Metrics Dashboard Design and KPI Tracking: Building the Dashboard That Drives Decisions

Dashboard Design Principles

A well-designed dashboard is one page, visual, and focused. **Dashboard Elements** Good dashboard includes: 1. Time period at top (January 2025) 2. Key metrics (9-12 metrics max) 3. Trends (up/down arrows or sparklines) 4. Comparisons (vs target, vs prior month) 5. Color coding (green = good, red = bad, yellow = watch) Example layout: Top row (Revenue): - MRR £100K (target £100K) ↑ +8% MoM - Growth 8% MoM (target 10%) ↓ -2% - ARR £1.2M (running rate) Middle row (Retention): - Monthly churn 2.1% (target 2%) ↗ +0.1% - NRR 115% (target >110%) ✓ - Customer count 125 (up from 123) Bottom row (Health): - Burn £40K/month (budget £35K) ↑ - Runway 20 months (target 12+) ✓ - Headcount 15 (planned 16 by month-end) **Visual Design Principles** 1. Color coding - Green: On target or exceeding - Yellow: Close to target, watch - Red: Missing target, needs action 2. Trend indicators - ↑ Improving (good or bad depending on metric) - ↓ Declining (good or bad depending on metric) - → Flat (depends on metric) Example: - Revenue ↑ = Good - Churn ↑ = Bad - Burn ↓ = Good - Growth ↓ = Bad (unless declining from unsustainable level) 3. Sparklines vs numbers - Number: 125 customers - Sparkline: Small chart showing 120, 121, 123, 125 (trend over time) - Combined: 125 ↑ [small upward chart] 4. Meaningful precision - Don't show: MRR £100,247.38 (too precise, not meaningful) - Show: MRR £100K (rounded, clear) - Exception: Customer churn 2.1% (precision matters for small %) **Dashboard Update Frequency** Daily (real-time): - For very fast-moving metrics (viral products, e-commerce) - Usually not necessary for SaaS Weekly: - Standard for SaaS dashboards - Sufficient for board visibility - Captures weekly patterns Monthly: - For detailed analysis (P&L, full financials) - Board review once per month Most effective: Weekly dashboards for team, monthly detailed for board. **Tools for Building Dashboards** Spreadsheet (Google Sheets, Excel): - Pros: Simple, familiar, flexible - Cons: Manual updates, error-prone, not real-time - Best for: Early-stage, simple dashboards BI Tools (Tableau, Looker, Mode): - Pros: Automated updates, real-time, beautiful - Cons: Expensive (£500-5K+/month), requires setup - Best for: Growth-stage, complex dashboards, data team Investor platforms (Carta, Visible, Lattice): - Pros: Built for investors, includes cap table - Cons: Expensive, less flexible - Best for: Raising capital, board management Native product dashboards (Mixpanel, Segment, Stripe): - Pros: Real-time, integrated - Cons: Only own metrics, not integrated across company - Best for: Specific function (product usage, payments) Most early-stage SaaS start with spreadsheet, graduate to BI tool at £5-10M ARR. **Example Dashboards by Role** CEO Dashboard: | Metric | Value | Target | Trend | |--------|-------|--------|--------| | MRR | £100K | £105K | ↑ 8% | | Growth | 8% MoM | 10% MoM | ↓ | | Churn | 2.1% | 2% | ↑ | | Burn | £40K | £35K | ↑ | | Runway | 20 mo | 12+ mo | ✓ | | Customers | 125 | 130 | ↑ | Sales Dashboard: | Metric | Value | Target | Trend | |--------|-------|--------|--------| | Pipeline | £500K | £600K | ↓ | | Deal win rate | 25% | 30% | ↓ | | Sales quota attainment | 85% | 100% | ↓ | | New customers | 8 | 10 | ↑ | | CAC | £6K | £5K | ↑ | Engineering Dashboard: | Metric | Value | Target | Trend | |--------|-------|--------|--------| | Deploy frequency | 5/week | 5/week | ✓ | | Bug escape rate | 2% | <2% | ✓ | | Performance (P95) | 150ms | <200ms | ✓ | | Feature adoption | 60% | >50% | ✓ | | Tech debt | 15% | <20% | ✓ | Different dashboards for different functions. One CEO dashboard per board, departmental dashboards for teams.

Selecting and Defining KPIs

KPI = Key Performance Indicator (the few metrics that matter most). **What Makes a Good KPI?** Good KPI: 1. Tied to business goal (why are we measuring this?) 2. Measurable (can track quantitatively) 3. Actionable (can do something to improve it) 4. Owned by a person/team (clear responsibility) Bad KPI: - Vanity metric (looks good but doesn't drive decisions) - Lagging-only (can't act on it) - Unclear ownership (nobody responsible) Example good KPI: - "Monthly churn <2%" (goal: retention, measurable, sales/CS owns it, actionable: improve onboarding) Example bad KPI: - "Website visitors" (vanity metric, doesn't drive revenue, hard to act on) **KPI by Function** Sales KPIs: - New customer acquisition (target: 10/month) - Sales productivity (£1M ARR per salesperson) - Win rate (% of qualified leads that convert) - Sales cycle (average days from first call to close) - Pipeline coverage (pipeline / quota ratio, target 3-4x) Marketing KPIs: - Cost per acquisition (target: <£5K) - Marketing pipeline contribution (% of qualified leads from marketing) - CAC payback period (target: <12 months) - Brand awareness (survey/research metric) - Organic traffic (% of traffic from organic search) Product KPIs: - Feature adoption (% of customers using key features) - User engagement (daily/monthly active users) - Product health score (composite of engagement + NPS + churn) - Time to value (days until customer sees first value) - Support deflection (% of issues resolved by product/docs) CS KPIs: - Customer retention (% of customers who renew) - Net Revenue Retention (revenue from existing customers) - Customer satisfaction (NPS, CSAT) - Time to resolution (average days to resolve support ticket) - Health score accuracy (% of at-risk customers flagged correctly) Finance KPIs: - Monthly Recurring Revenue (growth indicator) - Gross Margin (profitability) - Burn Rate (runway indicator) - Payback period (customer acquisition efficiency) - Cash runway (months until out of cash) **KPI Cadence** Frequency of review: Daily: Sales pipeline, website downtime, critical bugs Weekly: MRR, growth %, churn, new customers, support backlog Monthly: Unit economics, customer health scores, expenses vs budget Quarterly: NRR, market analysis, strategic progress Annual: Market share, competitive position, long-term goals Not all metrics reviewed daily. Focus on fastest-moving. **KPI Targets** Set targets based on: 1. Historical performance (how have we done?) 2. Benchmarks (what do peers do?) 3. Strategic goals (where do we want to go?) Example target-setting: Current: 6% MoM growth Benchmark: 8% MoM (fast SaaS) Goal: 10% MoM (aggressive) Targets: - Month 1: 7% (increase from 6%) - Month 2: 8% (continue increase) - Month 3: 8% (stabilize) - Month 4+: 9-10% (reach goal) Targets should stretch but be achievable (if miss consistently, morale drops). **Tracking KPI Progress** Chart KPIs over time: Example MRR tracking: Month: Jan, Feb, Mar, Apr, May MRR: 80K, 86K, 93K, 100K, 108K Target: 85K, 90K, 97K, 105K, 113K Status: On/Below/Below/Below/Below MRR tracking shows growth slowing (starting below target by Apr). Action: Investigate growth drivers (sales productivity? churn increase?). **KPI vs Vanity Metrics** Vanity metrics look good but don't drive decisions: Vanity: "Website visitors up 50% YoY" Real metric: "Website to trial conversion rate 5%" (shows quality) Vanity: "£10M fundraise!" Real metric: "Runway extended to 24 months" (shows impact) Vanity: "5000 customers" Real metric: "Average ARR per customer £10K, 95% NRR" (shows quality) Focus on metrics that matter, not those that sound good. **Reporting KPIs to Stakeholders** Monthly deck for board should include: Slide 1: Key metrics snapshot - MRR and growth % - Churn and NRR - Burn and runway Slide 2: KPI variance - Which KPIs on target? Which off? - Why are we off target? - What are we doing about it? Slide 3: Key wins (progress on goals) - "Launched feature X (increased adoption to 60%)" - "Improved CAC payback from 10 to 8 months" - "New customer segment launched (targeting Y)" Slide 4: Challenges (risks to KPIs) - "Growth slowing (need sales productivity improvement)" - "Churn up in SMB segment (investigating product fit)" - "Burn increasing faster than expected (hiring ahead of revenue)" This format shows KPI performance and explains the story.

Free — no card needed

See this in action for your business

AskBiz tracks these metrics automatically — just connect your data and start asking questions.

Leading vs Lagging Metrics

Not all metrics are equal. Some predict the future, others confirm the past. **Lagging Indicators (Outcome)** Measure what already happened: - Revenue (money received) - Profit (money left after expenses) - Customer churn (customers who left) - Market share (our % of market today) Problem with lagging only: - Can't act on them (event already occurred) - By the time you see decline, problem happened months ago Example: - Month 1: See revenue decline - But sales pipeline was weak 2 months ago (cause happened months earlier) - Now too late to fix current month **Leading Indicators (Input)** Predict what will happen: - Pipeline (potential revenue in future) - Website traffic/trial signups (future customers) - Feature adoption (future expansion revenue) - Customer health scores (future churn) - Employee satisfaction (future turnover/productivity) Advantage: - Can act on them (time to fix before problem) - Month-to-month early warning system Example: - Month 1: See weak pipeline (early warning) - Action: Hire sales support, increase marketing - Month 2-3: Pipeline fills, revenue protected **Leading/Lagging by Function** Sales: Leading: - Pipeline (£500K in opportunities) - Qualified leads (50 in sales process) - Sales conversation rate (% of leads who meet sales) Lagging: - Revenue (deals closed) - Win rate (% of deals that close) - Sales cycle length (how long deals take) Product: Leading: - Trial conversion rate (% who try who convert) - Feature adoption (% using new feature) - NPS trend (sentiment changing) Lagging: - Customer churn (customers who left) - Renewal rate (% who renew) - Revenue per customer (actual spend) Support: Leading: - Support ticket volume (early signal of issues) - First response time (quick feedback) - Issue resolution time (fixing fast) Lagging: - Customer satisfaction (CSAT, NPS) - Support-related churn (customers leaving due to support) - Repeat tickets (same issue, multiple times) **Dashboard Mix** Best dashboards mix both: Example dashboard: Leading (what we expect): - Sales pipeline: £500K (expect £80K revenue next month) - Trial signups: 200 (expect 10 to convert, assuming 5% conversion) - Feature adoption: 60% (expect expansion revenue) Lagging (what happened): - MRR: £100K (actual revenue achieved) - Churn: 2% (customers who left) - Revenue per customer: £8K/month Together they tell story: - Pipeline strong (leading) → expect good month - MRR hitting target (lagging) → pipeline prediction confirmed If pipeline weak but MRR strong: - Discrepancy (doesn't match) - Question: What's driving current revenue? Is pipeline weakening real? - Action: Investigate leading indicator (is pipeline actually weak, or measurement wrong?) **Using Leading Metrics to Steer** Example: Monitor trial signup rate Week 1: 30 signups (on track for 120/month) Week 2: 28 signups (below pace) Week 3: 22 signups (concerning trend) Action (at week 3): - Increase marketing spend - Improve website conversion - Run test campaign Result (if fixed by week 4): - Week 4: 32 signups (recovery) - Month total: 112 signups (slight miss, but recovered) If waited until month-end to see lower revenue: - Too late to fix current month - Would see impact in next month's revenue Leading metrics allow faster response. **Dashboard Example: Full Mix** Daily standup dashboard: | Metric | Type | Value | Target | Trend | |--------|------|-------|--------|--------| | Website traffic | Leading | 5K/day | 6K/day | ↓ | | Trial signups | Leading | 50/week | 60/week | ↓ | | Sales pipeline | Leading | £400K | £600K | ↓ | | Support tickets | Leading | 200/week | 180/week | ↑ | | **MRR** | **Lagging** | **£100K** | **£100K** | **✓** | | Churn | Lagging | 2.1% | 2% | ↑ | | Revenue/cust | Lagging | £8K/mo | £8K/mo | ✓ | | NPS | Lagging | 45 | >50 | ↓ | This shows: - Leading: Website traffic and pipeline weak (warning signs) - Lagging: MRR still on target (because pipeline was strong last month) - Action: Address traffic and pipeline decline now (before it impacts next month's MRR) **Connecting Leading to Lagging** Build model showing relationship: Website traffic → Trial signups → Customer acquisition → MRR (2-3 month lag) Month 1: - Traffic: 5K/day (down 10%) - Prediction: 2-3 months later, MRR will decline (if not fixed) Month 2: - Traffic still down - Actions in Month 1-2 not working - Prediction: MRR decline imminent (Month 3-4) Month 3-4: - MRR declines (prediction confirmed) - Too late to fix current month - But leading indicators warned 2 months earlier Use leading metrics to predict, act before problem reaches lagging metrics.

Common Dashboard Mistakes

How to avoid dashboard pitfalls. **Mistake 1: Too Many Metrics** Wrong: 50 metrics on dashboard (overwhelming) Right: 9-12 metrics max (focused) Every metric dilutes focus. More metrics = harder to see what matters. Solution: Start with 3 core KPIs, add others only if actionable. **Mistake 2: Metrics Without Context** Wrong: "MRR £100K" (no target, no trend) Right: "MRR £100K (target £105K) ↑ +8% MoM" Context matters: - Target (are we on track?) - Trend (is it improving?) - Prior period (what changed?) **Mistake 3: No Ownership** Wrong: "Churn 2%" (nobody responsible) Right: "Churn 2% (owned by VP CS) ↑" (clear owner) Clear ownership enables accountability. **Mistake 4: Lagging-Only** Wrong: Dashboard with only MRR, churn, profit (all lagging) Right: Mix of leading (pipeline, trial signups) and lagging Leading metrics allow faster response. **Mistake 5: Vanity Metrics** Wrong: "Website visitors up 50%" (feels good, doesn't drive decisions) Right: "Trial signups up 20%, conversion rate 5%" (actionable) Focus on metrics that drive decisions, not those that sound good. **Mistake 6: No Updates** Wrong: Dashboard built once, never updated Right: Daily or weekly updates Stale dashboards create distrust (are these numbers real?). Commit to update cadence and stick to it. **Mistake 7: Wrong Audience** Wrong: One dashboard for CEO, Sales, Product, Support (too detailed for each) Right: Dashboard tailored to audience CEO needs: Revenue, churn, burn, runway Sales needs: Pipeline, win rate, sales cycle Product needs: Adoption, NPS, support tickets Support needs: Ticket volume, resolution time, CSAT Different roles care about different metrics. **Mistake 8: Metric Creep** Wrong: Start with 5 metrics, end up with 30 (scope creep) Right: Keep metric count stable, rotate as priorities change Once a year, audit metrics: Still relevant? Or can we retire? **Effective Dashboard Checklist** - [ ] One page (or one screen) - [ ] 9-12 metrics max - [ ] Each metric has target and trend - [ ] Mix of leading and lagging - [ ] Color-coded (green/yellow/red) - [ ] Clear ownership for each metric - [ ] Updated on cadence (daily/weekly) - [ ] Tailored to audience (different for CEO, team, board) - [ ] Visual and easy to scan (no numbers overload) - [ ] Actionable (can do something if metric is off)