Gross Margin Optimization and Cost of Revenue: Improving Unit Economics
Master gross margin. Analyze COGS, reduce costs, improve profitability.
Key Takeaways
- Gross margin definition: (Revenue - COGS) / Revenue. COGS = direct costs to deliver product (hosting, support, payment processing). Example: £100K revenue, £30K COGS = £70K gross profit (70% margin). Benchmark: SaaS 70-85% (good), <50% problematic. Impact: High margin = more money for operating (growth, salaries). Low margin = need high volume to be profitable.
- COGS components: (1) Cloud hosting (5-15% typical), (2) Payment processing (2-3%), (3) Support costs (3-5%), (4) COGS of goods (if product), (5) Other direct costs. Optimize: Each component can be improved 10-30% (£100K COGS → £70-80K COGS possible). Tools: Monitor per-customer COGS (should decrease as scale, not increase).
- Operating leverage: As scale, COGS per customer decreases. Example: £10K hosting serves 100 customers (£100/customer), then 500 customers (£20/customer). Gross margin: 60% → 80% (huge improvement). Key: Don't waste leverage (avoid expensive features, optimize infrastructure). Target: 75%+ gross margin for sustainable SaaS business.
Analyzing and Optimizing Gross Margin and Cost of Revenue
Building sustainable unit economics through cost optimization. **Gross margin fundamentals** Definition: - Gross margin = (Revenue - COGS) / Revenue - COGS = Cost of Goods Sold (direct costs to deliver product) - Shows: How much of each revenue dollar is available for operations Formula example: Revenue: £100K COGS: £30K (direct costs) Gross profit: £70K (£100K - £30K) Gross margin: 70% (£70K / £100K) Interpretation: - 70% margin: 70 cents per pound available for operations - 30% to COGS: 30 cents per pound spent on delivery Benchmark by business model: | Model | Margin | Notes | |---|---|---| | Pure SaaS | 75-90% | Mostly hosting + payment | | SaaS + consulting | 60-75% | Some service delivery | | Marketplace | 20-40% | High payment to partners | | Product + support | 50-70% | Mix of product and labor | | B2B services | 30-50% | Labor-intensive | Target: 70%+ for sustainable SaaS **COGS components and analysis** Component 1: Cloud hosting (infrastructure) Typical cost: 5-15% of revenue Calculation example: - AWS bill: £10K/month - Revenue: £100K/month - Hosting as % of revenue: 10% Per-customer cost (scaling indicator): - 100 customers: £10K / 100 = £100 per customer/month - 500 customers: Same £10K (optimized) = £20 per customer (4x leverage!) - 1000 customers: £12K (minimal overhead increase) = £12 per customer (still very efficient) Key insight: Cloud hosting should decrease per-customer as scale (if architected right) Optimization tactics: Tactic 1: Right-sizing - Current: Overprovisioned (more capacity than needed) - Fix: Reduce instance sizes, use cheaper regions, rightsizing tools - Expected: Save 20-30% of hosting costs Tactic 2: Auto-scaling - Current: Static capacity, pay for peak even in off-hours - Fix: Auto-scale infrastructure (scale down at night, weekends) - Expected: Save 15-25% (depends on usage pattern) Tactic 3: Caching and optimization - Current: Every request hits database - Fix: Cache frequently accessed data, optimize queries - Expected: Reduce CPU/bandwidth, save 10-20% Tactic 4: Negotiate better rates - Current: Standard cloud pricing - Fix: Volume discounts (Amazon credits, reserved instances) - Expected: Save 10-15% at scale (£5M+ spend) Combined potential: Reduce hosting from 10% → 6-7% of revenue (30-35% reduction) Component 2: Payment processing Typical cost: 2-3% of revenue Calculation: - Stripe charges: 2.2% + £0.30 per transaction - £100K revenue, 500 transactions - Cost: (£100K × 2.2%) + (500 × £0.30) = £2,200 + £150 = £2,350 (2.35%) Optimization: - Negotiate lower rate (Stripe volume discounts) - Reduce failed payments (lower retries = fewer charges) - Optimize transaction size (batch vs individual) - Expected: Reduce from 2.35% → 2.0% (save £350/month on £100K revenue) Component 3: Support and hosting costs Typical cost: 3-5% of revenue Calculation: - Support team: 2 people × £30K = £60K/year - Customers: 500 (200 who use support heavily) - Cost per customer: £60K / 500 = £120/customer/year - As % of revenue: (£60K / £100K × 12 months) = 60% = 5% (if £100K/month) Optimization tactics: Tactic 1: Self-service - Current: Support team handles common questions - Fix: Build