Weekly Operating Review: KPI Variance Diagnosis and Action Plan

Analyzes weekly KPI data against targets, separates signal from noise, proposes ranked root-cause hypotheses, and produces owner-assigned actions and an escalation list for a team operating review.

You get: Headline Summary, KPI Scorecard and root-Cause Hypotheses

More like this

The prompt

weekly-operating-review-kpi-variance-diagnosis-and-action-plan.md
Role: You are an operations analyst and COO advisor who runs weekly business reviews. You are skeptical of single-week swings, precise about variance, and focused on actions.

Context: I lead {team_or_function} at {company_name}. We review performance weekly against targets. The reporting period is {reporting_period}. Our top priorities this quarter are {quarterly_priorities}. The audience for this review is {audience}.

Task: Diagnose KPI variances and produce an operating review that supports one decision: where should the team spend its attention and resources in the next {planning_window}? Do not just describe the numbers; explain likely causes, confidence, and what to do about them.

Inputs (collect before analysis):
- KPI table: metric name, actual, target, prior period, same period last year if available. I will paste it here: {kpi_data}
- Known events in the period (launches, outages, holidays, staffing changes): {known_events}
- Definitions or data quality caveats: {data_caveats}
- Thresholds for escalation (for example variance beyond a set percentage): {escalation_threshold}
If the KPI data is missing or ambiguous, ask clarifying questions first. State all assumptions in a separate list.

Workflow:
1. Validate the data: flag missing values, inconsistent units, or metrics that lack a target.
2. Compute variance to target and to prior period (absolute and percent). Show the arithmetic for any metric you flag.
3. Classify each metric as On Track, Watch, or Off Track using {escalation_threshold}. Mark whether the movement is likely signal or noise and explain why.
4. For each Off Track or Watch metric, give up to 3 root-cause hypotheses ranked by likelihood. Label each as Supported by data, Plausible, or Speculative, and name the single cheapest check that would confirm or reject it.
5. Identify interdependencies, where one metric's miss may explain another.
6. Propose actions: each with owner role, expected effect, effort (S/M/L), and deadline. Rank by impact over effort.
7. List items to escalate beyond the team and what you need from leadership.

Output format:
1. Headline Summary (3 bullets maximum)
2. KPI Scorecard (table: metric, actual, target, variance, prior period, status, signal or noise)
3. Root-Cause Hypotheses (table: metric, hypothesis, evidence label, validation check)
4. Action Plan (table: action, owner role, expected effect, effort, deadline, priority)
5. Escalations and Leadership Asks
6. Data Gaps and Assumptions
7. Questions to Ask in the Meeting (5 maximum)

Constraints: Use only the data I provide. Do not invent benchmarks, industry averages, or causes presented as fact. Do not claim to have live data access. Keep language direct and free of filler.

ChatGPT-specific instructions: If I upload a CSV or spreadsheet, use data analysis to calculate variances and show the calculation method. If I paste the data as text, parse it and confirm your parsed table before analyzing. Use markdown tables and keep the Headline Summary above the fold.

Acceptance criteria:
- Variance math is shown and checks out against the inputs
- Every Off Track metric has at least one hypothesis with a validation check
- Actions have owners, deadlines, and a priority rank
- Speculation is labeled as such
- The summary tells the audience in under 60 seconds where to focus

Quality checks (run silently, then show a short checklist at the end):
- Recalculate three variances at random and confirm them
- Confirm no metric is classified without a target
- Confirm no cause is stated as fact without data support

Example output

AI-written example

Weekly Operating Review: Customer Support, Brightlane Software, Mar 3–9

1. Headline Summary

  • Backlog (640 vs 500) and first response time (5.8h vs 4.0h) are Off Track and are the focus for the next 2 weeks.
  • Ticket volume rose 23% after the v4.2 release on Mar 4, with 2 agents on leave. Capacity, not quality, is the most likely driver.
  • CSAT and reopen rate are within threshold. CSAT rests on 212 responses, so treat the dip as noise for now.

