The Professor
Index Course index

Module 11 · Non-technical skills

Lesson 49 — Estimating and splitting problems

Breaking down big tasks and communicating trade-offs honestly.

Published
In this lesson
  1. Exercise 1 — The honest estimate
  2. Exercise 2 — The vertical split
  3. Exercise 3 — The explicit scope
  4. Exercise 4 — The visible plan
  5. Exercise 5 — The applied skill
  6. Submit

The method on TicketFlow's features. No solutions.md before submitting.

Exercise 1 — The honest estimate

  1. Take the "automatic seat assignment (the system picks the best free seats)" feature and estimate it with §1's method: the 3 parts (well known/medium/unknown), the range with confidence, and the re-estimation trigger.
  2. The multiplier: review the last task you estimated "2 h": how long did it REALLY take? Note the deviation and explain its cause (did scope grow? did the unknown weigh?).
  3. The history: create your docs/estimates/Q4.md with the course's last 5 tasks (estimated vs actual, the deviation's why). Which pattern do you see (infra tasks underestimated? data ones ×2)?

Exercise 2 — The vertical split

  1. Split exercise 1's feature into 4-6 VERTICAL pieces (each: model+service+endpoint+test, demonstrable). Happy path first, edges after (§3's gift card pattern).
  2. Piece 1 in detail: what does day 1 deliver (the admin sees suggested seats with the fake gateway?)? If your piece 1 is "just the model": rewrite it vertical.
  3. The limit: does any piece exceed 3 days? Split it with the question "what is this piece's happy path?" and verify each piece has ITS demonstrable test.

Exercise 3 — The explicit scope

  1. For the full feature: write §4's "yes, but" with the 3 pieces (conditions, scope in/out, options for the business). The out-of-scope must be EXPLICIT (automatic-assignment refunds: in or out? decide and communicate).
  2. The impossible-deadline scenario: the business asks for the feature in 3 days (your estimate: 7-10). Write the 3 structured options (smaller scope / displace something else / resources) with each one's real cost. Which do you recommend and why?
  3. Early communication: your re-estimation trigger fires on day 2 (the unknown component is taking longer). Write the EXACT message you send that day 2 (to the stakeholder, 5 lines): what do you say, what do you ask, what do you propose?

Exercise 4 — The visible plan

  1. Turn the pieces into the sprint plan: the piece/days/deliverable/demo table — with the demo day for each (the business sees progress, not smoke).
  2. The plan's risk: mark the 3 biggest risks with their mitigation (the provider's sandbox, the assignment's concurrency, the suggestion's UX) — the plan's risk/impact/mitigation table.
  3. The post-sprint: when you finish (or simulate the close), fill the history's actual column (§1): which piece deviated and why? The calibration loop closed.

Exercise 5 — The applied skill

  1. The BIG problem: "migrate checkout from the monolith to the separate payments service (53)". Estimate it with the method (the 3 levels) and split it into vertical phases (the strangle pattern? which phase delivers value?). Honest total range.
  2. The big problem's communication: write the stakeholder's "yes, but": which options do you present (phases? not doing it? buying instead of building?) with their cost/risk?
  3. The personal rule: write YOUR estimation rule in 3 lines (the one history taught you) and paste it into your onboarding (48: the doc future-you reads).

Submit

Paste the estimate with trigger, the vertical split with demos, the deadline's "yes, but" and the visible plan. Next: Lesson 50 — Code review.

^ top