If you have ever sat through a prioritisation meeting that felt like a tug of war, you already know why so many roadmaps drift. Everyone comes armed with a favourite initiative and a great story. The loudest voice or the most senior title often wins. Weeks later, teams discover there was no shared logic behind the decision, just a moment of momentum. That is how portfolios fill up with work that is hard to defend and even harder to deliver.
Prioritisation fails for a few predictable reasons. First, strategy is rarely translated into a short list of value drivers that people can rally around. Without those anchors, anything can sound important in isolation. Second, confidence in delivery is not treated as a decision input. A use case may be high value on paper, yet the data is not ready, the sponsor is lukewarm, or the operating model cannot carry it. Third, effort is underestimated because the scoping deck sees only the happy path. Multiply those three mistakes and you get a queue that looks impressive and moves slowly.
There is a simple way to remove drama and replace it with discipline. Score every candidate by Impact, Confidence, and Effort. Do it in the open. Do it the same way every time. Impact asks the only question that matters to the board. If this lands, how does it move revenue, cost, risk, or customer and employee experience. Confidence tests whether you can deliver what you are promising. Do we have the data? Are the definitions clear? Are users and sponsors ready? Have we done something like this before? Effort forces a real conversation about complexity and capacity. What will it take to stand this up and keep it alive.
When ICE scoring is used properly, it changes the dynamic in the room. Instead of debating preferences, you debate evidence. Instead of selling ideas, teams bring assumptions and are willing to update them. The grid gives you a rational way to rank opportunities and, just as important, to say no to work that will not create material value this quarter or next. It also helps you find the sweet spot for quick wins that build confidence without derailing the transformation.
Across a large portfolio, ICE scales because it is transparent and teachable. Product managers can run it in discovery. Data leaders can use it to triage demand. Executives can use it to test whether the proposed sequence really matches the strategy. You get a living queue that moves as new information arrives. If a use case slips because a dependency is late or the data is dirtier than expected, the next highest score steps up and delivery keeps flowing.
The magic is not the acronyms. It is the discipline around them. Tie Impact to your value levers and insist on numbers. Tie Confidence to real readiness checks like data availability, privacy posture, stakeholder commitment, and the ability of the business to absorb the change. Tie Effort to constrained capacity rather than ideal timelines. If you add a simple stage gate that asks for a one-page business case and a lightweight investment view, you turn a scoring exercise into a funding decision you can defend.
There will be pushback at first. Some people will feel like their ideas are being reduced to a score. What they will learn is that ICE gives them a fair path to yes. If they tighten the problem statement, improve the data readiness, secure a sponsor, and right size the scope, their score rises and their turn will come. Over a few cycles, the culture shifts. Teams stop competing for attention and start competing to improve their assumptions.
Once you use ICE across dozens or hundreds of use cases, patterns emerge. You will see which types of initiatives always score high Impact but low Confidence because of recurring data issues. Fix those issues and whole categories of value unlock. You will discover which sponsors consistently raise Confidence because their teams execute. Back them with bigger bets. You will notice where Effort is frequently misjudged and refine estimates for the next wave. The portfolio gets smarter because the system learns.
Start by translating your strategy into clear value levers and make them the backbone of Impact scoring. Introduce ICE scoring as the default way to rank all incoming ideas and run a short readiness check to make Confidence real. Add a simple business case and investment view so funded work has a measurable outcome and a baseline to track. Use performance attribution and ROI validation to report realised value against the forecast and feed that learning back into the next scoring round. With a month of practice you will feel the noise drop, the queue stabilise, and the outcomes improve.