Here's the dirty secret about ecommerce analytics: most store owners are drowning in data and starving for insight.
You have Shopify reports, Google Analytics, Meta Ads Manager, Klaviyo dashboards, and maybe a few other tools — each showing different numbers, using different definitions, and telling slightly different stories about your business.
The standard advice is to track more. More metrics. More dashboards. More data sources.
That advice is wrong.
For stores doing $100K to $10M, you don't need more data. You need the right data, understood correctly, checked regularly, and used to make decisions.
This guide cuts through the noise. You'll learn which metrics actually matter at your stage, how to read them without getting fooled by broken tracking, and how to build a simple analytics routine that takes 30 minutes a week.
Note: This guide assumes your tracking is reasonably functional. If you're not sure whether your numbers are even accurate — if Shopify says one thing and Meta says another — start with our Marketing Data Reliability guide first. That's the plumbing. This is the architecture.
The Analytics Reality Check
Before diving into metrics, let's acknowledge what's broken.
In 2026, 40-60% of conversions never reach your ad platforms due to iOS privacy changes, ad blockers, and cross-device journeys. This means your Meta Ads Manager, Google Analytics, and even your attribution tools are showing you an incomplete picture.
Your Shopify dashboard is accurate for what happened (orders, revenue). But it can't tell you why it happened — which ads, emails, or content drove those sales.
Your ad platforms can tell you what they think drove sales. But they're guessing on 40-60% of conversions and often taking credit for purchases they didn't cause.
This isn't a reason to give up on analytics. It's a reason to be strategic about what you measure and how you interpret it.
THE ANALYTICS REALITY
════════════════════════════════════════════════════════════════════════════
WHAT'S ACCURATE: WHAT'S ESTIMATED:
──────────────── ─────────────────
Shopify orders: ✓ Meta attributed sales: ~60% visible
Shopify revenue: ✓ Google attributed sales: ~60% visible
Email open rates: ✓ "Conversions" in ad platforms: partial
Total customers: ✓ Attribution models: incomplete
RULE OF THUMB:
→ Trust backend data (Shopify) for WHAT happened
→ Use ad platform data DIRECTIONALLY for why
→ Never make big decisions on a single data source
════════════════════════════════════════════════════════════════════════════THE ANALYTICS REALITY
════════════════════════════════════════════════════════════════════════════
WHAT'S ACCURATE: WHAT'S ESTIMATED:
──────────────── ─────────────────
Shopify orders: ✓ Meta attributed sales: ~60% visible
Shopify revenue: ✓ Google attributed sales: ~60% visible
Email open rates: ✓ "Conversions" in ad platforms: partial
Total customers: ✓ Attribution models: incomplete
RULE OF THUMB:
→ Trust backend data (Shopify) for WHAT happened
→ Use ad platform data DIRECTIONALLY for why
→ Never make big decisions on a single data source
════════════════════════════════════════════════════════════════════════════THE ANALYTICS REALITY
════════════════════════════════════════════════════════════════════════════
WHAT'S ACCURATE: WHAT'S ESTIMATED:
──────────────── ─────────────────
Shopify orders: ✓ Meta attributed sales: ~60% visible
Shopify revenue: ✓ Google attributed sales: ~60% visible
Email open rates: ✓ "Conversions" in ad platforms: partial
Total customers: ✓ Attribution models: incomplete
RULE OF THUMB:
→ Trust backend data (Shopify) for WHAT happened
→ Use ad platform data DIRECTIONALLY for why
→ Never make big decisions on a single data source
════════════════════════════════════════════════════════════════════════════The Only 5 Metrics That Matter (Under $10M)
You could track 50 metrics. You shouldn't. Here are the five that actually drive decisions at your stage:
1. Revenue and Orders (Daily/Weekly)
This is your ground truth. Total revenue and order count from Shopify — not from ad platforms, not from attribution tools. What actually hit your bank account.
Track it daily. Compare week-over-week. If revenue drops 20%, you have a problem. If it's up 20%, figure out why so you can repeat it.
REVENUE TRACKING
════════════════════════════════════════════════════════════════════════════
DAILY CHECK (30 seconds):
Today vs. same day last week → Within ±15%? Normal variance.
