Estimation Language

Option C is the professional estimation with uncertainty. It models the three-tier estimation approach:

• Tier 1: Current rough estimate — "3–5 weeks" — gives the PM something to plan with immediately
• Tier 2: Named unknowns — "third-party API integration", "auth system" — explains why the range is wide (not incompetence, but unresolved questions)
• Tier 3: De-risking offer — "2-day spike to nail down unknowns" — turns uncertainty into a time-boxed investigation
• Commitment to tighten — "tighter estimate by Friday" — the PM gets a date for a better answer

Estimation uncertainty phrases:
• "My estimate is X, assuming [condition]. If [alternative], add Y days."
• "This is a rough order of magnitude — I'd need to investigate [X] before committing"
• "I can give you a confident estimate after a spike. The spike itself will take 2 days."
• "Best case X, worst case Y, most likely Z"

Option D ("exactly 4 weeks") is the most dangerous — false precision destroys trust when wrong.

Option C is the professional deadline management response. It reframes "can you do it?" into scope × time trade-off discussion:

1. Sets up the framing — "I want to set realistic expectations so the business can plan accurately" — positions this as being helpful, not obstructive
2. States the honest baseline — "6 weeks, not 4" — with no hedging
3. Offers three concrete options:
— Cut scope to hit original deadline
— Full scope, extended deadline
— Phased delivery (best of both)
4. Asks the PM/stakeholder to choose — respects their ability to make the call based on business context you don't have

Option A ("we'll do our best") sets up failure — you ship late or poor quality, without forewarning. Option D ("somehow") is even worse. Option B is honest but unhelpful — it says no without giving options.

The key insight: Most deadlines are not fixed constraints — they're the PM's first preference. When you give three options, you allow the PM to make an informed decision. This is collaborative, not adversarial.

Option C is the professional explanation of why estimation is inherently uncertain. Key points:

• Names the fundamental cause — "discovering unknowns not visible during scoping" — this is accurate and not an excuse
• Positions it correctly — "property of creative/knowledge-based work", not a professional failure
• Avoids defensiveness — acknowledges the concern without being combative
• Proposes concrete improvements — ranges, spikes, calibration tracking
• Ends collaboratively — "would it help to discuss how we communicate confidence levels?" — moves to solution

The "confidence level" framing is particularly effective: instead of asking "will it take 2 weeks?", planning with "we have 80% confidence we can deliver in 3 weeks, and we'll know in 2 days whether the auth integration changes that" gives everyone useful information.

Analogies that help non-engineers understand this:
• "It's like estimating a home renovation — you find surprises once the walls come down"
• "It's a research and development activity, not manufacturing"