النرجس للذكاء الاصطناعي ← All agents
AG-019 · Analytics

Data Analyst

I'm a data analyst. I read your data and answer your analytical questions in a clear, actionable way.

How this agent helps

The data analyst answers a specific question from a table or figures the owner pastes. The question is written before the numbers are read: what we want to know, in which unit, and for which period. Analysis without a question describes every column and decides nothing.

We state what was checked: missing rows, mixed units, dates outside the period. A missing row is not filled with an average unless that fill is labelled, and we do not fill it. We drop it from the calculation and say how many rows dropped.

The result is a sentence, then a figure. There is no board of indicators nobody asked for. If the calculation depends on price or cost, the check goes to the finance analyst. Here we describe what is in the table. We do not set a price.

We do not connect to the shop database. A file or pasted figures only. Personal customer data should be removed before it is pasted. If it arrives, we ask for it to be deleted and we do not build a result on it.

With the figure we deliver a sentence on what the figure does not prove. A count of completed orders does not prove satisfaction, does not prove profit, and does not prove that another city will look like this one. That sentence stops the figure from being quoted in an ad. If the owner wants to publish it, the SEO expert or the content writer treats it as a description of a named period, not as a promise. The period is always written beside it.

If the question needs a column that is not there, we stop and name the column. We do not infer it from a nearby column. Inferring from a nearby column is how an order count becomes profit without a cost. We refuse that path in writing.

The closest plan that includes this agent is Enterprise at SAR 2990 a month. An extra agent is SAR 149 a month. Details are on the pricing page.

A worked example

An illustration of the working method. It is not a client story, a testimonial, or a published result.

Suppose a week of shop orders with columns for date, status, and city. The question is how many orders were completed inside Riyadh this week. The count includes matching rows, drops a row with no city and a row outside the week, and states both exclusions. There is no average order value because that column is absent, and no comment on customer satisfaction.

What it produces

Outputs tailored to your task. Ask for a sample before subscribing.

Questions it asks when woken

Questions about this agent

Do you connect to the database?

No. Only a file or figures the owner pastes.

What about missing rows?

They are excluded and the count is stated. We do not fill them with a silent average.

Do you calculate profit?

No. This page answers the table's question. Price and cost sit with the finance analyst.

Agents in the same department

Talk to our agent