Earned Value

What Is a Control Account in EVM? Project Controls Explained

A control account is the management control point where scope, budget, and schedule are integrated and where EVM performance is measured. It's the fundamental building block of earned value.

Will Doyle

Mar 08, 2026 · 5 min read

<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-wbs-x-obs-matrix">The WBS x OBS Matrix</a></li><li><a href="#why-control-accounts-are-the-heart-of-evm">Why Control Accounts Are the Heart of EVM</a></li><li><a href="#how-to-set-up-control-accounts">How to Set Up Control Accounts</a></li><li><a href="#worked-example-control-accounts-on-a-35m-building-project">Worked Example: Control Accounts on a £35M Building Project</a></li><li><a href="#how-variance-analysis-rolls-up">How Variance Analysis Rolls Up</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>A control account (CA) is the intersection of a <a href="/en/earned-value/definitions/work-breakdown-structure">Work Breakdown Structure</a> (WBS) element and an <a href="/en/earned-value/definitions/organisational-breakdown-structure">Organisational Breakdown Structure</a> (OBS) element, the point where scope, schedule, and budget come together under one person's responsibility. It's where <a href="/en/earned-value">earned value</a> is actually measured. Not at the programme level. Not in a summary dashboard. At the control account, where a named manager owns a defined chunk of work with a specific budget and timeline.</p><p>If you've ever tried to run EVM at the whole-project level and wondered why the numbers feel meaningless, this is probably why. You need control accounts.</p><p>This page is part of the <a href="/en/earned-value/definitions">earned value definitions glossary</a>. For how control accounts feed into variance reporting, see the <a href="/en/earned-value/cost-schedule-variance">cost and schedule variance page</a>.</p><h2 id="the-definition">The Definition</h2><div class="ge-formula-box ge-anim"><span class="ge-formula-label">Formula</span><code>Control Account = the intersection of WBS and OBS where scope, schedule, and budget are assigned to a single responsible manager</code></div><p>Every control account has four things:</p><ol><li><strong>Scope</strong>: a defined portion of the WBS (what work is included)</li><li><strong>Schedule</strong>: start date, end date, and milestones</li><li><strong>Budget</strong>: a portion of <a href="/en/earned-value/definitions/budget-at-completion">BAC</a> allocated to this work</li><li><strong>Owner</strong>: one person who's accountable for performance</li></ol><p>If it doesn't have all four, it's not a control account. It's just a cost code.</p><h2 id="the-wbs-x-obs-matrix">The WBS x OBS Matrix</h2><p>This is easier to understand visually. Picture the WBS as columns (what work) and the OBS as rows (who's responsible). Each cell where they intersect is a potential control account.</p><pre class="ge-ascii-diagram ge-anim"> WORK BREAKDOWN STRUCTURE (WBS) ┌───────────┬───────────┬───────────┬───────────┐ │Substructure│Superstruc-│ M&E │ External │ │ │ ture │ │ Works │ ┌────────────┼───────────┼───────────┼───────────┼───────────┤ O │ Package A │ │ │ │ │ B │ Manager: │ CA-01 │ CA-02 │ │ │ S │ J. Smith │ £3.2M │ £4.8M │ │ │ ├────────────┼───────────┼───────────┼───────────┼───────────┤ │ Package B │ │ │ │ │ │ Manager: │ │ CA-03 │ CA-04 │ CA-05 │ │ R. Patel │ │ £2.1M │ £5.6M │ £1.8M │ ├────────────┼───────────┼───────────┼───────────┼───────────┤ │ Package C │ │ │ │ │ │ Manager: │ CA-06 │ │ CA-07 │ CA-08 │ │ S. Davies │ £1.9M │ │ £3.4M │ £2.2M │ ├────────────┼───────────┼───────────┼───────────┼───────────┤ │ Prelims │ │ │ │ │ │ Manager: │ CA-09 (spans all WBS) │ │ │ K. Murphy │ £4.5M │ │ └────────────┴───────────┴───────────┴───────────┴───────────┘ 9 control accounts. Total BAC = £29.5M + contingency. Each CA has one owner. No overlap. No gaps. </pre><p>Not every intersection needs to be a control account. Some WBS/OBS combinations don't have meaningful work. The goal is enough granularity to manage effectively, but not so much that you're drowning in data. On a £35M project, 8 to 15 control accounts is typical. On a £200M programme, you might have 40 to 60.