When somebody in a DTC eCommerce team asks why margin dropped last month, the dashboard shows the drop and stops there. Eventually, the question becomes a Slack message to the one person who can pull the numbers. However, in most situations the answer lands two days later, after the business decisions have already been made. The iQ Business Analyst is built to end that loop.
Ask it in plain English, in the moment, and it comes back with the answer, a chart, and the SQL it ran, straight from your certified data. It works off your own business logic, so "margin" and "new customer" mean what they mean in your business, and the answer is one your sharpest analyst would sign off on. In this article, we will walk through what it answers, how it works, and how it stays accurate.
What the iQ Business Analyst Is
The iQ Business Analyst is the business-question half of Saras iQ, an AI analyst built for eCommerce that turns a plain-English question, such as what contribution margin was by channel last week, into a written answer, a chart, and the SQL behind it, all grounded in your certified data. It is powered by Claude.
The word "analyst" is the part that matters. It is easy to pull the rows, but equally hard to interpret them the right way. The iQ Business Analyst does that interpreting by reading your certified data. It works in the same the way a teammate would and come back with the answer you asked for. It is one half of the Saras iQ AI Data Team, working alongside the iQ Data Engineer, which fields the questions about the data itself.
Core Capabilities of iQ Business Analyst
The iQ Business Analyst is there the moment someone has a question, at 11pm or five minutes before a Monday board call. These are the five capabilities:
How the iQ Business Analyst Works
Behind every answer the iQ Business Analyst generates, it runs a short pipeline. Five specialized agents each handle one step, and you can watch every step, so nothing is hidden behind a black box.
A question moves through five steps:
Most questions come back in about 40 to 60 seconds. The summary, SQL, and chart you can open on any answer come out of the last three steps. And because each agent owns one narrow job, there are fewer places for the answer to go wrong.
What It Answers: with a Focus on Contribution Margin
The iQ Business Analyst works across three areas of the business: contribution margin, customer cohorts and segmentation, and sales and marketing performance. Margin is where it does its best work.
Contribution margin, answered daily and down to the SKU
Contribution margin usually shows up once a month, after finance stitches it together from Shopify, the ad platforms, an ERP, and a spreadsheet. By then the decisions it should have shaped are already made. The iQ Business Analyst answers margin on demand, certified daily, weekly, or monthly, by channel across Shopify, Amazon, and retail in one answer, and down to the individual SKU. Every figure comes with the SQL behind it, so finance can trace it before it lands in a board deck.
The customer and marketing questions it handles too
Beyond margin, the iQ Business Analyst runs the customer and marketing questions teams ask every week. On the customer side it handles cohorts and RFM-based segmentation, including which customers are drifting toward churn, the kind of analysis behind customer segments that predict repeat purchase.
On the marketing side it reports channel and campaign performance like ROAS and blended CAC, measured against your warehouse rather than the ad platform's own numbers. Recurring leadership reporting, like a weekly business review, can run on the same engine.
How It Stays Accurate
Accuracy in the iQ Business Analyst is engineered before your team ever logs in. Four things make an answer trustworthy:
- A context layer trained on your business: our data experts load your business logic, table and column descriptions, SQL templates, and default rules, so iQ reads every metric the way you define it.
- Validation against your own questions: before go-live, iQ is tested against 30 to 100 of your historical questions through automated checks, manual review, and your own sign-off, confirming its SQL matches your certified numbers.
- It says so when it cannot answer: if the data is not there, iQ tells you plainly.
- Security by design: the language model never connects to your warehouse, iQ writes SQL first and fetches with read-only access, none of your data trains the models, and Saras Analytics is GDPR and SOC 2 compliant.
Because those definitions are set once and applied to every query, the same question returns the same number for everyone who asks. The full mechanics of the context layer are covered in <u>how to make AI answers accurate for eCommerce</u>.
Where You Use It
You reach the iQ Business Analyst on three surfaces, and all three run on the same certified data and business logic underneath.
The Slack experience has its own walkthrough, and so does iQ MCP.
The iQ Business Analyst in Action
The clearest sign the iQ Business Analyst works is what happens once a whole team has it. Over 30 days, Ridge answered 1,069 questions across 18 people spanning the CEO, marketing, product, operations, and finance, at roughly 35 questions a day, with zero analyst requests. Every one was self-serve.
.png)
The pattern holds on the margin side too, where daily reporting replaces the monthly scramble.
.png)
The Payoff by Role
The value lands differently depending on where you sit.
Conclusion
The brands pulling ahead are the ones deciding fastest on numbers they trust, while everyone else is still reconciling last month's. Every week your team spends chasing whose figure is right is a week a competitor spends acting on theirs. Put the iQ Business Analyst in front of your own numbers and see what your team stops waiting for. Talk to our data consultants.
.png)
.svg)
%201%20(1).svg)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.webp)
.avif)
.avif)
.avif)
.avif)
.avif)
.avif)
%20(1).avif)
