Skip links

The First 90 Days: Re-baselining a Data Center Programme After Contract Award

5 min read. For project directors and commercial leads on newly awarded data center projects.

Key takeaways

  • A tender programme is a sales document. Treat the first 90 days after award as the window to turn it into a delivery document.
  • Re-baseline while float still exists: sequence logic, procurement lead times and interface milestones are all still negotiable.
  • A baseline nobody believes in is worse than no baseline at all, because it hides the real critical path.

Every contractor knows the tender programme was written to win the job. Durations were compressed, procurement was assumed frictionless, and commissioning was drawn as a tidy block at the end. None of this is scandalous; it is how tenders work. The scandal is what happens next on many data center projects: the tender programme quietly becomes the contract baseline, and everyone spends the next two years explaining variances against a plan that was never meant to be delivered.

Why the first 90 days matter

Immediately after award, three things are still true that will not be true six months later. First, float exists, and it can be spent deliberately on de-risking the sequence rather than absorbed silently by early slippage. Second, procurement is still open: switchgear, generators and chillers with long lead times can still be ordered against a realistic need date rather than an optimistic one. Third, the client relationship is at its most constructive. Renegotiating logic and milestones in month two is a planning conversation. The same conversation in month eight is a claim.

What a real re-baseline covers

  • Sequence logic. Every activity gets real predecessors and successors. Open ends and date constraints are removed or justified one by one.
  • Procurement reality. Quoted lead times from actual suppliers replace tender assumptions, with delivery dates driving the MEP installation sequence.
  • Commissioning integration. The commissioning sequence is built backwards from IST and connected to the MEP completions it depends on.
  • Interface milestones. Every handover between trades gets a named owner and a date in the programme.
  • Honest durations. Production rates are sanity-checked against comparable projects, not against what the tender needed them to be.

Renegotiating a milestone in month two is a planning conversation. The same conversation in month eight is a claim.

The baseline test

A useful test for any data center baseline: would the people who have to deliver it sign their name to it? If the MEP contractor does not believe the energization date and the commissioning manager does not believe the IST window, the baseline is not a plan. It is a reporting fiction, and it will hide the real critical path until the moment it is most expensive to discover.

Leave a comment