</p><h2 id="why-control-accounts-are-the-heart-of-evm">Why Control Accounts Are the Heart of EVM</h2><p>Every EVM metric is calculated at the control account level first, then rolled up. That's a critical point most people miss. You don't calculate project-level <a href="/en/earned-value/definitions/cost-performance-index">CPI</a> and hope for the best. You calculate CPI for each control account, identify which ones are off track, and investigate those specifically.</p><p>Here's why that matters commercially. A project-level CPI of 0.94 tells you performance is 6% below plan. Useful, but what do you do about it? A control account breakdown tells you:</p><div class="ge-table-wrap ge-anim"><table class="ge-table"><thead><tr><th>Control Account</th><th>Budget</th><th>CPI</th><th>Status</th></tr></thead><tbody><tr><td>CA-01 Substructure (Smith)</td><td>£3,200,000</td><td>1.02</td><td>On track</td></tr><tr><td>CA-02 Superstructure (Smith)</td><td>£4,800,000</td><td>0.97</td><td>Minor concern</td></tr><tr><td>CA-04 M&E (Patel)</td><td>£5,600,000</td><td>0.83</td><td>Problem</td></tr><tr><td>CA-05 External Works (Patel)</td><td>£1,800,000</td><td>1.05</td><td>On track</td></tr><tr><td>CA-09 Prelims (Murphy)</td><td>£4,500,000</td><td>0.91</td><td>Concern</td></tr></tbody></table></div><p>Now you know. The project's CPI problem lives in M&E (CA-04) and prelims (CA-09). Smith's packages are fine. You need to sit down with Patel about M&E subcontractor productivity and with Murphy about prelim burn rate.</p><p>Without control accounts, you'd be staring at a single 0.94 figure and guessing where the problem is. I've seen commercial managers spend weeks doing deep-dives into the wrong package because they didn't have this breakdown. Weeks. Completely wasted.</p><h2 id="how-to-set-up-control-accounts">How to Set Up Control Accounts</h2><h3>Step 1: Build the WBS</h3><p>The WBS breaks the project scope into manageable chunks. On UK construction projects, the WBS typically follows physical elements:</p><pre class="ge-ascii-diagram ge-anim"> Level 1: Project Level 2: Substructure | Superstructure | M&E | Externals | Prelims Level 3: Piling | Foundations | Ground beams | Drainage Level 4: Individual activities (too granular for CAs) </pre><p>Control accounts usually sit at WBS Level 2 or 3. Level 4 is too detailed, you'd spend more time administering EVM than managing the project.</p><h3>Step 2: Build the OBS</h3><p>The OBS mirrors your management structure:</p><pre class="ge-ascii-diagram ge-anim"> Level 1: Project Director Level 2: Construction Manager | Commercial Manager | Design Manager Level 3: Package Managers (by trade or section) </pre><h3>Step 3: Map Intersections</h3><p>Each control account must have:</p><ul><li>One WBS element (no splitting a WBS element across two CAs)</li><li>One OBS element (one responsible person)</li><li>A budget that sums to BAC when all CAs are totalled</li><li>Start and end dates from the <a href="/en/earned-value/definitions/accepted-programme">Accepted Programme</a></li></ul><h3>Step 4: Assign Work Packages</h3><p>Within each control account, create work packages, the lowest level of detail. These are the units of work that get scheduled, costed, and measured. A control account might contain 5 to 20 work packages.</p><h2 id="worked-example-control-accounts-on-a-35m-building-project">Worked Example: Control Accounts on a £35M Building Project</h2><span class="ge-worked-label">Worked Example</span><div class="ge-callout ge-anim"><p><strong>Scenario:</strong> A £35M NEC4 Option C new-build office complex in Birmingham. Three buildings, shared basement car park, extensive landscaping. Programme: April 2025 to September 2027. The commercial team sets up 12 control accounts across 4 organisational packages.</p><p><strong>Control Account Structure:</strong></p><div class="ge-table-wrap ge-anim"><table class="ge-table"><thead><tr><th>CA</th><th>WBS Element</th><th>OBS Owner</th><th>Budget</th><th>Duration</th></tr></thead><tbody><tr><td>CA-01</td><td>Demolition & Enabling</td><td>T. Williams</td><td>£1,800,000</td><td>Apr-Jul 2025</td></tr><tr><td>CA-02</td><td>Substructure (all buildings)</td><td>T. Williams</td><td>£4,200,000</td><td>Jun-Nov 2025</td></tr><tr><td>CA-03</td><td>Superstructure Building A</td><td>J. Okafor</td><td>£5,100,000</td><td>Sep 2025-May 2026</td></tr><tr><td>CA-04</td><td>Superstructure Building B</td><td>J. Okafor</td><td>£4,800,000</td><td>Nov 