Data to AI™ / Copilot/Similarweb
Give sales and marketing an agent in Teams that answers how a prospect's or rival's website is doing, so market questions stop landing in one analyst's inbox.
Better answers, fewer tokens.
Microsoft 365 Copilot
How did our traffic and engagement compare with our two closest competitors last month?
Last month ourshop.com had 1.42M visits, behind rivalstore.com at 2.05M but ahead of shopnext.com. Our Bounce Rate was the lowest of the three.
| Domain | Visits | Bounce Rate | Pages per Visit |
|---|---|---|---|
| rivalstore.com Monthly Visit Data | 2,050,000 | 48.1% | 5.2 |
| ourshop.com Monthly Visit Data | 1,420,000 | 41.7% | 5.9 |
| shopnext.com Monthly Visit Data | 980,000 | 52.3% | 4.4 |
Monthly Visit Data has one row per Country, so Visits are summed across countries while Bounce Rate and Pages per Visit are weighted by Visits.
Similarweb estimates website traffic and engagement for your own domain and for competitors. Dataddo extracts visits, bounce rate, pages per visit, visit duration, country split and traffic sources per domain, so marketing can compare sites in plain language.
@Market Watch Which competitor Domain gained the most Visits last month, and did its Bounce Rate go up or down?
Product marketers ask the agent in the Teams channel during the weekly sync and discuss the answer in the thread, without anyone opening Similarweb.
@Market Watch Give me a table of the top five countries by Share of Visits for our Domain, with Rank and Average Time, for a slide.
In Microsoft 365 Copilot, a regional sales lead gets the table while preparing a partner pitch, so the slide shows the same figures marketing uses.
Use the dataddo tools to describe Traffic Sources Overview - Mobile Web and map Source Type and Visits into our metrics config.
In VS Code, GitHub Copilot in Agent mode reads real field names and sample rows, so a marketing engineer wires the data into a dashboard without waiting for an export.
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 | |
|---|---|---|---|
| Who can ask | Only the people with a Similarweb seat can look up a competitor, so everyone else sends requests and waits. | The Copilot Studio agent is published to Teams, and colleagues reach the extracted Similarweb data by typing @ and selecting it. | Market answers for the whole team |
| When the agent calls Dataddo | A general assistant answers traffic questions from public web snippets of unknown age. | The server description in Copilot Studio tells the agent to call Dataddo for website traffic, engagement and competitor questions. | Traffic questions use your data |
| Which domains are covered | Ad hoc lookups cover whichever competitor someone remembered, so comparisons skip sites from one month to the next. | The tracked list is fixed by the Similarweb sources you create, one per Domain, so every answer covers the same set. | A consistent competitor set |
| Monthly market check | An analyst rebuilds the country and engagement comparison from scratch before each business review. | Monthly Visit Data holds Visits, Share, Rank, Average Time and Bounce Rate per Country, ready for the agent to summarize. | Review prep in minutes |
| Token consumption | Copilot spends tokens exploring raw Similarweb 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 Similarweb datasets marketing teams use most, with their real field names. Copilot queries them by name through the Dataddo semantic layer.
| Dataset | Metrics and attributes |
Ask Copilot
|
|---|---|---|
| Visits - all traffic |
|
What is the average weekly Visits figure per Domain for the selected Country? |
| Monthly Visit Data |
|
How do Pages per Visit and Average Time compare by Country for our Domain? |
| Average visit duration - all traffic |
|
Is weekly Average Visit Duration trending up or down for each Domain? |
| Bounce rate - all traffic |
|
Which days show a Bounce Rate spike for our Domain in the selected Country? |
| Pages per visit - all traffic |
|
Which competitor Domain shows the highest Pages Per Visit in the selected Country? |
| Traffic Sources Overview - Desktop |
|
What share of desktop traffic per Source Type is Organic Visits versus Paid Visits? |
| Traffic Sources Overview - Mobile Web |
|
Which Websites appear in the mobile web traffic data next to our Domain? |
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 →Set up a Similarweb source in Dataddo with an API key and datasets such as Monthly Visit Data, and attach it to a Copilot destination or AI Model. In Copilot Studio, add https://headless.dataddo.com/mcp-data as a Model Context Protocol tool with OAuth 2.0, then publish the agent to Teams and Microsoft 365 Copilot.
Yes. Dataddo connects as an MCP tool on a Copilot Studio agent. Once you publish the agent to Microsoft 365 Copilot and Teams, people reach the Similarweb data by mentioning it with @.
Write a clear server description in Copilot Studio. It tells the agent when to call Dataddo, for example only for website traffic, engagement and competitor benchmark questions.
An API key generated in Similarweb under Settings > Account > API, Standard API / Batch API. You enter it once in a Dataddo authorizer. The Dataddo docs do not state which Similarweb plan is required.
Yes, through Monthly Visit Data, which returns one row per Country with Visits, Share and Rank. The other datasets use the Country filter set on the source, with world as the default.
Yes. Add the Dataddo server to .vscode/mcp.json, start it, sign in and use Copilot Chat in Agent mode. On GitHub Copilot Business and Enterprise, an organization admin first enables the MCP servers in Copilot policy.
No. The agent reads data that Dataddo extracted through the semantic layer. Dataddo never accepts SQL from the model, and nothing is written back to Similarweb.