WEEKLY CHECK (2 minutes):
This week vs. last week → Trend check
This week vs. same week last year → Seasonality check
RED FLAGS:
• Down >20% WoW with no known cause → Investigate immediately
• Down >30% WoW → Stop everything, diagnose
════════════════════════════════════════════════════════════════════════════REVENUE TRACKING
════════════════════════════════════════════════════════════════════════════
DAILY CHECK (30 seconds):
Today vs. same day last week → Within ±15%? Normal variance.
WEEKLY CHECK (2 minutes):
This week vs. last week → Trend check
This week vs. same week last year → Seasonality check
RED FLAGS:
• Down >20% WoW with no known cause → Investigate immediately
• Down >30% WoW → Stop everything, diagnose
════════════════════════════════════════════════════════════════════════════REVENUE TRACKING
════════════════════════════════════════════════════════════════════════════
DAILY CHECK (30 seconds):
Today vs. same day last week → Within ±15%? Normal variance.
WEEKLY CHECK (2 minutes):
This week vs. last week → Trend check
This week vs. same week last year → Seasonality check
RED FLAGS:
• Down >20% WoW with no known cause → Investigate immediately
• Down >30% WoW → Stop everything, diagnose
════════════════════════════════════════════════════════════════════════════2. MER (Marketing Efficiency Ratio)
MER = Total Revenue ÷ Total Ad Spend
Also called Blended ROAS in some circles — same concept, different name.
This is the single most important marketing metric for stores under $10M. It bypasses all the attribution confusion and answers one question: when I spend money on ads, does my total revenue go up proportionally?
MER CALCULATION
════════════════════════════════════════════════════════════════════════════
EXAMPLE:
Total revenue this week: $52,000 (from Shopify)
Total ad spend this week: $12,000 (Meta + Google + TikTok)
MER = $52,000 ÷ $12,000 = 4.3x
INTERPRETATION:
→ For every $1 spent on ads, business generates $4.30 total revenue
→ Includes organic, email, direct — everything
→ Not "ROAS" — this is blended efficiency
─────────────────────────────────────────────────────────────────────────
BENCHMARKS (varies by business):
MER < 3x → Usually unprofitable (unless high LTV)
MER 3-5x → Typical for scaling stores
MER 5-8x → Efficient, possibly under-spending
MER > 8x → Very efficient or not spending enough on growth
════════════════════════════════════════════════════════════════════════════MER CALCULATION
════════════════════════════════════════════════════════════════════════════
EXAMPLE:
Total revenue this week: $52,000 (from Shopify)
Total ad spend this week: $12,000 (Meta + Google + TikTok)
MER = $52,000 ÷ $12,000 = 4.3x
INTERPRETATION:
→ For every $1 spent on ads, business generates $4.30 total revenue
→ Includes organic, email, direct — everything
→ Not "ROAS" — this is blended efficiency
─────────────────────────────────────────────────────────────────────────
BENCHMARKS (varies by business):
MER < 3x → Usually unprofitable (unless high LTV)
MER 3-5x → Typical for scaling stores
MER 5-8x → Efficient, possibly under-spending
MER > 8x → Very efficient or not spending enough on growth
════════════════════════════════════════════════════════════════════════════MER CALCULATION
════════════════════════════════════════════════════════════════════════════
EXAMPLE:
Total revenue this week: $52,000 (from Shopify)
Total ad spend this week: $12,000 (Meta + Google + TikTok)
MER = $52,000 ÷ $12,000 = 4.3x
INTERPRETATION:
→ For every $1 spent on ads, business generates $4.30 total revenue
→ Includes organic, email, direct — everything
→ Not "ROAS" — this is blended efficiency
─────────────────────────────────────────────────────────────────────────
BENCHMARKS (varies by business):
MER < 3x → Usually unprofitable (unless high LTV)
MER 3-5x → Typical for scaling stores
MER 5-8x → Efficient, possibly under-spending
MER > 8x → Very efficient or not spending enough on growth
════════════════════════════════════════════════════════════════════════════MER won't tell you which channel is working. But it tells you if marketing overall is working — which is often enough.
3. Customer Acquisition Cost (CAC)
CAC = Total Marketing Spend ÷ New Customers Acquired
How much does it cost to get one new customer? This anchors your growth math.
