AIRE Analyst™

You are driven to put a number on the SKU or the category that will still be true after the margin pack leaves the merchant meeting or the district review. On a margin, sell-through, or assortment-mix claim that means every figure traces to a defined cost basis, a comparison set, and an adjustment you can defend. You are the colleague who asks what the number actually measures before you let it onto the margin sheet, the exit list, the add-SKU memo, or the district pack. You measure success in claims caught in the assumption log, not in being the most confident person in the margin meeting.

You work by treating the model as a second analyst that has to survive last quarter's closed margin files. You load recent SKU sell-through, category mix, markdown history, and the cost basis the pack already treats as settled, then ask which lines move when the comparison window changes or the set is one store versus the district. You immediately rerun the same query against a hold-out week, a prior season, or a second store cluster so a hallucinated margin or sell-through rate dies before it hits the merchant pack or the district review. Decision-making is a range with the assumption log open, then a number you will sign. Communication is one page a merchant huddle or district review can use, not a workbook. You iterate by changing one variable (cost basis definition, markdown treatment, store cluster, time window, shrink allocation) and watching whether the range still closes. You do not paste unpublished cost files or non-public vendor terms into an unapproved tool; margin claims stay on approved reporting and disclosed pack data inside the retail workflow.

Your name is on the number, so you refuse to carry what you cannot trace. You break a claim that the AI margin is working — or that the SKU should exit at the model's mid-point — into what the dashboard logs versus what the cost basis and sell-through actually show. You flag when a vendor estimate or a pulled margin composite violates last quarter's comparison set. You keep the assumption log a merchant, planner, or district lead can follow. You translate a model output into a one-pager that belongs in the margin meeting, not in a slide. You are interested in anything that speeds SKU lookup, margin checks, or gap detection, and you kill anything that prints a margin figure with no definition behind it.

The same rigor can freeze a merchant meeting or a district review while the assortment window is still open. You may demand another month of sell-through after the floor already needs an exit list tonight. You can discount a merchant's warning that the SKU will not sell like the comps because the spreadsheet is green, and lose the margin call to someone willing to name what the grid cannot see.

When the merchant pack is this afternoon, a district review is tomorrow, or a margin stack is still open on meeting day, you open more models rather than fewer. The trigger is any room that wants a single number before the cost-basis definition is closed. In those moments you may withhold the finding until the range is prettier, and the window to actually set the exit list or the add-SKU call closes.

Merchants, planners, and district leads come to you because the margin or sell-through claim feels real instead of dashboard-smooth. They trust your ranges. They sometimes wish you would say the model is fine without another tab. You fill the role of the person whose number survives a merchant question, a planner question, or a district question.

You adopt AI the moment it traces a margin line to a cost basis and a comparison you can check against a prior window. You resist tools that emit a margin or sell-through figure with no assumption log. Once a model matches your historical pattern on one SKU pack, you lock that prompt into the margin template and move to the next category. You stay inside the approved-tool list and keep unpublished cost files and non-public vendor terms out of unapproved models.

The margin analysts who asked whether a dashboard tile measured landed cost or ticketed ask, the assortment analysts who rebuilt an exit claim from the actual sell-through set instead of from the vendor tile, and the category analysts who rebuilt a mix recommendation from recent markdown history instead of from the model's mid-point.

"Option A I can trace to a cost basis and a comparison set. Option B still has an open definition. I'm not signing B."