How we use AI

DashDash Hub writes the plain-language reads you see on your dashboards and in your weekly report. Those sentences are generated by AI. This page explains exactly what that means: what the AI sees, what it does with it, what it is not allowed to do, and how to switch it off.

This page is a plain-English companion to our Privacy Policy. Where the two describe the same thing, the Privacy Policy governs.

1. What the AI actually does

It writes commentary on numbers we have already calculated. Specifically:

• the short read under each chart on your dashboards;
• the commentary at the top of each dashboard;
• the written sections of your weekly report.

It does not calculate your metrics. Every figure on your dashboards is computed by us from your connected sources; the AI receives those finished numbers and writes about them. If a number is wrong, the cause is the data or our calculation — not the AI.

2. What the AI is sent

For each dashboard and each period, we send a summary of that dashboard's numbers:

• totals for the period and the period before it, and the change between them;
• breakdowns you can already see on that dashboard — by channel, by campaign, by funnel stage, by status;
• your monthly targets, where you have set them, and how the month is pacing;
• the names of the charts on the page and the words we use for each metric, so the writing matches what you are looking at;
• how many working days, weekend days and public holidays the period contained.

Campaign names, channel names and search terms are sent as they appear in your own accounts, because the commentary has to be able to name them. If your campaign names contain information you would rather not share, rename them or turn AI off for your account.

3. What the AI is not allowed to do

• It cannot invent numbers. Every figure it cites is checked against the data we sent it. If a cited number does not match, the commentary is rejected and the page shows a plain summary instead of publishing it.
• It cannot act. It writes words. It does not change your campaigns, your budgets, your targets or your settings, and it is not connected to anything that can.
• It cannot see other customers. Each analysis covers one brand and one period. Your data is never combined with another customer's.
• It does not decide who sees what. Access to your dashboards is controlled by your account, not by the AI.

4. Who processes it

Provider	What they do	What they receive
Anthropic	Runs the AI model that writes the commentary.	The summary described in section 2, with personal contact details removed.

Anthropic processes the summary to generate the text and returns it to us. The full list of providers we use to run the service is in the Privacy Policy.

5. How to turn it off

AI insights are on by default. You can have them switched off for your account at any time by contacting us. With AI turned off:

• none of your data is sent to the AI provider;
• the written commentary disappears from your dashboards and reports;
• every chart, number, table and export keeps working exactly as before.

Turning AI off costs you the writing, not the product.

6. It can still be wrong

7. Human oversight

The commentary is generated automatically and published to your dashboard without a person reading it first — we tell you this plainly rather than implying a review that does not happen. What we do instead is constrain it: it may only write about the numbers it was given, its figures are checked against those numbers, and it follows a fixed set of analytical rules we maintain. If you see commentary that is wrong or misleading, tell us and we will correct the underlying rule.

8. Changes to this page

We will update this page when what we send, who processes it, or what the AI is allowed to do changes. The date at the top shows the last update.

9. Contact

Questions about how we use AI, or to have it switched off for your account: TBC.