Skip to content
toby
IrisData analyst

Iris, your AI marketing analyst

Most founders open Google Analytics, see a chart go down and close the tab. Iris is Toby’s AI marketing analyst: she reads your connected tools, shows which channels bring visitors, conversions and revenue, and when something drops she tells you by how much, since when and what usually causes it. She works only with real data. A company with no connected tool has no numbers in Toby, and Iris will say so rather than invent one.

At a glance

Mission
Reads your analytics and revenue tools, shows what each channel brings and flags drops early.
Default autonomy
Ask before publishing
Tools
Google Analytics 4Google Search ConsoleStripeShopifyNotionMeta AdsGoogle Ads

What Iris does

Example outputs

What Iris prepares for you, shown on fictional businesses. Yours are built from your site, your Brand Brain and your data.

RecommendationExample

An anomaly alert with likely causes

Title: Investigate the 31% drop in sign-ups Why: sign-ups fell from 212 to 146 over the last 7 days (−31% vs the week before), while visitors are stable (−3%) and organic clicks are flat. The traffic is still there; something between the visit and the sign-up changed. Likely causes, most probable first: 1. Tracking: the sign-up event stopped firing after a site change. Check GA4’s real-time report while you create a test account. 2. A broken form: test the sign-up on mobile Safari and Chrome. 3. Fewer high-intent visitors: compare the pricing page’s share of visits with the previous week. Cross-check: Stripe shows 38 new trials this week vs 41 last week, so real sign-ups may be fine and the tracking broken. Effort: Low.
Example for Lumio, Toby’s fictional demo invoicing app, with fictional figures.
Channel analysisExample

A weekly read of the channel table

Last 30 days, from GA4 Organic Search: 6,420 visits, 158 sign-ups, 2.5% Direct: 3,110 visits, 71 sign-ups, 2.3% Referral: 540 visits, 39 sign-ups, 7.2% Organic Social: 1,980 visits, 12 sign-ups, 0.6% What matters: 1. Referral converts nearly 3× the average but brings 4% of visits. 70% of it comes from two accounting blogs: worth a partnership conversation (Noah). 2. Social brings visits but almost no sign-ups. The posts link to the home page; linking to the free template page instead is a cheap test (Kai, Lina). 3. No paid channel appears: ad spend isn’t connected, so no return on ad spend can be calculated yet.
Example for a fictional company. Figures are illustrative.
Tracking planExample

A “How did you hear about us?” setup

Where: the last step of sign-up, optional, one click. Question: “How did you hear about Lumio?” Options, in random order: Google search · ChatGPT or another AI assistant · A friend or colleague · My accountant · LinkedIn · X · Reddit · A podcast · Other (free text) Why these options: your analytics shows 3,110 direct visits a month with no known source, and the AI-visibility test found Lumio mentioned in one of three assistant answers. This question tells you which of those channels is real. Review: once about 50 people have answered, share the results with Iris and she compares them with what GA4 attributes.
Example for Lumio. You add the field in your own sign-up form; Toby doesn’t change your site.

The playbooks

  1. 1

    Anomaly alert

    A move of more than 20% in traffic or conversions is rarely noise. The alert comes with the neighboring metrics (traffic, conversions, organic clicks) and the usual causes in order, so you get a starting diagnosis, not just a red number.

  2. 2

    Track the conversion that matters

    From Toby’s library of proven plays: when analytics is missing or doesn’t track the key action (sign-up, purchase, lead). The deliverable is the exact events to set up and where. Every other data-driven decision depends on it.

  3. 3

    Ask “How did you hear about us?”

    When attribution is unclear because of word of mouth, podcasts, AI answers or dark social: the question, the answer options and where to add it. It reveals channels analytics can’t see.

  4. 4

    Send more traffic to the page that converts best

    When one page, often pricing or a feature page, converts far above average but gets little traffic: internal links from the top five traffic pages and a navigation change.

How approval works

  • Iris only reads. Analytics, Search Console, Stripe and Shopify connect with read-only access, and reading runs without asking.

  • If she writes a report into a Notion page you shared, that write is a “To approve” card unless you have put Iris and the company on Autopilot.

  • No connected tool means no numbers: Iris never estimates your traffic or revenue and presents it as data. Example figures appear only in demo companies.

  • Your data stays in your company’s workspace and isn’t shared with other companies, even under the same account.

Security and trust

Tools Iris works with

GA4, Search Console, Stripe and Shopify are ready, read-only connectors, synced every 6 hours. Meta Ads and Google Ads spend will appear once those platforms approve Toby’s app; until then, return on ad spend can’t be calculated from your accounts.

All integrations

Frequently asked questions

What does an AI marketing analyst actually do?

It does the reading most teams skip: checking the numbers every day, spotting a drop early, comparing channels, and explaining what changed in plain language. Iris reads your connected tools, raises alerts with likely causes, and gives the other agents real figures to work with.

Which tools does Iris connect to?

Google Analytics 4, Google Search Console, Stripe and Shopify, as ready-made connectors with read-only access, plus Notion for reports. Other tools can be added through “Add any tool”, with an MCP server or an API connector. Meta Ads and Google Ads wait for the platforms’ app review.

What if I haven’t connected anything?

Then Toby shows no figures for your company, and Iris says so. She can still help you set up tracking: which conversion to measure, which events to create, and how to connect GA4 with a service account. The demo companies show example data so you can see how the dashboard looks.

Can Iris do multi-touch attribution?

Not as a modeling tool. Iris works from GA4’s channel grouping, your revenue tools and simple methods that hold up for small teams, such as the “How did you hear about us?” question. She will tell you when the data can’t support a precise answer.

How often is the data updated?

Connected tools sync every 6 hours. The automatic drop detection compares the last 7 days with the 7 before, so alerts follow your weekly trend rather than one bad afternoon.

Is my data safe?

Access keys are encrypted and never sent back to the browser, connections are read-only for analytics and revenue tools, and each company’s data is isolated. You can disconnect a tool at any time.

Put Iris to work on your business

Create your account, paste your URL and get your growth audit, your Brand Brain and your first three recommendations, already prepared.

Start free trial

7-day free trial, then $99/month per company. Cancel anytime. Already have an account? Sign in