Quantitative Decision Making for Tech Careers and Side Projects: Moving Beyond Gut Feeling

A developer in Lagos is offered a six-month contract paying ₦2.5m and a permanent role at ₦1.2m per month. The contract is clearly worth more on paper — yet most people pick the salary because risk feels safer. A freelancer weighs building a SaaS product against taking on three more retainers; a junior developer debates React versus Python; a senior manager considers relocating to Abuja or staying remote. In every one of these decisions, the winning choice is not the one that feels right — it is the one that survives arithmetic.

This guide gives you a complete quantitative toolkit for tech career and side-project decisions: expected value, decision trees, opportunity cost modelling, portfolio thinking, and practical risk frameworks — each with worked examples you can adapt to the Nigerian and remote-freelance context.

Expected Value: The Core Number Behind Every Decision

Expected value (EV) is the sum of every possible outcome, each weighted by its probability. It answers a deceptively simple question: if I ran this decision 100 times, what would the average outcome be? The formula is:

EV = (Probability₁ × Value₁) + (Probability₂ × Value₂) + …

Where probabilities are written as decimals and all values are in the same currency — for career decisions, mentally convert everything to net income or saved time. Compare options by their EV, but keep the risk checks in mind (Section 5): EV optimises averages, and averages only matter if you can survive the bad outcomes.

Worked Example A: Freelance Pitch vs. Salary Benchmark

You are considering a freelance engagement quoted at ₦800,000. You estimate a 55% chance the client pays in full, a 25% chance it pays after heavy delays (which burns you in stress and follow-ups, call it ₦700,000 of effective value), and a 20% chance it defaults or partial-pays (₦150,000). The EV:

EV = (0.55 × 800,000) + (0.25 × 700,000) + (0.20 × 150,000) = 440,000 + 175,000 + 30,000 = ₦645,000

If your alternative is a retainer worth ₦500,000 with 95% reliability, the EV difference narrows to ₦645,000 versus ₦475,000 — and once you factor the risk (Section 5), the "safer" option may genuinely win. This is the discipline the formula forces: it makes the quiet downside of "obviously better" offers visible.

Decision Trees: Stacking Your Options into One Picture

A decision tree maps a decision to its branches, their costs, probabilities, and outcomes — then you "roll back" from the leaves to the root picking the highest EV at each decision point. Build one when a decision has sequential stages, like the classic choice of job versus freelance versus further learning.

How to Build a Decision Tree

1. Root node: write your decision (e.g., "Should I accept the contract, take the salaried role, or defer work to study?").
2. Action branches: draw one branch per option.
3. Chance nodes: for each option, list the realistic outcomes (e.g., "contract pays in full", "contract delays 3 months", "contract collapses").
4. Costs and payoffs: label every branch with its net value (income minus costs and the opportunity cost of your time).
5. Probabilities: assign your best-estimate probability to each chance outcome; sum them to 1 per chance node.
6. Roll back: for each node, compute EV = Σ(probability × payoff); choose the branch with the highest rolled-back EV.

Worked Example B: Salary, Contract, or Learning

A developer weighs three paths over the next 12 months:

• Contract (₦2.5m for 6 months): 60% chance the work is steady and completes (

  ₦2,500,000

  ); 40% chance scope and payment grind down to

  ₦1,200,000

  . EV = 0.6(2,500,000) + 0.4(1,200,000) = ₦1,980,000.
• Salary (₦1.2m/month): near-certain, factor 95% the role lasts the year and 5% early-disclosure risk, EV ≈ ₦13.7m over 12 months — but compare on equal 12-month terms, where the contract needs a second gig for the remaining 6 months, roughly ₦2.4m–₦3m expected if clients are available.
• Learning (6 months at low income to master AI product skills): 35% chance it unlocks a paid role at ₦2m/month in month 7–12 (

  ₦12m

  minus the ₦1.2m forgone learning income =

  ₦10.8m

  ), 40% chance it lands a modest

  ₦1.2m/month

  role (net

  ₦4.8m

  ), 25% chance you return to a standard role (net

  ₦2.4m

  ). EV = 0.35(10,800,000) + 0.40(4,800,000) + 0.25(2,400,000) = 3,780,000 + 1,920,000 + 600,000 = ₦6.3m.

