Product
Data drill-down: from a number to the reason behind it
A total tells you what happened. It doesn't tell you why — for that, you have to narrow the data, cut it a different way, and look at the row that actually moved.
A total is an answer to a question you didn't ask
"Revenue is down 8% this month" is a real number and a useless one. It tells you something moved without saying what, where, or because of whom — a single figure sitting on top of every product, every location, every vendor you carry, with all of that detail averaged away. Stock counts hide the same blind spot: a healthy total on hand can still mean one location is bare while another is sitting on too much of it. The number that would explain either one is buried a few layers down inside a live table somewhere, and a dashboard that only shows the total has no way to hand it to you. Getting from that number to the reason behind it takes three moves: narrow the data, cut it a different way, and open the one row that actually moved.
Narrow it: filters
The first move is narrowing the data instead of staring at the whole of it. Ask Atlas for a report and it queries your synced store data directly — a sales overview, inventory levels, a demand forecast, a replenishment list, lost revenue from stockouts, or stock availability — and every constraint you add shows up as a chip you can take back off. Say "add gross profit," "show the last 90 days," or "just the top 20," and only that changes — the rest of the report stays exactly as it was. Remove a chip and that piece of the filter goes away with it. Nothing about what's included is a guess, because it's sitting on screen where you can see it and take it back off.
Cut it a different way
A narrower dataset still hides the explanation if it's cut along the wrong axis. The same report can be asked for by product, category, vendor, or location, and the total splits differently depending on which one you pick. Revenue down 8% broken down by location might show one store carrying the entire drop while every other location held steady. Broken down by vendor instead, that same 8% might trace to a single supplier's products running short across every store at once. Same total, same period, same 8% — a different explanation depending on which way you asked for it to be cut.
Open it: ask about the row that moved
Once a report narrows down to the row that actually moved, the next question is usually about that row specifically — and the answer doesn't require leaving the report to go find it. Bigger answers already arrive as sortable, expandable tables you can pop open and export to CSV, and sparkle marks sit right next to the numbers on screen: click one to ask about that exact value, no retyping required. A table of top sellers can carry the same instinct one step further — alongside the numbers themselves, a follow-up built from what's already on screen, like asking why one specific product in the list is slowing down, so the row that caught your eye is also the one you can ask about directly.
Or just ask
None of this requires knowing which filter or which grouping to reach for first. Type what you want to know, in plain English, and Atlas builds the report from your live store data — no measure pickers, no filter trees to learn. Suggested questions greet you on every visit, and a report you've asked for before is one click from running again, so the second time you chase the same number down, it's already waiting. Refine the report the same way you asked for it, by saying what should change next, and once it looks the way you want, save it next to the rest of your reports and schedule it to keep arriving — daily, weekly, or monthly — instead of asking again from scratch. Ask a question about a number instead of building a report around it, and Atlas answers directly from the same live data and hands you the action that follows. Either way, the path from a total to the reason behind it is a conversation, not a spreadsheet.