self-serve knowledge base, chatbot - Expected: Reduce support tickets 30%, lower cost to 3.5% Tactic 2: Improve product (reduce support need) - Current: Confusing UI, bugs, missing docs - Fix: Improve UX, fix bugs, add help text - Expected: Fewer support requests, reduce to 3% Tactic 3: Community support - Current: All support from company - Fix: Build community (Slack, forums, peer support) - Expected: Customers help each other, reduce cost to 2% Component 4: COGS of goods (if applicable) For product-based businesses: Typical cost: 30-50% of revenue Example: - Product cost: £30 per unit - Price: £100 per unit - COGS: 30% Optimization: - Negotiate better supplier rates - Improve manufacturing efficiency - Economies of scale (buy in bulk) - Expected: Reduce from 30% → 25% (17% improvement) **Gross margin trends and targets** Early stage (pre-revenue to £1M ARR): - Gross margin: 60-75% typical - Focus: Build product, not yet optimized - Target: 65%+ Growth stage (£1M-10M ARR): - Gross margin: 70-80% - Focus: Optimize infrastructure, improve efficiency - Target: 75%+ Scale stage (£10M-100M ARR): - Gross margin: 75-85% - Focus: Leverage (infrastructure costs fixed, spreading) - Target: 80%+ Mature stage (£100M+): - Gross margin: 80-90% - Focus: Pure leverage (scale) - Target: 85%+ **Gross margin by customer segment** Enterprise customers: - Margin: 80-90% (economies of scale) - Reason: Hosting cost per customer low, payment processing low (large transactions), support efficient Mid-market customers: - Margin: 75-80% (good leverage) - Reason: Some hosting cost, moderate support SMB/self-serve customers: - Margin: 60-70% (less leverage, more support) - Reason: Support cost per customer higher (relative), more payment processing cost Strategy: Enterprise most profitable, focus there **Operating leverage in action** Scenario: SaaS company scaling Month 1: £10K revenue - Hosting: £2K (20% COGS) - Payment: £250 (2.5% COGS) - Support: £1.5K (15% COGS) - Total COGS: £3.75K (37.5% COGS, 62.5% margin) Month 6: £30K revenue - Hosting: £3K (10%, optimized, same infrastructure) - Payment: £750 (2.5%, scales with revenue) - Support: £2K (6.7%, better processes) - Total COGS: £5.75K (19.2% COGS, 80.8% margin) Month 12: £100K revenue - Hosting: £7K (7%, more optimized) - Payment: £2.5K (2.5%, same rate) - Support: £4K (4%, self-serve reduces ticket volume) - Total COGS: £13.5K (13.5% COGS, 86.5% margin) Trend: Margin improves from 62.5% → 80.8% → 86.5% (operating leverage in action) **Margin dashboard and monitoring** Monthly metrics: | Item | Amount | % of revenue | Target | Status | |---|---|---|---|---| | Revenue | £100K | 100% | £120K | -17% | | Hosting | £7K | 7% | 6% | Over | | Payment processing | £2.5K | 2.5% | 2% | Over | | Support | £4K | 4% | 3% | Over | | Other COGS | £1K | 1% | 1% | On | | Total COGS | £14.5K | 14.5% | 12% | Over | | Gross profit | £85.5K | 85.5% | 88% | Under | Analysis: - Revenue below target (-17%, investigate) - Gross margin below target (14.5% vs 12% target) - Hosting high (7% vs 6%), needs optimization - Payment processing high (2.5% vs 2%), negotiate rates Actions: - Fix hosting (optimize, reduce from 7% to 6%) - Negotiate payment (get down from 2.5% to 2%) - Support efficient (already 4%, on track to 3% with scale) Expected improvement: - Reduce COGS from 14.5% to 12% (2.5% improvement) - Gross margin: 85.5% → 88% (target achieved) **Common margin mistakes** Mistake 1: Ignore COGS as scale - Problem: Assume COGS stays same (doesn't increase as revenue grows) - Reality: Some COGS grows with revenue (payment processing, support), some is fixed (hosting) - Fix: Track COGS as % of revenue, optimize components separately - Impact: Margin degradation caught early Mistake 2: No per-customer tracking - Problem: Know average margin (80%), don't know if expensive customers have low margin - Fix: Calculate margin per customer segment - Impact: Identify unprofitable segments, adjust pricing or support Mistake 3: Neglect support cost - Problem: Build product, skip on support efficiency - Fix: Invest in self-serve, community, automation - Expected: Support cost reduce from 5% to 2% (3% margin improvement, material) Mistake 4: Expensive features - Problem: Build features that increase hosting cost (no ROI analysis) - Fix: Analyze feature cost/benefit (does new feature justify hosting increase?) - Impact: Margin discipline