Skip to content This library is free, offered as dana. If it helps your work, donate here →
penguin tree ai
← back to libraryDonate here

This is one of 3,622 free prompts in the penguin tree ai library. Sign up once and the whole library opens, with new prompts by email.

open the library →

Metric target setter

Produces a defensible target range for any KPI or OKR metric using both top-down and bottom-up methodologies, with a recommended target, stretch and commit thresholds, driver-tree math, seasonality adjustments, and…

download .md

Purpose

Turn a metric baseline and growth context into a reasoned target range — with a recommended number, commit floor, and stretch ceiling — derived from two independent methodologies (top-down and bottom-up) so the target is both ambitious and defensible when challenged in a planning review.

When to Use

Invoke this skill when the user asks to:

  • Set a quarterly or annual target for a specific KPI or OKR
  • Decide whether a proposed target is realistic or a sandbagged/moonshot number
  • Reconcile a top-down leadership goal with what the team can actually deliver bottom-up
  • Build a defensible rationale for a target before a planning or board review
  • Calibrate commit vs. stretch thresholds for an OKR
  • Recalculate a target after a baseline shift, seasonality change, or new initiative

Inputs Required

Before starting, collect:

  • Metric name: The exact metric being targeted (e.g., MRR, activation rate, NPS, qualified leads)
  • Current baseline: The most recent actual value and the period it covers
  • Historical trend: Last 3–6 periods of the metric if available (values or growth rates)
  • Target period: The window the target covers (quarter, year, specific end date)
  • Growth context: Known drivers changing the metric — new features, headcount, budget, market shifts
  • Top-down expectation: Any goal handed down from leadership or a business plan (if one exists)
  • Metric type: Whether it's a rate/percentage, a count/volume, a currency figure, or a ratio

If historical trend or a top-down expectation is missing, note the assumption and rely on the available methodology, flagging reduced confidence.

Instructions

Step 1: Establish the Baseline and Run-Rate

Normalize the starting point before projecting anything.

  • Confirm the baseline period matches the target period cadence (annualize or de-annualize as needed)
  • Calculate the trailing growth rate: period-over-period % change across the historical trend provided
  • Distinguish organic run-rate (what happens with no new action) from the raw baseline
  • Flag any anomalies in recent periods (one-time spikes, outages, promotions) that distort the trend

Step 2: Bottom-Up Projection (Build the Driver Tree)

Decompose the metric into the levers the team actually controls.

  • Break the metric into its multiplicative or additive drivers (e.g., Revenue = Traffic × Conversion × AOV; Activated Users = New Signups × Activation Rate)
  • For each driver, estimate a realistic delta from the growth context — cite the initiative or capacity change behind each
  • Recompose the drivers into a bottom-up projected value for the target period
  • Produce a conservative and an achievable bottom-up number based on driver assumptions holding partially vs. fully

Step 3: Top-Down Projection (Ambition and Constraint)

Work backward from the business goal and external benchmarks.

  • If a leadership expectation exists, treat it as the top-down anchor; if not, derive one from the required growth to hit a stated business outcome
  • Cross-check against benchmarks: category growth norms, prior-year performance, comparable-company rates
  • Apply a compounding growth rate to the run-rate to express the top-down ambition as a monthly/quarterly path, not just an endpoint

Step 4: Reconcile the Two Methods

The target lives in the tension between them.

  • Compare the bottom-up achievable number against the top-down ambition number
  • If bottom-up ≥ top-down: the top-down goal is safe; consider raising the stretch
  • If bottom-up < top-down: quantify the gap and identify what additional driver movement or investment would close it
  • If the gap exceeds ~30%, flag it explicitly as an unresourced target that will require either new investment or a rescoped goal

Step 5: Apply Seasonality and Confidence Adjustments

  • Adjust the path for known seasonality (Q4 retail lift, summer B2B slowdown, enrollment cycles)
  • Assign a confidence level (High / Medium / Low) based on data availability, trend stability, and driver certainty
  • Widen the range for Low confidence; tighten it for High confidence with stable history

Step 6: Set the Three-Point Target

Deliver a range, not a single number:

  • Commit (floor): The number the team is confident hitting — roughly the conservative bottom-up figure
  • Target (recommended): The number to plan and staff against — the reconciled midpoint
  • Stretch (ceiling): The upside if drivers overperform — near or above the top-down ambition
  • Express each as an absolute value AND as a % change from baseline

Output Format

  • Structure: Sequential sections following the methodology, ending in a three-point target table and a one-paragraph recommendation
  • Length: 700–1,000 words
  • Format: Prose with an embedded driver-tree breakdown and a final target table (Commit / Target / Stretch × absolute value / % change / confidence)
  • Sections: Baseline & run-rate → Bottom-up projection → Top-down projection → Reconciliation → Seasonality & confidence → Three-point target table → Recommendation
  • Tone: Analytical and decisive — comfortable stating a number and defending the math, not hedging

Quality Checklist

Before delivering output, verify:

  • [ ] Both a bottom-up and a top-down number are calculated independently before reconciliation
  • [ ] The driver tree shows the specific levers and the delta assumed for each — not a single blended growth rate
  • [ ] The final output includes all three points (Commit, Target, Stretch) as both absolute values and % change from baseline
  • [ ] Any gap over ~30% between bottom-up and top-down is explicitly flagged as unresourced
  • [ ] A confidence level is assigned with a stated reason tied to data or driver certainty
  • [ ] The recommendation states one number to plan against — it does not leave the choice open

Notes

  • Targets built only top-down get sandbagged in execution; targets built only bottom-up get anchored to last year's ceiling — the reconciliation step is where the real value is
  • A healthy stretch target sits roughly 20–30% above commit; if commit and stretch are nearly identical, the range is too safe to be useful
  • When history is thin (new metric, new product), lean harder on driver-tree math and benchmarks, and default confidence to Low with a wider range
  • Revisit the target if the baseline moves more than 10% before the period starts — a stale baseline invalidates the whole projection

How to use this skill

This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.

Installing in Claude Code:

  • ~/.claude/skills/<folder-name>/SKILL.md — auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)
  • <your-project>/.claude/skills/<folder-name>/SKILL.md — auto-loads only when working in that project (recommended for project-specific skills)

Restart Claude Code or open a new session to register.

Using in Claude.ai, ChatGPT, Gemini, or other AI tools:

  1. Drag this file into a new chat, or copy and paste its contents
  2. Send your task or request as your next message — the AI will apply the skill where it fits

For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills