Best Practice

⭐ Pick the few metrics that actually matter

Tracking everything means acting on nothing. Pick the 3-5 metrics that actually predict a healthy store and ignore the vanity ones.

🎯 Why it matters

Dashboards overflow with numbers, and most are noise that creates the illusion of control. A healthy DTC store is governed by a few metrics: contribution margin, blended MER/CAC, repeat rate, and cash runway. The skill is choosing the few that drive decisions and deliberately ignoring vanity metrics. AI proposes your shortlist from your model; you confirm what maps to real decisions.

💬 My take

Pick the few metrics that actually matter

Tracking everything means acting on nothing, most dashboard numbers are noise that feels like control. A healthy store runs on a handful: contribution margin, blended efficiency (MER/CAC), repeat rate, and cash runway.

AI can propose your shortlist and name the decision each drives; you confirm which map to choices you actually make, because that's a judgment about your business.

Fastest path: one prompt, end to end

🤖 AI prompt — paste into ChatGPT / Claude

You are a DTC analytics advisor.
My stage: [just launched / growing / scaling]  Main goal this quarter: [profit / growth / cash]
Metrics I currently look at: [list]

Do this:
1. Recommend the 3-5 north-star metrics for MY stage and goal (e.g. contribution margin, blended MER, new-customer CAC, repeat/retention rate, cash runway). For each: what decision it drives and the rough healthy direction.
2. Name the vanity metrics I should STOP fronting (raw revenue, follower count, total orders, platform-reported ROAS alone) and why each misleads.
3. Give the one metric that, if it moved the wrong way, should trigger an immediate look.
Don't invent my numbers; this is about which metrics to track, not their values.
Output: shortlist (metric | decision it drives | healthy direction) + vanity-metric stop-list + the tripwire metric.

Or do it in 4 steps

Anchor on contribution margin and blended efficiency. Contribution margin (see contribution-margin) tells you if orders make money; blended MER (total revenue ÷ total ad spend) or new-customer CAC tells you if acquisition is sane across all channels, not per-platform ROAS, which double-counts.
Add a retention signal. Repeat-purchase rate or 90-day cohort revenue. A store that can't bring customers back has to buy every sale forever, this metric decides whether growth compounds or just churns.
Watch cash runway as a tripwire. Months of cash at current burn (see cashflow-basics). It doesn't drive daily decisions but it's the one that, if it drops, overrides everything else.
Actively ignore vanity metrics. Raw revenue (can be unprofitable), follower count (doesn't pay bills), total orders (says nothing about margin), and platform-reported ROAS in isolation (each channel over-claims). Fronting these is how stores feel successful while losing money.

Worked example (labeled): two stores both do $100k/month. Store A: 55% contribution margin, 2.1x blended MER, 40% repeat rate, 6 months cash. Store B: 55% margin, 1.4x MER, 12% repeat, 1 month cash.

Same headline revenue, wildly different health, and only the north-star metrics, not the revenue number, reveal that B is one bad month from trouble.

Review the shortlist monthly; if a metric never changes a decision, drop it and free the attention.

⚖️ Do & Don't

Do

Anchor on your core two — contribution margin (do orders make money) and blended MER/CAC (is acquisition sane across all channels).
Add one retention signal, repeat-purchase rate or 90-day cohort revenue, so you know growth compounds instead of just churning.
Keep cash runway as a tripwire — it won't drive daily calls, but the moment it drops it overrides everything else.
Let AI propose the shortlist and name the decision each metric drives; you confirm which ones map to choices you actually make.

Avoid

Don't front raw revenue — a store can do $100k a month and still be one bad month from trouble.
Don't judge acquisition on per-platform ROAS alone; each channel over-claims, so use blended MER or new-customer CAC.
Don't keep a metric that never changes a decision — review the shortlist monthly and drop the dead weight.
Don't let follower count or total orders onto the board; they don't pay bills and say nothing about margin.

💡 Quick tips

Two stores at identical revenue can have opposite health — only the north-star set (margin, MER, repeat, cash) reveals which is fine.
Tell AI your stage and this quarter's goal (profit / growth / cash) so the shortlist fits your actual situation.
Name one tripwire metric whose wrong-way move triggers an immediate look, before it becomes a crisis.

🏢 Brand in focus

BrandReported a net loss of about $92M on roughly $439M revenue in 2019, the year before its IPO

Casper — headline revenue that hid the real numbers

Casper.com · Casper fits because it's the cautionary version of this note: it grew revenue fast enough to IPO in 2020, and revenue was exactly the vanity number that looked healthy. The metrics that actually governed the business — contribution margin and customer acquisition cost — told the opposite story; heavy marketing spend and thin per-order economics meant it lost money as it grew. The lesson is the note's core: a big revenue number can sit on top of unhealthy north-star metrics, and only tracking margin, CAC and cash tells you whether growth is real. The watch-out is that Casper's problem wasn't tracking revenue — it's that revenue was fronted while the governing metrics were ignored until the losses were public.

📚 Go deeper — resources

article

Amplitude — What is a North Star Metric (the one-metric framework)

article

Investopedia — Key Performance Indicator (KPI), explained

tool

Google Sheets (free — build and review your 5-metric shortlist)

🏷️ Tags

moneymetricskpiboth

🔗 Related