If your CAC is $45 and your average first order is $60, you're making $15 profit on the first transaction (before product costs). That might be fine if customers come back. It's a disaster if they don't.
4. Average Order Value (AOV)
AOV = Total Revenue ÷ Total Orders
Simple but powerful. Small AOV improvements compound quickly.
If you have 1,000 orders/month at $75 AOV and you increase it to $85, that's $10,000 more monthly revenue with zero additional ad spend.
Track AOV weekly. Test bundles, upsells, and free shipping thresholds to move it.
5. Repeat Purchase Rate (and LTV)
Repeat Rate = Customers with 2+ Orders ÷ Total Customers
This is the metric most small stores ignore — and it's often the difference between profitable and unprofitable.
A 30% repeat rate means 30% of your customers buy again. At that rate, you can afford a higher CAC because you're getting multiple purchases per customer.
A 10% repeat rate means you're basically starting from scratch every month. Every sale has to be profitable on its own.
The LTV Connection: Repeat rate is the foundation of Customer Lifetime Value (LTV). In 2026, LTV is how smart stores justify high acquisition costs.
LTV CALCULATION (SIMPLIFIED)
════════════════════════════════════════════════════════════════════════════
LTV = AOV × Purchase Frequency × Customer Lifespan
EXAMPLE:
─────────
AOV: $85
Purchases per year: 2.4
Average customer lifespan: 2.5 years
LTV = $85 × 2.4 × 2.5 = $510
─────────────────────────────────────────────────────────────────────────
THE LTV:CAC RATIO
If your LTV is $510 and CAC is $45:
LTV:CAC = $510 ÷ $45 = 11.3x
BENCHMARKS:
LTV:CAC < 3x → Unprofitable acquisition
LTV:CAC 3-5x → Healthy, keep scaling
LTV:CAC > 5x → Very healthy, or under-investing in growth
════════════════════════════════════════════════════════════════════════════LTV CALCULATION (SIMPLIFIED)
════════════════════════════════════════════════════════════════════════════
LTV = AOV × Purchase Frequency × Customer Lifespan
EXAMPLE:
─────────
AOV: $85
Purchases per year: 2.4
Average customer lifespan: 2.5 years
LTV = $85 × 2.4 × 2.5 = $510
─────────────────────────────────────────────────────────────────────────
THE LTV:CAC RATIO
If your LTV is $510 and CAC is $45:
LTV:CAC = $510 ÷ $45 = 11.3x
BENCHMARKS:
LTV:CAC < 3x → Unprofitable acquisition
LTV:CAC 3-5x → Healthy, keep scaling
LTV:CAC > 5x → Very healthy, or under-investing in growth
════════════════════════════════════════════════════════════════════════════LTV CALCULATION (SIMPLIFIED)
════════════════════════════════════════════════════════════════════════════
LTV = AOV × Purchase Frequency × Customer Lifespan
EXAMPLE:
─────────
AOV: $85
Purchases per year: 2.4
Average customer lifespan: 2.5 years
LTV = $85 × 2.4 × 2.5 = $510
─────────────────────────────────────────────────────────────────────────
THE LTV:CAC RATIO
If your LTV is $510 and CAC is $45:
LTV:CAC = $510 ÷ $45 = 11.3x
BENCHMARKS:
LTV:CAC < 3x → Unprofitable acquisition
LTV:CAC 3-5x → Healthy, keep scaling
LTV:CAC > 5x → Very healthy, or under-investing in growth
════════════════════════════════════════════════════════════════════════════THE PROFIT EQUATION
════════════════════════════════════════════════════════════════════════════
CAC × (1 / Repeat Rate) = What you're actually paying per customer
EXAMPLE A: Low repeat rate
──────────────────────────
CAC: $40
Repeat rate: 15%
True customer cost: $40 × (1/0.15) = $40 × 6.67 = ~$267 in marketing
to generate 6.67 orders from cohort
EXAMPLE B: High repeat rate
──────────────────────────
CAC: $40
Repeat rate: 40%
True customer cost: $40 × (1/0.40) = $40 × 2.5 = ~$100 in marketing
to generate 2.5 orders from cohort
→ Same CAC, very different economics
════════════════════════════════════════════════════════════════════════════THE PROFIT EQUATION
════════════════════════════════════════════════════════════════════════════
CAC × (1 / Repeat Rate) = What you're actually paying per customer
EXAMPLE A: Low repeat rate
