Operating Expense Management and Control: Managing Spending
Master opex management. Control spending, optimize allocation, maintain discipline.
Key Takeaways
- Opex definition: Operating expenses (all non-COGS costs). Breakdown: Payroll (60-70% typical), tools (5-10%), marketing (10-20%), operations (5-10%). Control: Budget by category, require approval for spending. Higher amounts need higher approval. Example: <£1K anyone approves, £1-5K manager, >£5K CFO/CEO. Discipline: Prevents overspending, forces prioritization.
- Payroll control: Largest expense, hardest to cut. Approach: Hiring freeze before layoffs (less painful). Measure: Revenue per employee (should increase as scale). Benchmark: £100K-£300K revenue per employee typical. Ratio: If £100K revenue, 1-3 employees (depends on business). Monitor: Headcount plan (planned hires vs actual), salary increases, benefits cost.
- Tool expense management: Many low-cost SaaS tools add up (£1K/month easy). Audit: Quarterly review of all subscriptions (cancel unused). Negotiate: As company grows, negotiate volume discounts on key tools. Target: Tool cost 3-5% of revenue (5-10% is overspending). Combined payroll + tools = 65-85% of opex.
Managing and Optimizing Operating Expenses
Building expense discipline and operational efficiency. **Operating expenses fundamentals** Definition: - All non-COGS business expenses - Includes: Payroll, marketing, tools, rent, legal, etc. - Key metric: Opex as % of revenue (lower is better) Typical opex breakdown (£100K/month): | Category | Amount | % of opex | |---|---|---| | Payroll | £60K | 60% | | Marketing | £15K | 15% | | Tools/SaaS | £8K | 8% | | Rent/facilities | £4K | 4% | | Professional services | £2K | 2% | | Travel | £2K | 2% | | Other | £9K | 9% | | **Total** | **£100K** | **100%** | Opex as % of revenue: - Early stage (£100K revenue): 100-150% opex (losing money, investing) - Growth stage (£1M revenue): 60-80% opex (approaching breakeven) - Scale stage (£10M+ revenue): 30-40% opex (profitable) Target: <60% opex by breakeven (£40K opex on £100K revenue = profitable) **Payroll management** Payroll breakdown: - Salaries/wages: Base compensation - Benefits: Health insurance, retirement, PTO (20-25% of salary) - Taxes: Employer taxes (15-20% of salary) - Recruitment: Cost to hire (5-10% of salary annually) Total cost per employee: - Base salary: £50K - Benefits (25%): £12.5K - Taxes (15%): £7.5K - Total compensation cost: £70K Hiring plan impact: - Current: 10 employees @ £70K average = £700K - Planned: Add 3 engineers @ £80K = £240K additional - New total: £940K (+34% payroll increase) Headcount control: Philosophy: "Hire slowly, fire fast" - Hiring: Thoughtful process (do we need this role? Can we wait?) - Firing: Quick if not working (don't waste compensation on wrong fit) Approval process: - New role: CEO/board approval (significant expense) - Salary: Based on market (£50-100K range for engineers) - Benefits: Standard across company (no cherry-picking) - Raises: Annual review (3-5% typical), promotions (10-15%) Cost control mechanisms: - Hiring freeze: Stop new hires when runway tight - Salary bands: Prevent overpaying for roles - Benefits cap: Offer good benefits, but cap cost (e.g., 80% health insurance) - Bonus cap: Align with company performance Revenue per employee: - Metric: Total revenue / headcount - Target: £100K-300K per employee (depends on business) - Example: £1M revenue, 5 people = £200K per employee (healthy) - Growth: As revenue grows faster than headcount, RPE improves Monitoring: - Monthly: Headcount report (actual vs plan) - Quarterly: RPE analysis (is productivity improving?) - Annual: Salary benchmarking (are we competitive?) **Tool and software expense management** Audit example: 50-person company typical tools | Tool | Monthly | Annual | Use | |---|---|---|---| | Slack | £600 | £7.2K | Communication | | Google Workspace | £200 | £2.4K | Email, docs, sheets | | Salesforce | £2000 | £24K | CRM | | Asana | £500 | £6K | Project management | | Figma | £400 | £4.8K | Design | | GitHub | £300 | £3.6K | Code repository | | AWS | £3000 | £36K | Infrastructure | | Stripe | £0 (2.2% processing) | Varies | Payments | | Calendly | £100 | £1.2K | Scheduling | | Zoom | £200 | £2.4K | Video meetings | | Quadrant (analytics) | £150 | £1.8K | Analytics | | Amplitude | £200 | £2.4K | Product analytics | | Loom | £100 | £1.2K | Video recording | | Notion | £100 | £1.2K | Internal wiki | | Intercom | £250 | £3K | Customer chat | | Other | £500 | £6K | Various | | **Total** | **£8.5K** | **£102K** | | Analysis: - AWS: £36K (largest, necessary for infrastructure) - Salesforce: £24K (necessary for sales) - Everything else: £42K (smaller tools adding up) Opportunity: Review "everything else" - Cancel unused: Quadrant (analytics), Loom (not used) - Renegotiate: Intercom (too expensive for use case, switch to Zendesk) - Consolidate: Multiple analytics tools (keep one) - Expected savings: £10-15K annually (10-15% of non-core tools) Approval process for new tools: - Under £100/month: Approver (person needing tool) - £100-500/month: Manager approval - £500-1000/month: Finance approval - >£1000/month: CEO approval Requirement: Business case (what problem does this solve? What's ROI?) Example: - Tool: Chorus (sales call recording) - Cost: £500/month - Business case: Improve sales coaching, reduce ramp time - Expected: 2 week faster ramp per rep × £0.5K per week = £1K savings per hire - ROI: £1K saving × 5 new hires per year = £5K benefit, £6K cost = -£1K (not approved) Alternative: Use free Loom instead of Chorus **Marketing spend control** Budget allocation: - Paid ads (Facebook, Google, LinkedIn): £10K/month - Content marketing (blog, webinars, guides): £3K/month (freelancers, tools) - Events (sponsorship, booth): £2K/month - Email campaigns (automation, list management): £1K/month - Sales enablement (collateral, training): £1K/month - Public relations/analyst relations: £1K/month - Total: £18K/month (18% of £100K opex) Optimization: ROI tracking: - Paid ads: £10K spend, 20 customers acquired, £500 CAC, 12-month payback = acceptable - Content: £3K spend, 5 customers/month acquired organically (hard to track), assume low CAC = excellent - Events: £2K spend, 3 customers from events, £667 CAC, 20-month payback = expensive Actions: - Maintain paid ads (working) - Invest in content (best ROI, scale) - Reduce events (slow payback, reconsider) - Kill low-ROI channel if any Expected allocation: - Paid ads: £8K (more efficient, scale) - Content: £5K (invest, best ROI) - Events: £1K (minimal) - Other: £4K - Total: £18K (same budget, different allocation = better ROI) **Facility and administrative costs** Rent/facilities: - Office space: £4K/month - Utilities: £500/month - Internet: £200/month - Insurance: £300/month - Total: £5K/month With remote work (shift to hybrid): - Office downsized 50% (London expensive) - New rent: £2K/month - Utilities: £200/month - Internet: £100/month - Insurance: £150/month - Total: £2.5K/month (50% savings) Remote work benefit: 30-person team, 30% want office space occasionally = book shared space (£500/month) instead of permanent lease Savings: £2.5K from opex (2.5% of opex, material at scale) **Expense approval process** Policy example: | Amount | Approval | Speed | Use case | |---|---|---|---| | <£500 | Manager | Next business day | Supplies, small tools | | £500-2K | Finance | 2-3 days | Larger purchases | | £2K-10K | CFO | 3-5 days | Significant spend | | >£10K | CEO/Board | 1-2 weeks | Major commitments | Controls: - Budget: Department must have budget approved (no surprise spending) - Receipts: Keep all receipts (audit trail, tax deduction) - PO: Use purchase orders for >£1K (binding commitment) - Vendor vetting: Check vendor reliability (don't work with unknown suppliers) **Opex monitoring dashboard** Monthly metrics: | Category | Budgeted | Actual | Variance | Status | |---|---|---|---|---| | Payroll | £60K | £61K | +£1K | +1.7% | | Marketing | £15K | £16K | +£1K | +6.7% | | Tools | £8K | £8.2K | +£0.2K | +2.5% | | Rent/facilities | £5K | £5K | £0 | On | | Professional | £2K | £1.5K | -£0.5K | -25% | | Travel | £2K | £1K | -£1K | -50% | | Other | £8K | £7.5K | -£0.5K | -6.3% | | **Total** | **£100K** | **£100.2K** | **+£0.2K** | **+0.2%** | Analysis: - Payroll slightly over (1 person not budgeted?) - Marketing over (need to investigate, reduce elsewhere) - Professional services under (good, less need for contractors) - Travel under (team not traveling) - Overall: On budget (total £100.2K vs £100K) Quarterly review: - Trend: Is opex creeping up? (should decrease as % of revenue) - Efficiency: RPE (revenue per employee) improving? - Allocation: Is money going to right places (growth channels)? Target: Opex from 100% revenue → 60% by breakeven (2-3 years) **Common opex mistakes** Mistake 1: No budget discipline - Problem: Spend without planning, budget useless - Fix: Monthly actuals vs budget, variance review - Impact: Catch overspending early, adjust course Mistake 2: Ignore small expenses - Problem: £100/month tools × 50 = £5K/month waste - Fix: Quarterly tool audit (cancel unused) - Impact: Easy 5-10% savings Mistake 3: Hiring without planning - Problem: Hire for growth, then burn rate spikes - Fix: Headcount plan (when to hire, for what) - Impact: Controlled payroll growth Mistake 4: No approval limits - Problem: Anyone can spend anything (chaos) - Fix: Tiered approval (amount determines who approves) - Impact: Accountability, discipline