Gross Margin Deep Dive: Optimizing Cost Structure and Profitability

Understanding Cost Structure

Gross margin is the profit remaining after paying the direct costs of delivering your service. **Gross Margin Formula and Benchmark** Gross Margin = (Revenue − COGS) ÷ Revenue Example: Revenue: £10M COGS: £2M (cost to deliver the product) Gross Profit: £8M Gross Margin: 80% Interpretation: For every £1 of revenue, you keep 80p after paying delivery costs. SaaS benchmarks: Early-stage (< £1M ARR): - Gross margin: 60-75% (not yet optimized) - Problem: High support costs per customer (small base) - Plan: Improve margin as scale Growth-stage (£1-10M ARR): - Gross margin: 75-85% (improving) - Problem: Infrastructure costs increasing with scale - Plan: Negotiate better terms, improve support efficiency Scale-stage (£10-100M ARR): - Gross margin: 80-90% (optimized) - Benefit: Scale provides leverage on costs - Strategy: Maintain or improve margin Mature (> £100M ARR): - Gross margin: 85%+ (highly optimized) - Leverage: Large enough to negotiate best rates If below 70%: Your product is too expensive to deliver profitably. **Components of COGS (Cost of Goods Sold)** What counts as COGS for SaaS: 1. Infrastructure Costs (40-50% of COGS) - Cloud hosting (AWS, Azure, Google Cloud) - Database services (scale with data) - CDN (content delivery) - Backup and disaster recovery Example: £10M revenue SaaS - AWS costs: £400K/year (4% of revenue) - This is typical for typical SaaS - Some data-intensive SaaS: 8-10% of revenue 2. Payment Processing (10-15% of COGS) - Stripe: 2.2% + £0.30 per transaction - Local payment processors: 2-5% depending on country - Chargeback fees: 0.1-0.3% Example: £10M revenue - Stripe fees: 2.2% × £10M = £220K - = 2.2% of revenue 3. Support Salaries (allocated, 20-30% of COGS) - Customer support team cost allocated to COGS - How much to allocate? Support cost ÷ Revenue Example: 10-person support team - Cost: £400K/year (salaries, benefits) - Revenue: £10M - Allocation: £400K ÷ £10M = 4% of revenue 4. Third-party SaaS (5-10% of COGS) - Twilio (SMS/calls): If using for product features - Stripe (for payment): Already counted above - SendGrid (email): If core to product - Data providers: If core to product Example: £10M revenue SaaS - Twilio for SMS feature: £100K/year (1% of revenue) Total COGS example: - AWS: 4% - Stripe: 2.2% - Support (allocated): 4% - Third-party SaaS: 1.5% - Total COGS: 11.7% of revenue - Gross Margin: 88.3% What NOT to include: - Sales salaries (included in S&M, not COGS) - R&D/engineering salaries (included in R&D, not COGS) - Admin salaries (included in G&A, not COGS) - Office rent (included in G&A, not COGS) - Marketing spend (included in S&M, not COGS) **Cost Structure Analysis** Analyze your COGS to find improvement opportunities: Example company, £5M revenue: COGS Breakdown: - AWS: £300K (6% of revenue) - Stripe: £120K (2.4% of revenue) - Support: £200K (4% of revenue) - Third-party: £50K (1% of revenue) - Total COGS: £670K (13.4% of revenue) - Gross Margin: 86.6% This is healthy. But let's look for improvement opportunities: AWS costs (£300K): - Analyze: Are databases optimized? Redundancy needed? - Benchmark: 6% is average, industry range 3-8% - Opportunity: Optimize databases, reduce redundancy → save £50K (1% of revenue) Stripe fees (£120K): - Analyze: Using Stripe for all payments? International customers? - Benchmark: 2.4% is high (standard 2.2%) - Opportunity: Negotiate better rates or use local processors for international → save £30K (0.6%) Support (£200K): - Analyze: 10 support reps × £20K/year = £200K. Is this right? - Benchmark: 4% is typical, range 2-6% - Opportunity: Build self-service knowledge base, reduce rep headcount by 2 → save £40K (0.8%) Third-party (£50K): - Analyze: Necessary or nice-to-have? - Opportunity: Build in-house for 1 service → save £20K (0.4%) Total potential savings: £140K (2.8% of revenue) New gross margin: 89.4% (from 86.6%) This is material. Margin improvements compound. **Operational Leverage** Key insight: COGS grows slower than revenue at scale. Example: Year 1: £5M