each time |
| Commenter details | A comments export carries Author Display Name and channel links into every chat it is pasted into. | Untick fields such as Author Display Name when refining attributes, and Claude only sees the comment text and counts in the flow. | Fewer personal fields in chats |
| Token consumption | Claude spends tokens exploring raw YouTube 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. | Less 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 YouTube 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 the schedule you set, for example daily. YouTube Organic datasets do not use a date range, so every run pulls all currently available data.
YouTube 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 YouTube Organic source in Dataddo with datasets such as Owned Channel Videos Basic User Activity and Owned Channel Videos Traffic Sources, and attach it to a Claude destination or AI Model. Add https://headless.dataddo.com/mcp-data as a custom connector in Claude and sign in, or use claude mcp add in Claude Code.
The Google account you authorize needs at least admin-level permissions. If you lack them, a teammate with the authorizer role in Dataddo can authorize YouTube for you.
Yes, with Owned Channel Videos Demographics. It gives Views Percentage per Age Group and Gender for each video, along with Country Code, so Claude can describe who watches what.
No. YouTube Organic replaces the YouTube Analytics and YouTube Video connectors, so build new Claude setups on YouTube Organic datasets.
Yes. Run claude mcp add with the Dataddo URL and sign in with /mcp. Claude Code keeps the credentials in your system keychain and queries the same YouTube entities from the terminal.
YouTube Organic datasets do not use a date range, so each scheduled run pulls all currently available data. Claude reads what the latest run of the flow holds.
Yes. In the connector's tool permissions, set refresh to ask first while query and describe_entity stay allowed. Claude then checks with you before requesting fresh data.