──────────────────────────
CAC: $40
Repeat rate: 15%
True customer cost: $40 × (1/0.15) = $40 × 6.67 = ~$267 in marketing
to generate 6.67 orders from cohort
EXAMPLE B: High repeat rate
──────────────────────────
CAC: $40
Repeat rate: 40%
True customer cost: $40 × (1/0.40) = $40 × 2.5 = ~$100 in marketing
to generate 2.5 orders from cohort
→ Same CAC, very different economics
════════════════════════════════════════════════════════════════════════════THE PROFIT EQUATION
════════════════════════════════════════════════════════════════════════════
CAC × (1 / Repeat Rate) = What you're actually paying per customer
EXAMPLE A: Low repeat rate
──────────────────────────
CAC: $40
Repeat rate: 15%
True customer cost: $40 × (1/0.15) = $40 × 6.67 = ~$267 in marketing
to generate 6.67 orders from cohort
EXAMPLE B: High repeat rate
──────────────────────────
CAC: $40
Repeat rate: 40%
True customer cost: $40 × (1/0.40) = $40 × 2.5 = ~$100 in marketing
to generate 2.5 orders from cohort
→ Same CAC, very different economics
════════════════════════════════════════════════════════════════════════════Level 2: When You're Ready for More
Once you've mastered the five core metrics, add these for deeper insight:
Contribution Margin (The Profitability Check)
Revenue is a vanity metric. Profit is sanity.
A store can have great MER and still go bankrupt if COGS (Cost of Goods Sold) or shipping costs spike. Contribution Margin tells you what's actually left after the direct costs of making sales.
CONTRIBUTION MARGIN CALCULATION
════════════════════════════════════════════════════════════════════════════
Total Revenue
- COGS (product costs)
- Shipping costs
- Ad Spend
─────────────────────────
= Contribution Margin
EXAMPLE:
─────────
Revenue: $52,000
COGS (35%): -$18,200
Shipping: -$4,200
Ad Spend: -$12,000
─────────────────────────
Contribution: $17,600 (33.8% margin)
─────────────────────────────────────────────────────────────────────────
WHY THIS MATTERS:
Your MER might be 4.3x (looks great!)
But if COGS + shipping eat 43% of revenue...
You're only keeping 33% before overhead.
→ Track monthly at minimum
→ If margin shrinks while MER stays flat, you have a cost problem
════════════════════════════════════════════════════════════════════════════CONTRIBUTION MARGIN CALCULATION
════════════════════════════════════════════════════════════════════════════
Total Revenue
- COGS (product costs)
- Shipping costs
- Ad Spend
─────────────────────────
= Contribution Margin
EXAMPLE:
─────────
Revenue: $52,000
COGS (35%): -$18,200
Shipping: -$4,200
Ad Spend: -$12,000
─────────────────────────
Contribution: $17,600 (33.8% margin)
─────────────────────────────────────────────────────────────────────────
WHY THIS MATTERS:
Your MER might be 4.3x (looks great!)
But if COGS + shipping eat 43% of revenue...
You're only keeping 33% before overhead.
→ Track monthly at minimum
→ If margin shrinks while MER stays flat, you have a cost problem
════════════════════════════════════════════════════════════════════════════CONTRIBUTION MARGIN CALCULATION
════════════════════════════════════════════════════════════════════════════
Total Revenue
- COGS (product costs)
- Shipping costs
- Ad Spend
─────────────────────────
= Contribution Margin
EXAMPLE:
─────────
Revenue: $52,000
COGS (35%): -$18,200
Shipping: -$4,200
Ad Spend: -$12,000
─────────────────────────
Contribution: $17,600 (33.8% margin)
─────────────────────────────────────────────────────────────────────────
WHY THIS MATTERS:
Your MER might be 4.3x (looks great!)
But if COGS + shipping eat 43% of revenue...
You're only keeping 33% before overhead.
→ Track monthly at minimum
→ If margin shrinks while MER stays flat, you have a cost problem
════════════════════════════════════════════════════════════════════════════Cohort Analysis (The Long Game)
In 2026, looking at a single week's data can be misleading. What matters isn't just what your CAC was last week — it's whether the customers you acquired 6 months ago are still buying.