revenue - COGS: £670K (13.4%) - Gross profit: £4.33M Year 2: £7.5M revenue (50% growth) - AWS: increased to £375K (but scales with usage, not revenue %, so might only be 5% now) - Stripe: £180K (same %, but on higher base) - Support: £240K (added 1 rep, not proportional) - Third-party: £70K - Total COGS: £865K (11.5% of revenue, improved!) - Gross profit: £6.64M Margin improved from 86.6% to 88.5% just by scaling. This is operational leverage. Revenue grows 50% (£5M → £7.5M), but COGS only grew 29% (£670K → £865K). The reason: - Some costs fixed (support reps are hired in increments) - Infrastructure gets more efficient (per-unit costs decrease) - Payment processing % doesn't change but might negotiate better terms This is why growing SaaS becomes more profitable. Not because you raise prices, but because costs don't scale linearly. **Margin Expansion Opportunities** Lever 1: Negotiate Infrastructure Costs AWS costs typically 4-6% of revenue. Leverage points: - Reserved instances (commit to 1-3 year usage, get 30-50% discount) - Spot instances (unused compute, 60-90% cheaper, but variable) - Move to cheaper regions (some regions cheaper than US) - Optimize database (unused indexes, bad queries cost a lot) - CDN compression (smaller files = cheaper CDN) Potential savings: 20-30% of infrastructure costs Example: - Current AWS: £300K/year - Optimize and negotiate: £220K/year - Savings: £80K (1.6% of revenue) Lever 2: Optimize Payment Processing Stripe charges 2.2% for US cards. But can do better: - Local payment processors (1.5-2.5%): Lower costs in some regions - ACH payments (£0.25-1%): Much cheaper than credit cards - Direct bank transfers: Essentially free - Volume discounts: Negotiate with Stripe if >£1M/month Example: - 70% customers on Stripe (2.2%): 70% × 2.2% = 1.54% - 20% on local processors (1.8%): 20% × 1.8% = 0.36% - 10% on ACH (1%): 10% × 1% = 0.10% - Blended rate: 2% - Savings: 0.4% of revenue (£20K on £5M) Lever 3: Improve Support Efficiency Support is often 3-5% of revenue. Improvement opportunities: - Self-service knowledge base (reduce tickets by 20-30%) - Chatbot for common questions (handle 30-50% of tickets) - Community forums (customers help each other, reduce support load) - Tiered support (community/email/chat, then paid for phone) - Hire in lower-cost regions (£10K/year support rep in Philippines vs £20K in UK) Example: - Current: 10 support reps × £20K = £200K (4% of revenue) - Build knowledge base: Reduce tickets 20% → need 9 reps = £180K (3.6%) - Add chatbot: Reduce tickets 15% more → need 8 reps = £160K (3.2%) - Hire 2 reps in Philippines (£10K each): 6 UK reps (£120K) + 2 Phil (£20K) = £140K (2.8%) - Savings: £60K (1.2% of revenue) Total is material: support is often your biggest COGS opportunity. Lever 4: Increase Prices If margins low, pricing might be too low. Example: 70% margin (vs 85% benchmark) - This suggests pricing 20% too low - Raise prices 10%: Gross margin improves to 75% - Raise prices 15%: Gross margin improves to 78% But risk: Volume might decrease if price-sensitive market. Test with: - New customers only (don't risk existing) - Negotiate with at-risk customers (give them discount to renew) - Tiered pricing (better tier has higher price) Example: - Current: £100/mo for 1000 customers = £1.2M annual revenue - Increase 10%: £110/mo, might lose 5% of customers = 950 customers × £110 × 12 = £1.254M (4% growth in revenue) - Gross margin improves from 70% to 73% + 4% growth = strong result **Margin Dashboard** Track monthly: | Metric | Target | Actual | Trend | |--------|--------|--------|--------| | Gross Margin % | 80% | 78% | ↓ | | AWS % of revenue | 4% | 4.5% | ↑ (concerning) | | Stripe % of revenue | 2% | 2.2% | = | | Support % of revenue | 3% | 4% | ↑ (hiring) | | Third-party % | 1% | 1.2% | = | Watch for trends: - If margin declining, drill into which cost category increasing - If AWS increasing, might need to optimize or negotiate - If support increasing faster than revenue, hiring too fast Margin should improve as you scale. If not, something's wrong (either pricing, costs, or mix changing).