Prioritizing as a Team: Eisenhower, ICE, and Cost of Delay Compared

Why does everything feel equally important in a team?

Many companies have more good initiatives than time: the new quote form, the switch to a CRM, the overdue price list, the website. Each has an advocate, and each is reasonable on its own. Without a shared rule, whatever is pressing or whoever carries the most weight prevails. In the end, five initiatives are started and none is finished.

There are many methods for prioritizing. Three are particularly widespread: the Eisenhower matrix, ICE scoring, and cost of delay. This leads to the guiding question of this article: Which prioritization method fits which decision in a team, and how can you tell that you are using it wrongly?

To answer it, we first clarify the terms. Then we present the three methods, with cost of delay explained in more detail as the thinking model. This is followed by the research findings, a comparison with two perspectives, and a recommendation on how the three methods work together.

Whoever starts everything at once pays, for each initiative, the waiting time of all the others.

What does prioritizing mean?

Prioritizing means setting an order in which limited time, limited money, or limited people are allocated to tasks. It always includes deciding what will be done later or not at all. A list on which everything is priority one is not prioritization but a collection.

Three terms are often mixed up. A task is important if much depends on its result. It is urgent if it can only be done within a short time. Value is the benefit the result brings, such as additional contribution margin, hours saved, or a risk avoided. The three methods differ mainly in which of these terms they put at the center.

The Eisenhower matrix: important and urgent

The matrix is named after former US President Dwight D. Eisenhower, to whom the distinction between urgent and important problems is attributed. It became known mainly through Stephen Covey (1989). It divides tasks into four quadrants: important and urgent, do now; important and not urgent, schedule; urgent and less important, delegate; neither, drop.

Its greatest benefit is making a familiar pattern visible. Zhu, Yang, and Hsee (2018) showed in five experiments that people prefer tasks that seem urgent even when another task objectively pays more and all other reasons are equal. They call this the mere urgency effect. Mills, Cubitt, and Starmer (2026) confirmed the effect in an independent replication with 2,374 participants. What this means for an owner’s calendar is described in the essay on owner time management.

The limits are just as clear. The matrix knows only two levels per axis and no numbers. It says that two initiatives are important and not urgent, but not which comes first. It is therefore not sufficient for comparing larger initiatives that take weeks and compete for the same people.

ICE scoring: impact, confidence, ease

ICE scoring comes from growth marketing and is attributed to Sean Ellis (Ellis and Brown 2017). Each idea receives three scores from 1 to 10: impact, the expected effect; confidence, the certainty that this effect will occur; and ease, how easy implementation is. The three values are averaged or multiplied, and the idea with the highest value comes first.

The advantage is speed. A list of 30 ideas can be scored in half an hour, and the confidence factor forces a distinction between proven and hoped-for effects. The disadvantage is the arbitrariness of the points. What distinguishes an eight from a six is often unknown, and whoever likes an idea gives it higher scores. For many small experiments, for example in marketing or in workflow improvements, that is acceptable. For an investment that ties up a quarter, it is too little.

The thinking model: cost of delay

In his book on flow in product development (2009), Donald Reinertsen put a question at the center that the other two methods do not ask: What does it cost to have an initiative finished one week later? This cost of delay includes lost contribution margin, extra work that keeps occurring, risks that persist longer, and opportunities that close.

To set the order, Reinertsen divides the cost of delay by the duration of the initiative. In agile settings, the result is called Weighted Shortest Job First, or WSJF: the initiative with the highest value per week of work first. The principle is simple. A short initiative with medium cost of delay beats a long one with high cost because it is finished earlier and then stops costing money.

An analogy makes this tangible. At the supermarket checkout, you let the customer with two items go ahead when your own cart is full. You wait one minute longer, he saves ten. Total waiting time falls even though nobody becomes faster.

In the example, the CRM switch has the highest cost of delay per week, but it comes only third because it takes six weeks. The price list is finished after one week and costs nothing from then on. One limit is important: the figures are estimates. What matters is not the exact amount but the order of magnitude and the fact that everyone involved must disclose their assumptions.

What research shows about scoring methods

There is little robust research on simple prioritization methods in small companies. A recent study, however, shows where the weakness of all scoring methods lies. Manzur, Bazán, Fernandez, and a further author (2025) examined WSJF in its common form, in which cost of delay is composed of three scores for business value, time criticality, and risk reduction and divided by the size of the initiative. Their objection: the method treats the three values as independent, although they are not. In their analysis, risk reduction, for example, influenced how business value was interpreted.

In a case study, the authors compared the order produced by conventional WSJF with the actual effect of the initiatives in hindsight. The relationship was weak, with a correlation of 0.287. A method that models the dependencies between the values reached 0.711. The limits are clear: a single case, a computationally demanding method, and authors testing their own model. The message for a small company is still useful. Scores that are simply added up feign a precision they do not have. The reasoning behind the numbers matters more than the number.

The three methods compared: two perspectives

There are two perspectives on this comparison. The first says: simple rules are enough. A small team knows its initiatives, and every calculation only creates false precision, as the finding by Manzur and colleagues (2025) shows. The second says: precisely because estimates are uncertain, they must be written down. Whoever puts a figure on the cost of delay must say where it comes from, which makes it open to challenge and improvement.

I consider the second view the more important one for large initiatives and the first the right one in daily work. The value of the calculation lies not in its precision but in turning a discussion about opinions into a discussion about assumptions. For the question of what gets done first on Tuesday morning, the same calculation would be bureaucracy.

How the three methods work together in a company

1. Once a quarter: calculate large initiatives

   For the three to six largest initiatives, estimate cost of delay per week and duration, divide, and set the order. At most two initiatives run at the same time.

2. Once a month: score the idea list

   Score small improvements and marketing ideas with ICE. The top three go into the next month; the rest stay on the list.

3. Every week: Eisenhower for your own plan

   In weekly planning, each person checks whether their three weekly results lie in the important quadrant or only seem urgent.

The first step has the greatest effect because it limits the number of initiatives running at the same time. Why too much parallel work slows everything down is described in the comparison of OKR, Kanban, and weekly planning. How weekly results are set in 30 minutes is covered in the guide to weekly planning as a team. And whether there is time at all for the chosen initiatives is shown by capacity planning. An initiative that is at the top of the list but gets no hours is not prioritized, only praised.

What remains and how to start

We set out to ask
prioritization method fits which decision. The answer follows from the question each method asks. The Eisenhower matrix separates important from urgent and protects your week from the pull of urgency. ICE scoring sorts many small ideas quickly. Cost of delay shows, for a few large initiatives, which order leaves the least money on the table.

As shown, all three rely on estimates, and research warns against mistaking added-up scores for precision. This does not lead to abandoning methods but to using them deliberately: simply in daily work, calculated for the big decisions, and always with the willingness not to do something. The sober observation is: the best prioritization is the one after which an initiative is stopped.

For this week, one step is enough: write down all running initiatives, estimate the cost of delay per week and the remaining duration for each, and stop the initiative with the lowest value until two others are finished.