Agency Realization Rate: Find Fixed-Fee Margin Leaks

How can a fixed-fee agency become busier and less profitable in the same week?

The team records more client hours. Utilization rises. The project fee stays fixed. If those extra hours sit outside the estimate or the agreed scope, the standard value of delivery grows while the money attached to it does not.

The agency got busier. Margin fell.

Agency realization rate is the control that makes this gap visible, but the term is used for several different calculations. Some firms compare revenue with the standard value of worked hours. Some compare billed value with billable value. Others fold invoice collection into the same label. Put those versions in one dashboard and a precise percentage can still be wrong for the decision.

For fixed-fee work, use one delivery realization rate and keep time completeness, invoicing, collections, and project margin beside it as separate views. This article shows the record, the formula, and the weekly decisions that follow. It is an operating framework, not accounting advice.

The short answer

Use this management formula for a fixed-fee engagement:

Fixed-fee delivery realization = fee assigned to delivered scope / standard value of approved client work for the same scope and period × 100

The denominator is not the invoice. It is the internal standard value of approved delivery, usually calculated from approved client hours by role multiplied by the agency's own standard role rates (a comparison policy, not a client charge).

The numerator is the fee the agency assigns to that delivered scope and period under its contract and finance policy. For a completed project, that may be the contracted fee plus approved change orders. During delivery, finance must apply one consistent period method. Do not swap in the latest invoice or the cash received because those records answer different questions.

Keep the result at engagement or portfolio level. Realization is not an employee productivity score.

Utilization and realization answer different questions

Billable utilization asks how much available capacity became approved client work. Realization asks how much commercial value the agency retained from the work it delivered.

The distinction matters because a strong utilization result can coexist with weak realization. Another hour on a time-and-materials project may create another billable unit. Another hour on fixed-fee work usually creates no new fee unless a scope change earns an approved change order.

While reading current product documentation, I found the cleanest warnings below the headline metrics. Runn says missing actuals can make fixed-price project measures inaccurate, while Harvest explains that rate-based billable value can differ from the fixed-fee invoice. Those details changed the order of this framework: establish the evidence and period first, then calculate the percentage. The help pages are useful on their own terms, but neither one establishes an accounting policy for an agency.

A fixed-fee example where more work lowers realization

Consider a labeled hypothetical. An agency agrees a $24,000 fixed fee. Its internal standard-value rate for the delivery mix is $150 per approved client hour.

The extra 60 hours may all be legitimate client work. They may also increase billable utilization. Yet delivery realization falls because the standard value of approved work grew and the fixed fee did not.

This does not prove the account team failed. I cannot tell from 72.7% alone whether the cause was an estimate miss, an intentional discount, a quality decision, unapproved scope, rework, or a weak standard-rate policy. That uncertainty belongs in the review. A percentage should open the investigation, not close it.

The result can also mislead in the other direction. Missing time makes the denominator too small. A low internal standard rate makes delivery look better than the agency's current economics justify. A front-loaded fee allocation can make one week look exceptional and the next one poor even when delivery is steady.

This is why I would not copy a universal realization benchmark. Current agency surveys cover different service mixes, commercial models, team structures, and cost bases. A borrowed target can hide a broken definition with a respectable-looking number.

Build a Four-View Agency Realization Record

One rate is useful. One rate without its controls is fragile.

The Four-View Agency Realization Record keeps the commercial signal tied to its evidence. Use one row per engagement and review period, then preserve the version when scope or finance policy changes.

The evidence view comes first. Current Runn documentation makes the reason explicit: fixed-price performance depends on actual work data, and missing actuals distort the measures. Productive's fixed-price documentation confirms the other half of the problem: the service amount stays fixed even when more hours are added.

The billing view stays separate because an invoice can lag delivery or lead it. Harvest's current reporting guidance notes that rate-based billable amounts may not match a fixed-fee invoice. A delayed invoice is not automatically delivery leakage, and delivery leakage does not disappear because an invoice went out.

The margin view also stands alone. Teamwork defines project profit as revenue less cost and margin as profit divided by revenue in its profitability report. Realization can point toward commercial leakage, but it cannot replace direct cost or profit.

Revenue state needs particular care. IFRS 15 ties revenue to satisfaction of performance obligations and, for work recognized over time, an appropriate measure of progress. An invoice date, cash date, and revenue-recognition date can differ. Ask the person responsible for the books to define the revenue field for the agency's jurisdiction and policy, then use that field consistently.

Diagnose the leak before changing the fee

A low rate has several possible owners. Sending every exception to the sales lead guarantees that data, delivery, and billing problems will be treated as pricing problems.

Use this reconciliation:

The recurring operator pattern is quiet absorption. An account lead agrees to a small addition, the delivery team completes it, and nobody changes the commercial record because the request felt too small to escalate. Repeat that pattern and the fee loses meaning one polite exception at a time.

