In a market increasingly dependent on software, IT demand prioritization is what Rolando Bonaccorsi, a leader in AI and data science applied to business and operations, identifies as the source of much of the friction between technology and business teams. Without explicit, mutually agreed-upon criteria, the order of the queue begins to reflect who pushes the hardest rather than what delivers the greatest return for the company.
This is a familiar problem in organizations of all sizes. Each department believes its own requests should take priority over all others, the technology manager tries to please everyone, and the result is a collection of half-finished projects, none of which is completed by the originally agreed-upon deadline. The overall perception is that the team is slow, when the real problem is the dispersion of effort.
How Does the Dispute Begin?
A list of requests without visible rules creates a decision-making vacuum that someone will always fill, with or without formal authorization. Usually, it is the department with greater access to company leadership or the manager who is most persistent in following up. As a result, the company’s technical capacity is allocated based on internal influence rather than the value delivered to the company’s end customers.
The friction increases because no one can clearly explain the rationale behind the chosen order of requests. Those whose demands are postponed interpret the delay as a sign of disregard, while the technology department spends much of its time defending decisions instead of carrying out the work. In Rolando Bonaccorsi’s view, the political cost of this cycle often exceeds the technical cost of the requests themselves because it erodes trust between departments.
Criteria That Support Decision-Making
Simple models work better than elaborate scoring formulas. Combining business impact, estimated effort, the risk of doing nothing, and the window of opportunity already makes it possible to compare very different requests, provided that the criteria’s weights are agreed upon with the departments in advance and applied consistently in every case, without undisclosed exceptions.
The difficult part is not choosing the model but maintaining it when someone pushes for an exception. According to Rolando Bonaccorsi, consistency matters more than the model’s sophistication because an exception granted once becomes a precedent, and the criteria lose their authority when resolving the next conflict between two equally legitimate requests.
Declared Urgency Versus Actual Urgency
Virtually every request arrives labeled as urgent, completely draining the word of its meaning within the company. Distinguishing urgency from importance requires asking, in each case, what would happen if the request were postponed until the following month. This question reshapes the conversation because it forces the requester to describe a concrete, verifiable consequence, along with a deadline and estimated impact.
Many requests do not withstand this question and turn out to have much more flexible deadlines than initially declared. The few requests that genuinely cannot wait then receive the appropriate attention, as the queue is no longer clogged with artificial emergencies created in haste. As Rolando Bonaccorsi explains, this filter protects genuine priorities.
Queue Transparency Solves Half the Problem
Publishing the queue along with its criteria, position, and justification completely changes the tone of the relationship between technology and business departments. Those who can see the entire picture understand why their own request ranks fifth and can negotiate based on sound arguments rather than complain about a lack of attention from a team that never stops working. A visible queue also reduces informal pressure applied outside the agreed-upon process.
Periodic reviews with representatives from each department complete the arrangement and distribute responsibility for prioritization among everyone involved. Decisions made collectively reduce the individual pressure placed on the technology team and increase everyone’s commitment to the agreed-upon deliverables. Rolando Bonaccorsi points to a lightweight prioritization committee as a viable alternative to handling requests based on who insists the most.