Cohort analysis groups customers by when they were acquired and tracks their behavior over time.
COHORT ANALYSIS: THE QUESTION THAT MATTERS
════════════════════════════════════════════════════════════════════════════
COHORT: Customers acquired in September 2025
Month 1 (Sep): 100 customers, $8,500 revenue
Month 2 (Oct): 22 customers rebought, $2,100 revenue
Month 3 (Nov): 18 customers rebought, $1,750 revenue
Month 6 (Feb): 12 customers rebought, $1,200 revenue
TOTAL VALUE FROM COHORT: $13,550 from 100 customers = $135.50 LTV
─────────────────────────────────────────────────────────────────────────
THE INSIGHT:
If you paid $40 CAC × 100 customers = $4,000 to acquire this cohort
And they generated $13,550 over 6 months
→ You made $9,550 contribution from that $4,000 investment
→ This justifies continuing to pay $40 CAC
If the cohort only generated $5,000 over 6 months?
→ You barely broke even
→ Your CAC is too high, or your product doesn't retain
════════════════════════════════════════════════════════════════════════════COHORT ANALYSIS: THE QUESTION THAT MATTERS
════════════════════════════════════════════════════════════════════════════
COHORT: Customers acquired in September 2025
Month 1 (Sep): 100 customers, $8,500 revenue
Month 2 (Oct): 22 customers rebought, $2,100 revenue
Month 3 (Nov): 18 customers rebought, $1,750 revenue
Month 6 (Feb): 12 customers rebought, $1,200 revenue
TOTAL VALUE FROM COHORT: $13,550 from 100 customers = $135.50 LTV
─────────────────────────────────────────────────────────────────────────
THE INSIGHT:
If you paid $40 CAC × 100 customers = $4,000 to acquire this cohort
And they generated $13,550 over 6 months
→ You made $9,550 contribution from that $4,000 investment
→ This justifies continuing to pay $40 CAC
If the cohort only generated $5,000 over 6 months?
→ You barely broke even
→ Your CAC is too high, or your product doesn't retain
════════════════════════════════════════════════════════════════════════════COHORT ANALYSIS: THE QUESTION THAT MATTERS
════════════════════════════════════════════════════════════════════════════
COHORT: Customers acquired in September 2025
Month 1 (Sep): 100 customers, $8,500 revenue
Month 2 (Oct): 22 customers rebought, $2,100 revenue
Month 3 (Nov): 18 customers rebought, $1,750 revenue
Month 6 (Feb): 12 customers rebought, $1,200 revenue
TOTAL VALUE FROM COHORT: $13,550 from 100 customers = $135.50 LTV
─────────────────────────────────────────────────────────────────────────
THE INSIGHT:
If you paid $40 CAC × 100 customers = $4,000 to acquire this cohort
And they generated $13,550 over 6 months
→ You made $9,550 contribution from that $4,000 investment
→ This justifies continuing to pay $40 CAC
If the cohort only generated $5,000 over 6 months?
→ You barely broke even
→ Your CAC is too high, or your product doesn't retain
════════════════════════════════════════════════════════════════════════════Pull cohort data from Shopify or Klaviyo monthly. It's the only way to know if your acquisition spending is actually building long-term value.
