Re-Baselining in EVM: When and How to Reset Your Baseline

<div class="ge-article-wrapper"><nav class="ge-toc" aria-label="Table of contents"><p class="ge-toc-label">In this article</p><ul class="ge-toc-list"><li><a href="#the-definition">The Definition</a></li><li><a href="#the-before-and-after">The Before and After</a></li><li><a href="#when-re-baselining-is-legitimate">When Re-Baselining Is Legitimate</a></li><li><a href="#when-re-baselining-is-hiding-problems">When Re-Baselining Is Hiding Problems</a></li><li><a href="#worked-example-45m-project-at-month-14">Worked Example: £45M Project at Month 14</a></li><li><a href="#the-re-baseline-process">The Re-Baseline Process</a></li><li><a href="#common-mistakes">Common Mistakes</a></li><li><a href="#frequently-asked-questions">Frequently Asked Questions</a></li></ul></nav><article class="ge-article-body"><p>Re-baselining is the formal act of replacing your Performance Measurement Baseline (PMB) with a new one. It resets all your variances to zero. Your <a href="/en/earned-value/cost-schedule-variance">cost variance</a>, your <a href="/en/earned-value/cost-schedule-variance">schedule variance</a>, your cumulative <a href="/en/earned-value/definitions/cost-performance-index">CPI</a>, all wiped clean. That's either exactly the right thing to do or it's the commercial equivalent of resetting the odometer on a used car. The difference depends entirely on why you're doing it. </p><p>Re-baselining is part of the <a href="/en/earned-value/definitions">earned value definitions glossary</a>. For the full formula reference, see the <a href="/en/earned-value/formulas">earned value formulas page</a>. </p><h2 id="the-definition">The Definition</h2><p><strong>Re-baselining = formally replacing the current Performance Measurement Baseline (PMB) with a new baseline that reflects the current approved scope, schedule, and budget</strong></p><p>After re-baselining: </p><ul><li><a href="/en/earned-value/definitions/budget-at-completion">BAC</a> = new approved budget</li><li><a href="/en/earned-value/definitions/planned-value">PV</a> curve = new approved schedule</li><li>All historical variances become irrelevant</li><li><a href="/en/earned-value/definitions/cost-performance-index">CPI</a> and <a href="/en/earned-value/definitions/schedule-performance-index">SPI</a> reset to 1.0 (or close to it)</li></ul><p>There's no formula for re-baselining. It's a management decision with contractual and commercial consequences. And it should be rare. If you're re-baselining more than once on a project, something has gone fundamentally wrong with either the project or the project controls. </p><h2 id="the-before-and-after">The Before and After</h2><p>This is what re-baselining actually does to your EVM data. </p><pre class="ge-ascii-diagram ge-anim"> BEFORE RE-BASELINE (Month 14 of a £45M project) ================================================ Budget ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ BAC = £45M (£M) 45 ┤ │ PV (planned) ────── 40 ┤ ╱╱ │ ╱╱ 35 ┤ ╱╱ │ ╱╱ 30 ┤ ╱╱ │ ╱╱ 25 ┤ ╱╱ │ ╱╱ AC (actual) ━━━━━━ £22.1M 20 ┤ ╱╱ ━━━━━ │ ╱╱ ━━━━ 15 ┤╱━━━━ EV (earned) ∙∙∙∙∙∙ £17.8M ━━ ∙∙∙∙∙∙∙∙ 10 ┤ ∙∙∙∙∙∙∙∙∙∙ PV at M14 = £23.4M │ ∙∙∙∙∙ 5 ┤∙∙ │ 0 ┤────┬────┬────┬────┬────┬────┬────┬──── 0 2 4 6 8 10 12 14 Month CV = EV - AC = £17.8M - £22.1M = -£4.3M (OVER BUDGET) SV = EV - PV = £17.8M - £23.4M = -£5.6M (BEHIND PROGRAMME) CPI = 0.806 SPI = 0.761 Verdict: Project is 20% over budget and 24% behind schedule. These numbers have been getting worse for 8 months. AFTER RE-BASELINE (New baseline at Month 14) ============================================= Budget ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ NEW BAC = £52M (£M) 52 ┤ │ New PV ────── 45 ┤ ╱╱ │ ╱╱ 40 ┤ ╱╱ │ ╱╱ 35 ┤ ╱╱ │ ╱╱ 30 ┤ ╱╱ │ ╱╱ 25 ┤ ╱╱ │ ╱╱ AC ━━ £22.1M 20 ┤╱ ━━━━━━ │ ━━━━━━ 15 ┤━━━━ EV ∙∙ £22.1M (re-measured) │ ∙∙∙∙∙∙∙∙∙∙∙∙∙ 10 ┤∙∙∙∙ │ 5 ┤ │ 0 ┤────┬────┬────┬────┬────┬────┬────┬──── 0 2 4 6 8 10 12 14 Month CV = £22.1M - £22.1M = £0 (RESET) SV = £22.1M - £22.1M = £0 (RESET) CPI = 1.000 SPI = 1.000 History erased. Variances gone. The £4.3M overrun and £5.6M schedule slip are now invisible in the EVM data. </pre><p>That second chart should make you uneasy. The project hasn't improved. The same amount of work has been done for the same cost. But the numbers now say everything is fine. This is why re-baselining is sometimes called "rubber baselining", stretching the baseline to fit whatever's actually happened. </p><h2 id="when-re-baselining-is-legitimate">When Re-Baselining Is Legitimate</h2><p>There are genuine reasons to re-baseline. Not many. But they exist. </p><p><strong>1. Fundamental scope change.