A practitioner thread on fixed-budget projects offers a useful, modest response: keep an out-of-scope slide in the weekly client review and route each addition toward an estimate, deferral, or omission. It is an anecdote, not a benchmark. Still, the process is sound because it gives the client a decision before the agency turns the request into invisible labor.

The UK government's agile contracting guidance reaches a related conclusion from a public-procurement setting: an imperfect specification can produce late change requests, overruns, or scope reduction. That guidance is not private-agency legal advice. The transferable point is narrower: fixed price works against a defined boundary, and uncertainty needs an agreed commercial route.

Scope control is a choice architecture

"Out of scope" is not a complete management action. It tells the client what the agency will not do, but it offers no useful next state.

Give every new request four destinations:

1. Exchange it. Remove an equivalent piece of agreed scope and preserve the fee and delivery window.
2. Price it. Approve a change order with its own fee, acceptance rule, and timing.
3. Defer it. Put the request into a later phase or retained backlog with no current delivery promise.
4. Decline it. Protect the intended outcome when the addition would weaken quality, timing, or focus.

Basecamp documents a similar operating idea in Shape Up: fixed time, variable scope. Its teams set an appetite and reduce nonessential scope to fit it. Client work adds contracts, acceptance, and relationship duties that a product team does not carry, so the method cannot be copied whole. The useful principle survives: when one boundary stays fixed, another boundary needs an explicit decision.

If an architecture-practice owner asked how to interpret a low realization month, I would start with the exception log, not the individual timesheets. Lump-sum professional work can absorb revisions, approval delays, and coordination changes long before a monthly finance view explains the loss. The engagement remains the unit of control.

I think this is also the fairer employee policy. Approved time shows where delivery effort went; it does not prove who caused a commercial gap. A seller can under-scope, a manager can accept extras, a client can delay a dependency, an internal standard rate can be stale, and finance can assign the wrong period. The activity-is-not-productivity framework covers the wider evidence managers should use before drawing a performance conclusion.

Run the review weekly and close it monthly

The review should be short enough to happen and complete enough to produce a decision.

And monthly? Restate incomplete weeks, reconcile approved change orders, confirm the revenue field with finance, and compare closed engagements by service and commercial model. Do not mix retainers, fixed projects, and time-and-materials work into one target until each model has a stable definition.

One discovery in the source trail deserves emphasis: time capture is necessary, but it is not the commercial control. De Jong Phillips describes finding added requirements after it began tracking time, including work that client managers had not repriced. The lesson is not "track harder." It is that delivery evidence must reach the person who can change scope, fee, or future estimates.

The common failure is waiting for project completion because the final margin will be more accurate. It will. It will also arrive after the agency has delivered every recoverable decision. Weekly realization is an intervention signal; the monthly close is the authoritative restatement.

Where KordanoTime fits

KordanoTime opens access on December 1, 2026. The first release captures online and offline time in a daily timeline with projects, tasks, screenshots, and screen video. It includes schedules, attendance, approvals, leave, breaks, office-versus-remote reporting, payroll integrations or payroll-ready exports, QuickBooks or more than 60 integrations, mouse and keyboard jiggler detection, bring-your-own media storage for screenshots and screen recordings in Amazon S3 or Cloudflare R2, custom subdomains, multiple currencies, and admin impersonation.

That evidence base carries approved delivery into the realization review while the commercial judgement stays with the agency. You set the standard-value rate card, the scope version, the fee assigned to the period, and the exception route. KordanoTime captures and carries the time, project, schedule, approval, and export records that make those choices defensible.

Early Access includes 500 screenshots per company each month. The Team plan includes unlimited screenshots. GPS, geofencing, and employee location sit outside the first release and are planned for a 2027 release.

The practical product test is a fixed-fee exception: add approved client work after kickoff, trace who reviewed the time, export the final record, and confirm that the agency can explain the commercial decision without turning activity into an accusation.

If that is the review your agency wants to run, join the Founding list.

Frequently asked questions

What is a good realization rate for an agency?

There is no universal target that fits every service mix and commercial model. Build a starting band from the agency's own closed, complete engagements, segment it by fixed fee, retainer, and time-and-materials work, then document who can change the definition.

Can agency realization rate be above 100%?

Yes. A result above 100% can reflect efficient delivery, a favorable scope outcome, or an internal standard rate that is too low. Check time completeness, scope version, and rate-card version before celebrating it.

What is the difference between utilization and realization rate?

Utilization compares client work with available capacity, so it supports staffing and allocation decisions. Realization compares commercial value retained with the standard value delivered, so it supports scope, estimate, discount, and pricing decisions.

Should freelancer and contractor hours count in realization?

Include approved contractor delivery when it contributes to the scope being measured, and apply a documented standard-value rule to it. Keep contractor cost in the margin view and keep reimbursable pass-through expenses separate unless the agency's policy deliberately includes them.

How often should an agency calculate realization rate?

Calculate a provisional rate weekly when there is still time to correct scope, data, or delivery. Restate it at the monthly close after approvals, change orders, revenue treatment, and direct costs are complete.