Track software license allocation and usage
Analyze license usage data to surface underused or over-allocated software licenses before the next renewal.
Hiring gaps caught before a req sits open for months, not after.
Step 2 may send candidate pipeline details — potentially including names — to an external AI chat if you don't have a licensed Copilot. Strip candidate names before pasting into ChatGPT or Claude; compare requirement gaps and pipeline counts without identifying individuals.
The apps run in this order. At the AI step, pick whichever of these you have — they're alternatives, not a sequence.
Copilot works differently from the others: with a paid Microsoft 365 Copilot license it's built right into the app above, no copy-pasting needed. Without a license, use it the same way as ChatGPT or Claude — in its own chat window.
Copy the must-have and nice-to-have requirements from each open requisition's Word document into a single Excel Table, one row per requisition, with separate columns for each requirement type — don't leave them as free text you'll have to re-read later.
Export the current candidate pipeline from your ATS alongside it, matched to the same requisition IDs, so the two data sets can sit side by side in the next step.
Paste both the requirements table and the pipeline export, and define exactly what counts as a "gap" in your prompt — a missing must-have skill, a seniority mismatch, or a pipeline simply too thin to fill the role on time are three different problems that need different follow-up.
Vague prompts like "find issues with these reqs" tend to produce generic observations instead of an actionable flag list.
Prompt idea:
Compare the requirements listed in each open requisition against its current candidate pipeline. Flag any requisition where the pipeline is missing a stated must-have requirement, or has fewer than 3 active candidates.
Add an Urgency column to the Table and sort the flagged requisitions by it, not by how large the gap looks — a single missing must-have on a critical, time-sensitive role outranks a three-requirement mismatch on a role with a flexible start date.
Use conditional formatting (Home > Conditional Formatting > Color Scales) on the urgency column so the top priorities are visually obvious before you move to outreach.
Draft one short note per flagged requisition naming the specific missing requirement or pipeline shortfall, and route it to whoever owns that req — a shared summary document that buries five different owners' issues in one long list tends to get skimmed and ignored rather than acted on.
A gap report showing which open requisitions have candidate pipelines that don't match the stated requirements.
Steps 1 and 3 are plain data-gathering and Excel sorting. The AI step earns its place matching each requisition's stated must-haves against pipeline candidates semantically, not just counting rows.
Non-AI alternative: A straightforward Excel filter on candidate count per req covers the volume half of step 3's prioritization without AI.
Analyze license usage data to surface underused or over-allocated software licenses before the next renewal.
Build a Copilot Studio agent that answers employee benefits and policy questions from your HR SharePoint content, escalating only what actually needs a person.
Costs are rising but revenue is flat.
Marketing has one list of users, Sales has another. The names don't match perfectly.
Get practical help with AI consulting or ask about training courses for your team.