Communicating slips

Say the assumptions with the number: "Two weeks if FraudSvc review is ≤2 days; otherwise three." When assumptions break, renegotiate early.

Managers can help with escalation. They cannot help with silence. See project execution for the weekly status rhythm.

Walking through a status sentence

Yellow. Export query works in staging. FraudSvc API rejects our service account — escalated Tuesday, no reply yet. If unresolved by Friday I need either a temporary allowlist or we slip the self-serve milestone by one week and ship internal-only on the original date.

That is a decision request, not a confession.

When you are late, you have three levers: cut scope, move the date, or add people (usually the weakest late in a project — Brooks's law). Present the options with costs. Do not silently work weekends and hope; that hides the signal until you break.

SDE2s are expected to estimate a multi-week feature with a breakdown, call out risks, and warn early. Bring a table, not a vibe. When leadership asks "can we hit Friday?", answer with options, not bravery.

In interviews, when asked "how long would this take?", narrate the breakdown and the confidence, not a macho single number. Pair estimation with MVP scoping: often the right answer to a bad deadline is a smaller product, not a heroic calendar.

Common mistakes

• Silence until the original date, then a surprise.
• Weekend heroics that hide the slip.
• Adding people late and expecting a linear speedup.
• Letting a junior's optimism set the team date without a senior pass on the risky tasks.

Check yourself

FraudSvc has not approved your account and it is Wednesday. The self-serve milestone is next week. What do you send?

A yellow status with the blocker, the Friday decision date, and two options: allowlist, or slip self-serve and ship internal-only.

You are late. What three levers do you put on the table?

Cut scope, move the date, or add people — and say that adding people late is usually the weakest.

Key takeaways

• Assumptions travel with the number. When they break, renegotiate early.
• Status is a decision request: blocker, date, options.
• Three levers: scope, date, people. People late in a project rarely save you.
• Do not hide slips with weekends. The signal is the point.

Further learning

A clear talk on why early estimates have huge error bands:

Back to the map: estimation and planning.