</strong> If the client adds or removes a significant portion of the project, the original baseline no longer represents the approved scope. On NEC4, multiple implemented compensation events that collectively change the target by more than 15-20% often warrant a re-baseline rather than endless incremental BAC adjustments. </p><p><strong>2. Force majeure or suspension.</strong> If the project is suspended for 6 months due to factors outside anyone's control, the original schedule baseline is meaningless. Re-baseline to reflect the new reality. </p><p><strong>3. Contract restructuring.</strong> If the contract form changes (e.g., Option A converted to Option C mid-project), the entire baseline framework needs rebuilding. </p><p><strong>4. Original baseline was fundamentally flawed.</strong> Sometimes the tender programme was unrealistic, the cost plan was based on wrong quantities, or the WBS structure was unworkable. If you can't measure against the baseline because it was never right, replacing it is honest, not evasive. </p><h2 id="when-re-baselining-is-hiding-problems">When Re-Baselining Is Hiding Problems</h2><p><strong>"We're behind programme, so let's extend the baseline."</strong> No. That's not a re-baseline. That's concealing delay. If you're behind programme, the EVM data should show it. Fix the performance, not the measurement. </p><p><strong>"CPI has been below 0.85 for six months, let's reset."</strong> No. A low CPI means you're spending more than you should for the value earned. Re-baselining makes the CPI look better without changing the underlying cost problem. </p><p><strong>"The client doesn't like the numbers."</strong> Definitely not. EVM data should tell the truth. If the truth is uncomfortable, the response is corrective action, not cosmetic adjustment. </p><h2 id="worked-example-45m-project-at-month-14">Worked Example: £45M Project at Month 14</h2><span class="ge-worked-label">Worked Example</span><div class="ge-callout ge-anim"><p><strong>Scenario:</strong> A £45M NEC4 Option C flood defence scheme in East Anglia. At month 14, the project has been significantly impacted by a fundamental design change to the main pumping station, the client changed the design standard from a 1-in-100-year to a 1-in-200-year flood protection level. This was instructed as a compensation event in month 8 and implemented at £7M.</p><br><p><strong>Current state (pre-re-baseline):</strong></p><p>- Original BAC: £45M</p><p>- Adjusted BAC (after CE): £52M</p><p>- AC to date: £22.1M</p><p>- EV (against original scope): £17.8M</p><p>- CPI: 0.806</p><p>- SPI: 0.761</p><br><p>The CPI looks terrible. But here's the thing, much of the "overrun" is actually the team delivering the additional pumping station scope that wasn't in the original baseline. The EV is still being measured against the original £45M baseline, which doesn't include the new scope.</p><br><p><strong>The case for re-baselining:</strong></p><p>- Scope has genuinely changed by 15.5% (£7M / £45M)</p><p>- The original PV curve doesn't reflect the additional pumping station works</p><p>- The programme has been rescheduled to accommodate the new design</p><p>- Continuing to measure against the original baseline produces misleading metrics</p><br><p><strong>The re-baseline:</strong></p><p>- New BAC: £52,000,000</p><p>- New PV curve: reflects revised programme including pumping station redesign</p><p>- EV re-measured: £22,100,000 (re-assessed against new scope at month 14)</p><p>- AC: £22,100,000 (unchanged, actual spend doesn't change)</p><p>- New CPI: 1.000</p><p>- New SPI: 1.000</p><br><p><strong>Critical requirement:</strong> The project team retains the pre-re-baseline data in an appendix to the monthly report. The £4.3M cost variance and £5.6M schedule variance from the original baseline are documented, with explanations showing how much was due to the scope change and how much was genuine performance variance.