2025-Jul 2026</td></tr><tr><td>CA-05</td><td>Superstructure Building C</td><td>J. Okafor</td><td>£3,900,000</td><td>Jan-Sep 2026</td></tr><tr><td>CA-06</td><td>Basement Car Park</td><td>T. Williams</td><td>£2,600,000</td><td>Jul-Dec 2025</td></tr><tr><td>CA-07</td><td>M&E Installation</td><td>D. Khan</td><td>£5,400,000</td><td>Mar 2026-Jun 2027</td></tr><tr><td>CA-08</td><td>Fit-out & Finishes</td><td>D. Khan</td><td>£3,200,000</td><td>Sep 2026-Aug 2027</td></tr><tr><td>CA-09</td><td>External Works & Landscaping</td><td>P. Evans</td><td>£1,500,000</td><td>Apr-Sep 2027</td></tr><tr><td>CA-10</td><td>Prelims & Site Overheads</td><td>P. Evans</td><td>£2,800,000</td><td>Apr 2025-Sep 2027</td></tr><tr><td>CA-11</td><td>Design & Professional Fees</td><td>R. Begum</td><td>£1,200,000</td><td>Apr 2025-Sep 2027</td></tr><tr><td>CA-12</td><td><a href="/en/earned-value/definitions/contingency-reserve">Contingency Reserve</a></td><td>Project Director</td><td>£1,500,000</td><td>Not profiled</td></tr><tr><td></td><td><strong>Total (BAC)</strong></td><td></td><td><strong>£38,000,000</strong></td><td></td></tr></tbody></table></div><p>Note: the £38M total is £35M contract value plus £3M contingency and design fees. (On NEC4 Option C, the £35M is the target total of the Prices excluding fee.)</p><p><strong>Month 9 Variance Report (December 2025):</strong></p><div class="ge-table-wrap ge-anim"><table class="ge-table"><thead><tr><th>CA</th><th>PV</th><th>EV</th><th>AC</th><th>CV</th><th>SV</th><th>CPI</th><th>SPI</th></tr></thead><tbody><tr><td>CA-01</td><td>£1,800,000</td><td>£1,800,000</td><td>£1,740,000</td><td>+£60,000</td><td>£0</td><td>1.03</td><td>1.00</td></tr><tr><td>CA-02</td><td>£3,800,000</td><td>£3,500,000</td><td>£3,680,000</td><td>-£180,000</td><td>-£300,000</td><td>0.95</td><td>0.92</td></tr><tr><td>CA-03</td><td>£2,040,000</td><td>£1,900,000</td><td>£2,100,000</td><td>-£200,000</td><td>-£140,000</td><td>0.90</td><td>0.93</td></tr><tr><td>CA-04</td><td>£960,000</td><td>£800,000</td><td>£850,000</td><td>-£50,000</td><td>-£160,000</td><td>0.94</td><td>0.83</td></tr><tr><td>CA-06</td><td>£2,200,000</td><td>£2,340,000</td><td>£2,280,000</td><td>+£60,000</td><td>+£140,000</td><td>1.03</td><td>1.06</td></tr><tr><td>CA-10</td><td>£840,000</td><td>£840,000</td><td>£920,000</td><td>-£80,000</td><td>£0</td><td>0.91</td><td>1.00</td></tr><tr><td><strong>Totals</strong></td><td><strong>£11,640,000</strong></td><td><strong>£11,180,000</strong></td><td><strong>£11,570,000</strong></td><td><strong>-£390,000</strong></td><td><strong>-£460,000</strong></td><td><strong>0.97</strong></td><td><strong>0.96</strong></td></tr></tbody></table></div><p><strong>Analysis at the control account level reveals:</strong></p><ul><li><strong>CA-01 (Demolition)</strong>: Complete. Came in £60K under budget. Williams delivered well.</li><li><strong>CA-02 (Substructure)</strong>: CPI 0.95, SPI 0.92. Piling took longer than planned (unexpected obstructions). Behind programme but cost impact is moderate. Williams needs to recover 2 weeks in Q1 2026.</li><li><strong>CA-03 (Building A Super)</strong>: CPI 0.90 is the red flag. Okafor's concrete frame subcontractor is underperforming. The commercial team needs to review subcontract productivity rates and consider a formal early warning.</li><li><strong>CA-04 (Building B Super)</strong>: SPI 0.83 is concerning. Only just started and already 17% behind programme. Steel delivery delays. Okafor should raise an <a href="/en/nec4/early-warnings">early warning</a> under clause 15.</li><li><strong>CA-06 (Basement)</strong>: Ahead of programme and under budget. Good news.</li><li><strong>CA-10 (Prelims)</strong>: CPI 0.91. Prelim costs running 9% hot. Not unusual, but needs watching. Tower crane rental uplift might be driving it.</li></ul><p><strong>Without the control account breakdown</strong>, the project CPI of 0.97 looks almost fine. You might not investigate. With the breakdown, you see that CA-03's 0.90 is being masked by CA-01 and CA-06's strong performance. The problem is specific and actionable.</p></div><h2 id="how-variance-analysis-rolls-up">How Variance Analysis Rolls Up</h2><p>Control accounts don't exist in isolation. They roll up through the WBS to give programme-level and project-level views.