The tree does not just hand you the learning option as the EV winner — it shows the variance. Learning has the widest spread, and that variance is exactly what a risk framework (Section 5) lets you evaluate before you commit. That interplay is why decision trees beat gut feeling: the gut sees a salary's ₦14m and stops, the tree sees a ₦10.8m payoff that costs six months of living lean.

Opportunity Cost: What You Give Up Is a Real Expense

Opportunity cost is the value of your best alternative forgone. Every hour spent on Option A is an hour Option B loses, and every Naira committed to one direction is Naira that cannot compound elsewhere. Most tech professionals underweight this because it is invisible — but it is the single largest hidden cost in career math.

• Time: if you can earn ₦5,000/hour freelancing, a 200-hour side project carries a bare ₦1,000,000 opportunity cost before a single unit is sold. The product must plausibly return more than that to beat the freelance route.
• Focus and energy: your best alternative is not just "money" but momentum. Splitting focus across three projects lowers output on all three — calculate the EV of focusing on your top two instead.
• Forgone learning: choosing a role purely for its salary can cost the skill growth that would double your rate next year. Negate that with a "depreciation" line: how much value does this option lose each quarter as the market shifts?

Before any yes, ask: What is my highest-value alternative, and is this choice beating it on EV, risk, and growth combined? If the "yes" only wins on cash and costs you the growth, run the numbers again.

Portfolio Thinking for Skills and Projects

Careers are not single bets; they are portfolios. Continental airlines cry, pilots land; a fintech startup fails, the founder's consulting practice pays the rent. The quantitative habit is to spread bets across horizons and risk profiles deliberately.

• Spread the bets, but aim them: hold cashflow bets (services, retainers, part-time roles that always pay) alongside growth bets (products, deep-skill mastery, personal branding that pays later). The cashflow layer funds the patience the growth layer needs.
• Use a horizon lens: label each bet by horizon — 0–6 months (cashflow), 6–24 months (skill and portfolio compounding), 24+ months (products, authority, equity). Rebalance quarterly so no single horizon is empty.
• Set a kill criterion before you start: decide in advance "I will pivot out of this side project if it has not reached 500 users / ₦100k monthly revenue within six months." Pre-committed thresholds turn emotional sunk-cost debates into arithmetic, and they protect your portfolio from loss-aversion.
• Correlation matters: two side projects that both depend on the same two clients are one bet wearing a disguise. Genuine diversification needs different revenue sources, different audiences, or different skills — not the same bet twice.

Risk Frameworks: Surviving the Downside Before Enjoying the Upside

EV tells you the average; risk tells you whether you can survive the worst branch. Adopt these four guardrails to make every EV-spill over risk-aware.

1. Downside limits: define the worst outcome you can absorb before the upside is tempting. Never stake rent money, a family emergency fund, or the mortgage on a single project; cap "pure upside" investments (products, speculative contracts) at a fixed fraction of your monthly safety budget.
2. Conservative assumptions: every probability and payoff estimate you make has optimism baked in. Mentally discount unproven revenue by 25–40%, extend timelines by 30–50%, and recompute each EV with the pessimistic figures. The option that wins at both the realistic and pessimistic bounds is the genuinely robust one.
3. Inflation of timelines: software projects run long — assume reality will arrive at 1.5× your target time and 1.3× your cost. Recompute the decision tree with inflated timelines; if a supposedly winning option only wins at perfect-on-time assumptions, it is a poor bet.
4. The "survive to reinvest" test: will you still have capital, credibility, and energy to make the next decision even if this one fully fails? If the answer is no, the upside is not worth the risk regardless of EV.

Worked Examples: Four Decisions Run Through the Framework

1. Contract vs. Salary

A bargaining freelancer gets a 6-month contract at ₦2.5m versus a permanent role at ₦1.1m/month. Calculate EV as in Example B, but add the risk layer: the contract's 40% grind-down probability is your downside; the salary's near-certainty is your floor. If the cashflow difference (₦1.3m extra over six months) is less than your emergency buffer, the salary's guaranteed floor wins even though the contract has the higher headline. Recompute with the pessimistic discount and choose the floor when they are close.