</p><br><p>After analysis: £3.8M of the cost variance traces directly to the pumping station CE scope. The remaining £500K is genuine performance variance that the team needs to address going forward. That £500K doesn't disappear. It becomes a management action in the new baseline period.</p></div><h2 id="the-re-baseline-process">The Re-Baseline Process</h2><p>If you've determined that re-baselining is legitimate, do it properly. </p><ol><li><strong>Document the reason.</strong> Written justification approved by the project sponsor and client.</li><li><strong>Quantify the scope change.</strong> What's been added, removed, or changed?</li><li><strong>Rebuild the WBS and cost baseline.</strong> New BAC reflecting current approved scope.</li><li><strong>Rebuild the schedule baseline.</strong> New PV curve reflecting the current approved programme.</li><li><strong>Re-measure EV at the re-baseline date.</strong> Physical progress against the new scope.</li><li><strong>Archive the old baseline.</strong> Never delete historical data. Keep it in the project records.</li><li><strong>Note the re-baseline in every subsequent EVM report.</strong> "Project re-baselined at Month 14. Pre-re-baseline data available in Appendix C."</li></ol><h2 id="common-mistakes">Common Mistakes</h2><p><strong>Re-baselining without client approval.</strong> On NEC4, the target total of the Prices can only change through implemented compensation events. You can't unilaterally re-baseline your EVM system without the contractual basis. If the CEs have been implemented, the BAC change is already authorised. If they haven't, you've got no basis for a new baseline. </p><p><strong>Losing the historical data.</strong> The whole point of EVM is to track performance over time. If you re-baseline and delete the old data, you've destroyed the trend analysis. Archive everything. Keep both baselines visible in your reporting. </p><p><strong>Re-baselining too often.</strong> Once, maybe twice on a large programme. If you're doing it quarterly, your baseline is meaningless and your EVM system is a fiction. Fix the project, not the measurement. </p><p><strong>Not separating genuine performance variance from scope change variance.</strong> Before re-baselining, decompose the total variance into scope-driven and performance-driven components. Only the scope-driven variance should be absorbed into the new baseline. Performance variance needs to be carried forward as a known issue. </p><div class="ge-product-note ge-anim"><p><strong>How Gather helps.</strong> Gather's AI reads your site diaries daily and maps progress against your cost-loaded programme, giving you accurate earned value data without manual spreadsheet updates. <a href="https://gatherinsights.com/contact">Book a demo</a> to see it working on a live NEC4 project.</p></div><h2 id="frequently-asked-questions">Frequently Asked Questions</h2><h3>Does NEC4 have a formal re-baselining mechanism?</h3><p>Not as such. NEC4 adjusts the target through compensation events (clause 65). Each implemented CE changes the target total of the Prices, which is equivalent to incremental BAC adjustments. A full re-baseline, rebuilding the PMB from scratch, is an EVM management decision, not a contract mechanism. But it should always align with the contractual target. </p><h3>How does re-baselining affect <a href="/en/earned-value/definitions/estimate-at-completion">EAC</a> forecasts?</h3><p>After re-baselining, your <a href="/en/earned-value/definitions/estimate-at-completion">EAC</a> restarts from the new BAC. If EAC = BAC / CPI and your CPI is now 1.0, the EAC equals the new BAC. This means your forecast assumes perfect performance from this point forward. That's optimistic. I'd recommend running <a href="/en/earned-value/definitions/estimate-at-completion">EAC</a> with a blended CPI that incorporates pre-re-baseline performance for at least 3 months after the reset. </p><h3>What's the difference between re-baselining and Over Target Baseline (OTB)?</h3><p>An OTB is a specific type of re-baseline where the new BAC exceeds the original contract budget. It formally acknowledges that the project will cost more than originally authorised. In NEC4 terms, this would be a target increase through CEs. Re-baselining is the broader concept. It can involve scope reduction, scope increase, or schedule restructuring without necessarily changing the total budget. </p></article></div>