Cost of Delay: Prioritising Initiatives by the Cost of Waiting

Cost of Delay is the economic value your company loses each week that an initiative remains undelivered. To sequence initiatives with it, divide the cost of waiting by the duration of delivery and start first with whatever prevents the largest loss per week of work. The method replaces priorities set by volume with a calculation that executives and business units can examine together.

Most portfolios are sorted by impression, seniority or the date of the last escalation call. Cost of Delay gives the sequence an economic basis. This article explains the concept as developed by Donald Reinertsen, shows a calculation with sample figures, classifies the four typical delay profiles and describes how a company can introduce the method within one quarter.

What is Cost of Delay?

Cost of Delay quantifies how much value per unit of time is lost when a product, a feature or a change goes live later than it could. The term comes from product development and was popularised by Donald Reinertsen and his 2009 book “The Principles of Product Development Flow”. Reinertsen argues that companies usually know the cost and the benefit of an initiative but treat time as a quantity without a price. Yet the value of an initiative changes with its delivery date: a new offer that launches three months later loses three months of revenue, and a process improvement loses three months of savings.

The value has four sources, each of which can be expressed per week. The first is additional revenue or margin. The second is avoided cost, such as the removal of manual work. The third is avoided loss, for instance through the expiry of a legacy system or an impending contractual penalty. The fourth is learning value: whoever ships earlier receives customer data earlier and can base follow-on decisions on facts. The fourth item is hard to quantify and is therefore usually estimated as a surcharge on the others.

How do you calculate Cost of Delay?

You calculate Cost of Delay by estimating the expected value of the initiative per week after release and multiplying it by the number of weeks by which the start slips. The estimate is made jointly by the business unit, finance and the delivery team, so that those who answer for the value test the assumptions.

A four-step calculation is enough for the first round. In the first step the business unit names the metric the initiative changes, for example revenue per month, handling cost per transaction or downtime cost. In the second step the amount per week that the initiative generates after release is estimated. In the third step finance and the business unit check together whether the assumptions are plausible. In the fourth step the team records how the cost of waiting behaves over time, in other words which of the four delay profiles applies.

An exact figure is not required. Reinertsen argues that economic decisions tolerate rough estimates, because initiatives often differ in their Cost of Delay by a large multiple. If one initiative costs 50,000 euros per week and another 2,000 euros, the order stays the same even with a 30 per cent error.

How do you build a sequence with CD3?

The sequence emerges when you divide Cost of Delay by the duration of delivery and start the initiatives with the highest quotient first. This measure is called CD3, short for “Cost of Delay Divided by Duration”. It answers the question of which initiative prevents the largest loss per week of work. With equal duration the Cost of Delay decides, and with equal Cost of Delay the shorter initiative goes first.

The example shows the benefit of the method. The legacy system has the highest Cost of Delay at 30,000 euros and would start first in a ranking by value alone. Because it ties up ten weeks, it delays the other two initiatives by ten weeks. If the company begins with invoice checking, which is finished in one week, and moves the ordering flow ahead of the legacy system, value arises earlier. Adding up the cost of waiting until each initiative is finished gives 560,000 euros in CD3 order (10,000 times 1 week, 20,000 times 5 weeks, 30,000 times 15 weeks). If the legacy system goes first, the total is 710,000 euros (30,000 times 10, 10,000 times 11, 20,000 times 15). The CD3 sequence reduces the sum of lost euros by 150,000 without cancelling a single initiative.

The calculation applies to initiatives that one team works through one after another. Where several teams work in parallel, each team applies CD3 to its own queue. Where an initiative waits for a shared resource, Cost of Delay decides who receives it first.

Which four delay profiles exist?

Initiatives differ in how the cost of waiting grows over time. Kanban practice knows four classes of service that can be read as delay profiles, and they simplify the estimate. Joshua Arnold and Özlem Yüce link Cost of Delay to the question of how urgent an initiative is.

With the standard profile the loss accrues evenly per week, as with a new feature that keeps generating revenue. With the fixed-date profile the value is fully available until a deadline and zero afterwards, for example for a trade fair, a seasonal offer or the end of a contract. With the expedite profile a high loss arises from the very first day, for instance with a fault that affects customers directly, and the initiative goes ahead of everything else. With the intangible profile costs are barely visible at first and grow later, as with the upkeep of a technical foundation or the training of employees.

The profiles prevent a common mistake. An initiative with a fixed date gains nothing if it is finished two weeks before the deadline, but loses everything if it is finished a day after. It needs a plan with a buffer, and acceleration at any price does not pay off here. Intangible initiatives drop out of rankings by Cost of Delay easily and then become a risk to delivery capability. Reserve a fixed share of capacity for them.

When does Cost of Delay pay off in a company?

Cost of Delay pays off when more than five initiatives compete for the same teams and the sequence is disputed. In organisations with a few large initiatives, a decision by the executive team is often enough. As soon as business units present their requests at the same time and each side has good reasons, gut decisions fail. The article on clearing out the project portfolio shows how many initiatives a company should run at once before a ranking takes effect at all.

The benefit rises when the method is combined with a limit on parallel work. Whoever starts fewer things at once needs fewer weeks for each, and the cost of waiting per initiative falls with the shorter duration. How this works is described in the article on limiting work in progress. Where time between idea and release is lost is shown in the article on shortening time to market.

How do you introduce Cost of Delay?

Introduction succeeds within a quarter if it begins with a manageable portfolio and fixed dates. Five steps are enough.

1. Define the portfolio: choose an area with five to fifteen initiatives competing for the same teams.
2. Hold an estimation round: the business unit names the metric and the amount per week for each initiative, finance checks the assumptions, the delivery team estimates the duration.
3. Assign a profile: each initiative receives one of the four delay profiles and the deadlines are documented.
4. Fix the sequence: the executive team confirms the CD3 ranking and reserves capacity for intangible work.
5. Review quarterly: assumptions are compared with actual values after release so that the estimates improve.

The estimation round is the most valuable part because it exposes assumptions that otherwise stay unspoken. Whoever cannot justify the value of an initiative in euros per week usually has no clear business result either, and that should be settled before the start.

What are the limits of the method?

Cost of Delay delivers a defensible sequence. The figures are estimates and depend on the assumptions of those involved. Presenting them as measurements costs trust as soon as reality deviates. Treat the results as a reasoned working hypothesis and test them after release.

Dependencies also need attention. An initiative with a low Cost of Delay of its own can be the precondition for another with a high one, and then it belongs at the front. Security and stability work is hard to express in money and therefore receives a fixed share of capacity that does not compete in the ranking.

Cost of Delay with Niehues Consulting

Niehues Consulting is a consultancy for AI management and organisation design from a delivery perspective, based in Düsseldorf. Founder Marin Niehues builds organisations that deliver, in projects ranging from VC-funded scale-ups to DAX 40 corporations and in more than 20 countries on six continents. As delivery manager, project manager or product owner he runs portfolios with clear sequences, owners and metrics.

The approach follows the Niehues Delivery Model with its three phases. For prioritisation by the cost of waiting, three of its checking questions count: Which business result should change measurably? What is the baseline, and how do we measure success? Which metric shows after 90 days that the solution has landed? The answers form the value assumption that every CD3 calculation needs. The entry point is a free initial consultation of 30 minutes, which you can arrange via the contact page.