Skip to content

Solutions

Conversational analytics for retail and e-commerce

Conversational analytics lets a person ask a business question in plain words and get the answer as a number, a chart and a reason, without writing a query or waiting for an analyst. In PACX this is Ask PACX, and it answers from your own sales, stock and ad data.

Last reviewed

What does an answer look like?

An example with sample data. You type "How did Myntra do last week?". PACX answers: ₹1.38 Cr net sales, down 9% on the week before. It shows a chart of the last five weeks. It gives the reason: one kurta sold out in sizes M and L on Thursday, about ₹0.11 Cr of lost sales. It says where the answer came from: 11,480 Myntra order lines.

Why can the number be trusted?

An answer uses the same rule for each number as every other PACX screen, so the answer to a question and the number on the home screen are the same number. Each answer shows its working.

The answers come from governed data: each brand has its own workspace, and you choose who sees what.

What can you ask?

Ask about sales, stock or ads. Some examples:

  • Which products are selling slower this month?
  • How did Myntra do last week?
  • Which products run out this week, and where?
  • Which ads lose money?
  • Which stores are behind last year?

Can the answers come to you?

Yes. PACX can send a short message on WhatsApp each morning: yesterday’s net sales, the products about to run out, an ad that is losing money and the stock that has not sold.

Questions

What is conversational analytics?

Conversational analytics lets a person ask a business question in plain words and get the answer as a number, a chart and a reason, without writing a query.

Can I ask PACX questions in plain words?

Yes. Ask a question such as "Which products are selling slower this month?" and get the number and a chart, worked out from your own data.

Where do the answers come from?

From your own data: the order lines, stock and ad spend that PACX reads from your systems. Each answer shows its working.

Who can see the answers?

Each brand has its own workspace, and you choose who sees what.