</p><pre class="ge-ascii-diagram ge-anim"> PROJECT LEVEL CPI = 0.97 SPI = 0.96 │ ├── Substructure CPI = 0.97 (CA-01 + CA-02 + CA-06) │ ├── Superstructure CPI = 0.92 (CA-03 + CA-04 + CA-05) ← PROBLEM │ ├── M&E CPI = N/A (CA-07 not yet started) │ ├── Externals CPI = N/A (CA-09 not yet started) │ └── Overheads CPI = 0.92 (CA-10 + CA-11) ← WATCH </pre><p>The roll-up works both ways. Senior management sees the project-level summary. Package managers see their control account detail. The commercial manager lives in the middle, monitoring control account CPIs and flagging outliers.</p><p>On a well-run project, the monthly EVM report has a project-level dashboard on page 1 and control account detail from page 2 onwards. If your report only has the summary, you're missing the point.</p><h2 id="common-mistakes">Common Mistakes</h2><p><strong>1. Too many control accounts.</strong> More isn't better. Thirty control accounts on a £15M project means each one averages £500K. At that level, you're spending more time collecting data than using it. Keep it manageable. Eight to fifteen for a typical project.</p><p><strong>2. Too few control accounts.</strong> Three control accounts on a £50M project means each one averages £17M. That's too coarse. You can't identify problems until they're massive. A CPI of 0.95 on a £17M control account hides £850K of overspend somewhere.</p><p><strong>3. No single owner.</strong> If two people share a control account, nobody owns it. When CPI drops, they point at each other. One CA, one owner. Non-negotiable.</p><p><strong>4. Budget doesn't sum to BAC.</strong> Every pound of BAC must be allocated to a control account (including contingency, if that's held as a separate CA). If the sum of all CA budgets is less than BAC, you have undistributed budget floating around. If it's more, you've double-counted.</p><p><strong>5. Not aligning CAs to the programme.</strong> Each control account must map to activities in the <a href="/en/earned-value/definitions/accepted-programme">Accepted Programme</a>. If you can't pull <a href="/en/earned-value/definitions/planned-value">PV</a> from the programme for a specific CA, you can't calculate SPI. The WBS in EVM and the WBS in the programme must match.</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>How many control accounts should a project have?</h3><p>There's no fixed rule, but a practical guideline is one control account for every £2M to £5M of budget. A £20M project might have 6 to 10 CAs. A £100M programme might have 25 to 40. The test is: can you produce a meaningful <a href="/en/earned-value/definitions/cost-performance-index">CPI</a> for each CA, and can the CA owner actually explain their variance each month? If they can't, the CAs are either too granular or poorly defined.</p><h3>What's the difference between a control account and a work package?</h3><p>Work packages sit inside control accounts. A control account is the management control point, the level at which EVM is formally measured and reported. Work packages are the detailed units of work within a CA. A control account for "M&E Installation" at £5.4M might contain 15 work packages: HVAC ductwork, electrical first fix, fire alarm, plumbing rough-in, and so on. EVM can be measured at <a href="/en/earned-value/definitions/work-package">work package</a> level too, but it rolls up to the CA for reporting.</p><h3>Can a control account span multiple WBS elements?</h3><p>In strict EVM practice, no. Each CA should map to a single WBS element and a single OBS element. In practice on UK construction projects, some CAs do span multiple WBS elements (prelims is a common one. It touches everything). The key is that the CA must still have one clear owner and a defined scope that doesn't overlap with other CAs.</p><h3>What happens when a control account finishes over budget?</h3><p>The overspend is locked in and visible in the project's cumulative <a href="/en/earned-value/cost-schedule-variance">cost variance</a>. You can't move budget from other CAs to cover it retroactively (that's called "budget gaming" and it destroys EVM integrity). The overspend feeds into <a href="/en/earned-value/definitions/estimate-at-completion">EAC</a> and is a lesson for future forecasting. If the overspend is due to an unidentified risk, it may justify drawing on management reserve, but that's a project-level decision that changes BAC.</p></article></div>