2. Side Project vs. Freelance Clients

You want to build a SaaS checklist tool; your alternative is a retainer worth ₦400k/month with 90% reliability. Conservative estimates: the side project has a 30% chance of hitting ₦300k/month within 12 months, a 30% chance of breaking even (₦50k/month), and a 40% chance of flop. EV ≈ 0.3(3,600,000) + 0.3(600,000) + 0.4(0) = ₦1,080,000 a year versus the retainer's ₦4,320,000. On pure numbers the retainer wins hugely — so the intelligent choice is to build the side project while keeping the retainer (portfolio thinking, phase the hours), not to quit the retainer for it.

3. Learn X vs. Learn Y

You have 100 focused hours to invest. Skill X (edge computing) currently averages ₦1.2m for a project and has 2 known local clients; skill Y (AI agents) has 10 known demand signals, funnel-chats already inbound, and projects averaging ₦900k. Even though X has the higher individual rate, Y has the liquidity — the tested probability of landing work is what the EV emphasises. Rank skills by (demand probability × market rate × your speed to competence), not by headline pay.

4. Relocate or Stay Remote

You are offered an Abuja in-person role at ₦2.5m/month versus staying remote in Lagos at ₦1.2m/month. Frame it in quantities: the relocation's real EV must subtract rent difference, daily commuting cost and time, childcare/family support costs, and the risk that the role's travel-heavy culture deflates productivity. If the net gain is below ₦400k per month after every documented cost and a 25% uncertainty discount on the higher salary, the remote option's zero-risk floor remains the rational winner.

Conclusion

Quantitative decision making rewires how you choose: expected value turns vague anxieties into numbers, decision trees expose the variance behind a headline, opportunity cost makes invisible trade-offs visible, portfolio thinking smooths your worst quarter, and risk frameworks make sure you survive the worst branch before enjoying the best. Applied to contracts, side projects, skills, and relocation, this toolkit consistently beats gut feeling — because the gut predicts nothing, but the math always leaves a trail.

Your Next Actions

1. Write down the three biggest career decisions you face right now and compute an expected value for each option using the formula in this guide.
2. Build one decision tree this week for your most pressing choice (job vs. freelance vs. learning) with probabilities and penalties attached, and roll it back.
3. Track your best alternative for 30 days — every hour spent on one thing is a documented cost to another; audit your week and correct your allocation.
4. Map your current bets to horizons (cashflow, growth, long-term) and set pre-committed kill criteria for the weakest bet.
5. Apply the pessimistic discount to your best option's numbers and recheck: if it still wins, it is genuine.

Let the numbers decide what your instinct alone never could. If you need help turning decisions into launch-ready digital assets or market-tested projects, explore our services, sharpen your skills on learnTech, or talk to us on contact. For more frameworks like this one, keep reading the blog.

<!-- IMAGE GENERATION PROMPTS FOR THIS ARTICLE: 1. Clean corporate editorial photograph: a Nigerian professional in smart business casual standing beside a whiteboard covered in a decision tree diagram with boxes, arrows, percentages and Naira figures, photographed from a slight low angle, bright fluorescent office lighting, colour palette of white, navy and orange accents, mood of calm analysis and clarity, composition focused on the whiteboard math and the confident subject. 2. Isometric 3D illustration: a three-dimensional decision tree with a choice node splitting into branches labelled with percentages, each branch leading to small coins and icons (salary purse, contract handshake, graduation cap), pastel blue background, crisp corporate illustration style, informative mood, palette of blues, greens and warm amber for emphasis. 3. Minimalist data-visualization editorial graphic: a large clean expected-value formula (EV = P x V) rendered over a subtle grid, flanked by small bar charts comparing ₦ figures of contract vs salary vs learning paths, red and green accents on white and light grey, sharp sans-serif typography, professional finance-infographic style, neutral mood. 4. Cinematic flat-lay photograph: a desk scene with a calculator, a decision-tree sketched on graph paper, a smartphone showing a currency converter with Naira denominations, a cup of tea and reading glasses, warm morning window light, muted blues, greys and cream, orderly and contemplative, tight overhead composition. -->