2. KPI Scorecard (Off Track: >15% adverse; Watch: 5–15%; On Track: <5%)

MetricActualTargetVariancePriorStatusSignal/Noise
First response (h)5.84.0+1.8 (+45%)4.3 (+34.9%)Off TrackSignal
Backlog (tickets)640500+140 (+28%)410 (+56.1%)Off TrackSignal
CSAT91.2%92.0%−0.8 pt (−0.9%)92.4%On TrackLikely noise
Reopen rate6.1%6.0%+0.1 pt (+1.7%)5.9%On TrackNoise
Tickets created4,120nonen/a3,350 (+23.0%)UnclassifiedLikely signal

3. Root-Cause Hypotheses

MetricHypothesisEvidenceCheapest check
Response time, backlogVolume spike exceeded staffed capacitySupported by dataCompare tickets per agent-hour with last week
Volumev4.2 release drove a bug or how-to surgePlausibleTag breakdown for Mar 4–9 vs prior week
Response timeLeave gaps left some queues uncoveredPlausibleCheck response time by queue
BacklogAging tickets are stuck awaiting engineeringSpeculativeCount backlog tickets older than 7 days with the "engineering" status

4. Action Plan

ActionOwnerExpected effectEffortDeadlinePriority
Run tag and queue analysisSupport Ops AnalystConfirms the causeSMar 121
Publish v4.2 known-issues macro and help articleSupport LeadDeflects repeat tick
Generated from this prompt by AI. Your results will vary.

Tips and common edits

Works best with a clean table of at least 5 KPIs and two periods of data. Set the escalation threshold yourself so status labels match your business. Use the validation checks to confirm causes before committing resources.

Questions about this prompt

Process Bottleneck Diagnosis and Workflow Redesign With Pilot Plan

Maps a current team workflow, finds bottlenecks and handoff failures, designs a leaner future-state process with roles and automation candidates, and outlines a low-risk pilot with success metrics.

Process Transcript to Standard Operating Procedure

Converts a pasted process description or transcript into an SOP with purpose, roles, numbered steps, checklists and common mistakes.

Repetitive Task to Zapier, n8n or Make Automation Spec

Turns a repetitive task into a build-ready automation spec with triggers, AI steps, human checkpoints, failure handling and a test plan.

4-Week Small Business Launch Plan: $0 and Small-Budget Versions

Builds a 4-week launch plan for a product or service in a specific market, with channels and weekly tasks in a $0 version and a small-budget version.

Customer Support Macro Set for Refunds, Delays, Angry Customers and Escalation

Writes 10 ready-to-send support reply macros in your brand tone covering refunds, delays, angry customers and escalation.

Executive Decision Memo Builder With Options, Risks, and Recommendation

Turns a pending business decision into a one-page-style decision memo with scored options, explicit assumptions, risks, missing information, a recommendation, and next actions an executive can approve or reject.

Standard Operating Procedure From a Screen-Recording Transcript

Turns the transcript of a recorded walkthrough into a numbered SOP with prerequisites, steps, screenshots to capture, decision points, and common mistakes, ready for a team wiki.

Operations Bottleneck Diagnosis and 30-Day Process Fix Plan

Diagnoses where a workflow slows down using your process steps and metrics, then ranks causes and produces a 30-day improvement plan with owners, measures, and risks.

Excel and Google Sheets Formula Builder With a Plain-English Explanation

Writes the formula for what you want a spreadsheet to do, in both Excel and Google Sheets syntax, explains each part, and gives a test case so you can check it works.

Meeting Notes to Action Items, Owners and a Ready-to-Send Follow-Up Email

Turns raw meeting notes or a transcript into decisions, action items with owners and dates, open questions, and a short follow-up email you can send right after the call.

Explain This Error with Ranked Causes

Explains an error and stack trace in plain words, ranks the three most likely causes and gives a confirming check for each.

Monthly KPI Variance Review with Drivers and Action Plan for Leadership

Converts pasted monthly KPI results and targets into a variance analysis, driver hypotheses, and prioritized actions that a leadership team can decide on in one meeting.