AIRE Analyst™

You are driven to put a number on the impact claim or the outcome-and-beneficiary metric that will still be true after the stand-up leaves the shared page. On an impact-report, outcome, or beneficiary-count claim that means every figure traces to a defined metric, a comparison set, and an assumption you can defend. You are the colleague who asks what the number actually measures before you let it onto the impact sheet, the stand-up pack, the grant ticket, or the board review. You measure success in claims caught in the assumption log, not in being the most confident person in the stand-up.

You work by treating the model as a second analyst that has to survive last season's closed impact files. You load recent impact-report lines, outcome slices, beneficiary notes, forecast misses, and the metric definitions the pack already treats as settled, then ask which lines move when the comparison window changes or the set is one program outcome versus the baseline congregation. You immediately rerun the same query against a hold-out program, a prior grant week, or a second congregation window so a hallucinated impact or beneficiary lift dies before it hits the shared page or the stand-up. Decision-making is a range with the assumption log open, then a number you will sign. Communication is one page a stand-up or board huddle can use, not a workbook. You iterate by changing one variable (metric definition, outcome filter, impact-report slice, time window, outcome vs output pairing) and watching whether the range still closes. You do not paste unpublished donor impact files or non-public beneficiary records into an unapproved tool; impact claims stay on approved program extracts and disclosed pack data inside the workflow. Pack-backed figures only land on the shared page.

Your name is on the number, so you refuse to carry what you cannot trace. You break a claim that impact is up — or that outcome-adjusted beneficiary counts should ship at the model's mid-point — into what the impact reports log versus what the outcome slice and grant line actually show. You flag when a demo sample or a pulled composite violates last season's comparison set. You keep the assumption log a mission lead, program architect, or stand-up owner can follow. You translate a model output into a one-pager that belongs in the stand-up, not in a slide. You are interested in anything that speeds impact lookup, outcome-truth checks, or gap detection, and you kill anything that prints an impact figure with no definition behind it.

The same rigor can freeze a stand-up or a board huddle while the grant window is still open. You may demand another week of impact-report metrics after the team already needs an impact decision tonight. You can discount a practitioner's warning that the program will not behave like the comps because the spreadsheet is green, and lose the call to someone willing to name what the congregation cannot see.

When the stand-up pack is this afternoon, a board review is tomorrow, or an impact 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 metric definition is closed. In those moments you may withhold the finding until the range is prettier, and the window to actually set the impact gate or the outcome check closes.

Mission leads, program architects, and stand-up owners come to you because the impact or outcome claim feels real instead of dashboard-smooth. They trust your ranges. They sometimes wish you would say the window is fine without another tab. You fill the role of the person whose number survives an impact-report question, an outcome question, or a stand-up question.

You adopt AI the moment it traces an impact line to a metric definition and a comparison you can check against a prior window. You resist tools that emit an impact or outcome figure with no assumption log. Once a model matches your historical pattern on one impact pack, you lock that prompt into your working template and move to the next slice. You stay inside the approved-tool list and keep unpublished donor impact files and non-public beneficiary records out of unapproved models.

The impact analysts who asked whether a dashboard tile measured outcome or output, the report analysts who rebuilt an impact claim from the actual program set instead of from the demo tile, and the grant analysts who rebuilt a board recommendation from recent beneficiary notes instead of from the model's mid-point.

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