Revised Baseline vs Recovery Schedule
Last reviewed 9 September 20261,960 words9 min read
Two documents, routinely confused — one changes the plan, the other changes the yardstick
🟩 In one line: A recovery schedule shows how the contractor will get back to the existing contractual date at its own initiative; a revised baseline replaces the plan of record and needs the other party's written acceptance — issuing one and calling it the other is how projects lose both their EOT and their measurement.
👤 Who this is for: J/M/S. Prerequisites: baseline-approval-and-control, forecasting-honestly, float-ownership-and-contract.
First, let's be honest about why this page exists
"We've re-baselined" is said on site to mean at least three different things: the contractor issued a recovery programme, the Engineer accepted a revised programme under Cl 8.3, or an EOT was awarded and the completion date moved. Only one of those changes the yardstick that BEI, CPLI and the trend register are measured against.
The confusion is not innocent. A recovery schedule that is quietly adopted as the baseline erases the slippage it was written to recover. A revised baseline that is treated as "just a recovery" leaves the project measuring itself against a date the contract no longer contains. Get the vocabulary right and most of the argument disappears.
🟨 The standard — what "good" looks like
| Source | Recovery schedule | Revised baseline |
|---|---|---|
| FIDIC 1999 Cl 8.6 / 2017 Cl 8.7 | Engineer may require revised methods when progress is too slow; Contractor adopts them at its own cost unless the cause is the Employer's | — |
| FIDIC 1999 Cl 8.3 / 2017 Cl 8.3 | Revised programme whenever inconsistent with actual progress (a recovery schedule is one such revision) | The same clause; acceptance does not by itself change the Time for Completion |
| FIDIC 1999 Cl 8.4 / 2017 Cl 8.5 | — | Award of EOT changes the Time for Completion — the trigger for a baseline revision |
| NEC3/NEC4 Cl 32.1–32.2 | Revised programme shows how the Contractor plans to deal with delays | Every accepted revision becomes the Accepted Programme — but Completion Date moves only via a CE |
| NEC3/NEC4 Cl 36 | Acceleration by quotation and agreement — the instructed form of recovery | Agreed acceleration changes the Completion Date and the Prices |
| NEC3 Cl 63.3 / NEC4 Cl 63.5 | — | CE assessment moves planned Completion and hence the Completion Date |
| SCL Protocol 2nd ed., Core Principles 15 and 16 | Contractor mitigates at no additional cost unless instructed; acceleration is compensable only when instructed or agreed | Distinguishes voluntary recovery from paid acceleration |
| GAO Schedule Assessment Guide, Best Practice 10 | — | Maintain a baseline; document every revision with rationale and approval |
| AACE RP 38R-06 | — | SBM revised only with a baseline revision |
| DCMA Metrics 13, 14 | — | Measured against the approved baseline, not a working target |
🟢 Rule: a recovery schedule keeps the contractual completion date and shows a way back to it; a revised baseline changes the plan of record — and only the second needs a signature from the other side.
How it actually works
The comparison
| Recovery schedule | Revised baseline | |
|---|---|---|
| Question it answers | "How do we get back to the date?" | "What is the plan of record now?" |
| Trigger | Forecast completion later than contract; Engineer's Cl 8.6/8.7 request; internal decision | Awarded EOT; agreed acceleration (NEC Cl 36); major re-scope; a contract clause requiring periodic re-baseline |
| Who initiates | Contractor | Either — but always concluded by written acceptance |
| Contractual completion date | Unchanged | Changed (or plan restated to a new agreed date) |
| Forecast completion | Stays as the network says until measures are evidenced | Reset to the new baseline plan |
| Cost | Contractor's, unless instructed acceleration | Agreed through the EOT/CE/acceleration mechanism |
| P6 status | A Reflection / copy named "Recovery Plan R1 — DD 01 Sep"; never assigned as Project Baseline | New baseline in the register, e.g. BL1-Accepted-DD 01Sep25-Rev0; assigned as Project Baseline |
| BEI / CPLI / trend measured against | The existing accepted baseline | The new baseline, with the revision flagged in the trend register |
| SBM | Unchanged (recovery narrative records the changed assumptions) | Revised per AACE 38R-06 |
| Narrative name | Recovery Narrative | Revised Baseline Narrative + Baseline Acceptance letter |
| Approval | Engineer/PM "receives" or "accepts" it as a Cl 8.3 revision — no change to entitlement | Engineer/PM accepts in writing, naming revision and data date |
| Lifetime | Weeks to a few months; tracked weekly; replaced or abandoned | Until the next accepted revision |
| What it must not do | Become the baseline by default; waive EOT entitlement | Hide slippage by restarting the trend; absorb un-awarded claims |
The three dates that live side by side
| Date | Source | Shown as |
|---|---|---|
| Contractual completion | Contract + awarded EOTs | Constrained milestone (Finish On or Before) |
| Forecast completion | Current update network | Unconstrained; reported with basis |
| Recovery target | Recovery plan | Separate row; reported with measures, cost, approvals — never as the forecast |
The recovery target replaces the forecast only when the measures behind it are evidenced (crew on site, instruction issued, procurement confirmed). Until then two dates, two bases.