What NOT to Track (At Your Stage)
Just as important as what you track is what you ignore. These metrics cause more confusion than clarity for stores under $10M:
SIGNAL VS. NOISE: WHAT TO TRACK
════════════════════════════════════════════════════════════════════════════
SIGNAL (Track These) NOISE (Ignore These)
──────────────────── ────────────────────
✓ Revenue (Shopify) ✗ Social followers
✓ MER / Blended ROAS ✗ Page views
✓ CAC ✗ "Reach"
✓ AOV ✗ Platform-specific ROAS
✓ Repeat Rate / LTV ✗ Session duration
✓ Contribution Margin ✗ Bounce rate
✓ Cohort performance ✗ Daily CAC fluctuations
✗ Multi-touch attribution models
─────────────────────────────────────────────────────────────────────────
THE DIFFERENCE:
SIGNAL → Predicts or measures profitability
NOISE → Feels good but doesn't correlate with revenue
════════════════════════════════════════════════════════════════════════════SIGNAL VS. NOISE: WHAT TO TRACK
════════════════════════════════════════════════════════════════════════════
SIGNAL (Track These) NOISE (Ignore These)
──────────────────── ────────────────────
✓ Revenue (Shopify) ✗ Social followers
✓ MER / Blended ROAS ✗ Page views
✓ CAC ✗ "Reach"
✓ AOV ✗ Platform-specific ROAS
✓ Repeat Rate / LTV ✗ Session duration
✓ Contribution Margin ✗ Bounce rate
✓ Cohort performance ✗ Daily CAC fluctuations
✗ Multi-touch attribution models
─────────────────────────────────────────────────────────────────────────
THE DIFFERENCE:
SIGNAL → Predicts or measures profitability
NOISE → Feels good but doesn't correlate with revenue
════════════════════════════════════════════════════════════════════════════SIGNAL VS. NOISE: WHAT TO TRACK
════════════════════════════════════════════════════════════════════════════
SIGNAL (Track These) NOISE (Ignore These)
──────────────────── ────────────────────
✓ Revenue (Shopify) ✗ Social followers
✓ MER / Blended ROAS ✗ Page views
✓ CAC ✗ "Reach"
✓ AOV ✗ Platform-specific ROAS
✓ Repeat Rate / LTV ✗ Session duration
✓ Contribution Margin ✗ Bounce rate
✓ Cohort performance ✗ Daily CAC fluctuations
✗ Multi-touch attribution models
─────────────────────────────────────────────────────────────────────────
THE DIFFERENCE:
SIGNAL → Predicts or measures profitability
NOISE → Feels good but doesn't correlate with revenue
════════════════════════════════════════════════════════════════════════════Vanity metrics: Social followers, page views, "reach." These feel good but don't correlate with revenue.
Platform-specific ROAS: Meta ROAS, Google ROAS in isolation. The numbers are incomplete (40-60% missing) and each platform over-credits itself. Use MER instead.
Multi-touch attribution models: First-touch, last-touch, linear, time-decay — at your scale, the data isn't clean enough for these to be accurate. They create false precision.
Session duration and bounce rate: Occasionally useful for diagnosing site problems, but not metrics to track regularly. They don't predict sales.
Daily CAC fluctuations: CAC is noisy day-to-day. Track it weekly or monthly.
The Weekly Analytics Routine (30 Minutes)
Analytics only matters if you act on it. Here's a simple routine:
Monday Morning Check (15 minutes)
Pull these numbers from Shopify and your ad platforms:
WEEKLY ANALYTICS CHECK
════════════════════════════════════════════════════════════════════════════
METRIC THIS WEEK LAST WEEK CHANGE
────── ───────── ───────── ──────
Revenue (Shopify) $________ $________ ____%
Orders (Shopify) $________ $________ ____%
AOV (calculated) $________ $________ ____%
Ad Spend (all) $________ $________ ____%
MER (calculated) ____x ____x ____%
New Customers $________ $________ ____%
─────────────────────────────────────────────────────────────────────────
DECISION PROMPTS:
□ MER down >15%? → Check which channel dropped, creative fatigue?
□ AOV down >10%? → Check product mix, discount overuse?
□ Revenue up, orders flat? → Price increase working or fewer customers?
□ Revenue down, MER up? → Likely cut spend too much, test scaling
════════════════════════════════════════════════════════════════════════════WEEKLY ANALYTICS CHECK
════════════════════════════════════════════════════════════════════════════
METRIC THIS WEEK LAST WEEK CHANGE
────── ───────── ───────── ──────
Revenue (Shopify) $________ $________ ____%
Orders (Shopify) $________ $________ ____%
AOV (calculated) $________ $________ ____%
Ad Spend (all) $________ $________ ____%
MER (calculated) ____x ____x ____%
New Customers $________ $________ ____%
─────────────────────────────────────────────────────────────────────────
DECISION PROMPTS:
□ MER down >15%? → Check which channel dropped, creative fatigue?
□ AOV down >10%? → Check product mix, discount overuse?
□ Revenue up, orders flat? → Price increase working or fewer customers?