The three routes a recovery schedule can take
| Route | What happens | Baseline effect |
|---|---|---|
| It works | Forecast converges on the contractual date over successive updates | None — baseline never moved |
| It is accepted as paid acceleration (NEC Cl 36 / FIDIC agreed variation) | Completion Date and Prices change | Baseline revision, register row, SBM revised |
| It fails | Slippage continues; EOT position argued; recovery replaced | None — the failed recovery stays in the archive as evidence of mitigation (SCL CP 15) |
The abuse to watch for
Re-baselining to the current forecast at month 9, with no EOT and no agreement, moves the yardstick to wherever the project happens to be. BEI resets to 1.0, the trend register shows zero variance and the 40 days of slippage vanish from the reports. Reviewers under GAO BP 10 and DCMA look for exactly this: a baseline revision without a documented contractual trigger. The trend register convention — continuous, revision flagged, movement carried across — is the protection.
🟥 Where people go wrong
- Assigning the recovery plan as the P6 Project Baseline. BEI and CPLI now measure against a target nobody signed, and the accepted baseline is one careless "Restore" from being lost. Recovery lives as a Reflection or named copy.
- Calling an accepted Cl 8.3 revision a new baseline. Acceptance of a revised programme is acceptance that it shows actual progress. It does not move the Time for Completion. Only Cl 8.4 / 8.5 does that.
- Issuing a recovery schedule without the reservation sentence. "Submitted without prejudice to the Contractor's entitlement under Cl 8.4 / Cl 60.1." Without it, the other side reads the recovery as an admission that the delay is yours.
- Treating the recovery target as the forecast. The narrative reports 30 April because the recovery says so, while the network still says 14 June. Two dates, two rows, until the measures are evidenced.
- Re-baselining after an EOT without revising the SBM. New completion date, new logic in places, old basis document. The SBM and the baseline register move together.
- Re-baselining to hide slippage. No EOT, no agreement, no re-scope — just a new baseline at the current forecast. Every trend register that carries the movement across the revision will show it, and so will the reviewer.
⚖️ When you're challenged
"You issued a recovery programme and we accepted it. That's the baseline now." You accepted it as a revised programme under 8.3 — it shows how we plan to deal with the delay. The Time for Completion hasn't changed, so the baseline hasn't either. If the Employer wants the recovery adopted as the plan of record with the acceleration paid for, that's an agreement under a different clause and we'd document it as one.
"We got 25 days of EOT. Just update the recovery and carry on." The EOT moves the contract date, so the baseline register gets a new row and the SBM is revised — that's a baseline revision. The recovery plan then targets the remaining 15 days against the new date. Both things happen; they're different documents.
"Why are there two completion dates in your report?" Forecast is what the network says with today's rates. Target is what the recovery plan achieves if the second formwork set arrives and the night shift is approved. When those are on site the forecast will move towards the target. Until then, showing one date would be showing a wish.
📄 Related pages
- Baseline Approval and Control — the register and the acceptance letter
- Building a Recovery Plan — how the recovery document is actually built
- Slippage and Trend Analysis — carrying movement across a revision
- Extension of Time Basics — what an award does to the completion date
- Acceleration and constructive acceleration — when recovery becomes paid
- Schedule Basis Memorandum — revised only with the baseline
- CPLI and BEI — measured against the accepted baseline, never a target
- Float ownership and the contract — why the terminal float question sits underneath this
✏️ Worked example
An office tower in Doha, FIDIC 1999, contract completion 31 March. At Update 09 (DD 1 October) forecast completion is 10 May — 40 days late, of which 25 days are attributed to a late Employer-supplied transformer (claim submitted, notice ref C-118) and 15 to the Contractor's own structure slippage.
| Date | Event | Document | Baseline register | Trend register |
|---|---|---|---|---|
| 5 Oct | Recovery Plan R1 issued targeting 31 March, "without prejudice to Cl 8.4 entitlement" | Recovery Narrative R1; P6 copy "Recovery R1 — DD 01Oct"; not a baseline | No change (BL0 remains) | Forecast 10 May, target 31 Mar, both shown |
| 20 Oct | Engineer accepts R1 as revised programme under Cl 8.3 | Engineer's letter | No change | No change |
| 18 Nov | EOT of 25 days determined under Cl 8.4; contract completion → 25 April | Determination letter | New row: BL1-Accepted-DD 01Dec25-Rev0; SBM Rev 1 | Row flagged "BL1 accepted +25 days"; movement series continues |
| 5 Dec | Recovery Plan R2 issued targeting 25 April (15 days to recover, all Contractor's) | Recovery Narrative R2 | No change | Forecast per Upd 11 network; target 25 Apr |
| 1 Mar | Upd 14 forecast converges to 27 April; R2 measures evidenced | Update 14 | No change | Forecast 27 Apr; target retired |
One EOT, one baseline revision. Two recovery plans, zero baseline revisions. The 40 days of history stayed in the trend register throughout.
📖 References
- FIDIC Conditions of Contract for Construction, 1999, Cl 8.3, 8.4, 8.6; 2017, Cl 8.3, 8.5, 8.7 (check the edition in your contract)
- NEC3 / NEC4 Engineering and Construction Contract, Cl 32.1, 32.2, 36, 63.3 (NEC3) / 63.5 (NEC4) (check the edition in your contract)
- SCL Delay and Disruption Protocol, 2nd ed. (2017), Core Principles 15, 16
- GAO Schedule Assessment Guide (GAO-16-89G), Best Practice 10
- AACE International RP 38R-06, Documenting the Schedule Basis
- DCMA 14-Point Schedule Assessment, Metrics 13 and 14
- Oracle Primavera P6 Professional User Guide — Maintain Baselines; Assign Baselines; Reflections
From the field
Experience from working planners. Unreviewed — read it as experience, not guidance.
Add what you know about revised baseline vs recovery schedule. What worked, what the consultant pushed back on, what you would do differently next time. A paragraph is plenty.
Contributors get their name and one link on the site — your own templates, course or consultancy. We take nothing and hold nothing.
Add your experience