□ Revenue down, MER up? → Likely cut spend too much, test scaling
════════════════════════════════════════════════════════════════════════════WEEKLY ANALYTICS CHECK
════════════════════════════════════════════════════════════════════════════
METRIC THIS WEEK LAST WEEK CHANGE
────── ───────── ───────── ──────
Revenue (Shopify) $________ $________ ____%
Orders (Shopify) $________ $________ ____%
AOV (calculated) $________ $________ ____%
Ad Spend (all) $________ $________ ____%
MER (calculated) ____x ____x ____%
New Customers $________ $________ ____%
─────────────────────────────────────────────────────────────────────────
DECISION PROMPTS:
□ MER down >15%? → Check which channel dropped, creative fatigue?
□ AOV down >10%? → Check product mix, discount overuse?
□ Revenue up, orders flat? → Price increase working or fewer customers?
□ Revenue down, MER up? → Likely cut spend too much, test scaling
════════════════════════════════════════════════════════════════════════════Monthly Deep Dive (30 minutes)
Once a month, go deeper:
Contribution margin: Revenue minus COGS, shipping, and ad spend. Is it growing or shrinking?
CAC by channel: Which channel is getting you cheapest customers? (Use platform data directionally)
LTV:CAC ratio: Are you acquiring customers profitably over time?
Customer cohorts: Are customers acquired 3-6 months ago still buying? What's their cumulative value?
Product performance: Which SKUs are driving revenue vs. just selling?
How to Make Decisions with Imperfect Data
Since no single data source is perfectly accurate, use triangulation: make decisions when multiple signals agree.
Scaling a channel:
Platform ROAS looks good (3x+)
MER improved when you increased spend
Revenue increased week-over-week
All three agree? Scale with confidence.
Cutting a channel:
Platform ROAS looks bad (<1.5x)
MER dropped when you increased spend
Revenue flat or down despite more spend
All three agree? Cut it.
Mixed signals?
Don't make big moves
Run for 2-3 more weeks to get more data
Test smaller budget changes first
DECISION FRAMEWORK
════════════════════════════════════════════════════════════════════════════
PLATFORM MER REVENUE
DECISION ROAS TREND TREND ACTION
──────── ──── ───── ─────── ──────
Scale Good Up Up ✓ Scale 20%
Cut Bad Down Flat/Down ✓ Cut 30-50%
Hold Mixed Mixed Mixed → Wait 2 weeks
Investigate Good Down Down → Check tracking
Investigate Bad Up Up → Attribution lag?
════════════════════════════════════════════════════════════════════════════DECISION FRAMEWORK
════════════════════════════════════════════════════════════════════════════
PLATFORM MER REVENUE
DECISION ROAS TREND TREND ACTION
──────── ──── ───── ─────── ──────
Scale Good Up Up ✓ Scale 20%
Cut Bad Down Flat/Down ✓ Cut 30-50%
Hold Mixed Mixed Mixed → Wait 2 weeks
Investigate Good Down Down → Check tracking
Investigate Bad Up Up → Attribution lag?
════════════════════════════════════════════════════════════════════════════DECISION FRAMEWORK
════════════════════════════════════════════════════════════════════════════
PLATFORM MER REVENUE
DECISION ROAS TREND TREND ACTION
──────── ──── ───── ─────── ──────
Scale Good Up Up ✓ Scale 20%
Cut Bad Down Flat/Down ✓ Cut 30-50%
Hold Mixed Mixed Mixed → Wait 2 weeks
Investigate Good Down Down → Check tracking
Investigate Bad Up Up → Attribution lag?
════════════════════════════════════════════════════════════════════════════The Bottom Line
You don't need a PhD in data science to run a successful store. You don't need 15 dashboards or an attribution platform that costs $500/month.
You need:
Ground truth data — What actually happened (Shopify)
Efficiency signal — Is marketing working overall? (MER)
Unit economics — Are customers profitable? (CAC, AOV, LTV)
Profitability check — What's left after costs? (Contribution Margin)
A routine — Check the numbers weekly, go deeper monthly
The brands winning in 2026 aren't the ones with the most data. They're the ones who understand their data's limitations and make decisions anyway.
Start with five metrics. Add LTV and Contribution Margin when you're ready. Check weekly, decide monthly. That's 80% of the analytics game.
