Norway's QA regime: why the timing of a review matters as much as the review itself

Norway's independent QA regime shows that when a review happens determines whether it can still change anything: before the decision it works, after it, it only documents.

A review after the decision has already been made

In a steering committee meeting, the independent review is on the agenda, neatly sandwiched between other items. Except the budget was fixed two agenda items earlier. The reviewer can still raise points, but everyone at the table knows the decision will not change. Anyone who has sat through this once recognises it immediately: the review exists, but at the wrong moment.

That exact difference — before or after the decision — is why Norway's QA regime shows a result worth noticing.

What Norway does differently

For large government projects, Norway requires an independent quality assurance review, carried out before parliament makes the investment decision. Of the first forty projects that went through this regime, around 80 percent came in within budget (Samset & Volden, 2016). That stands out against the broader picture: research into 2,062 capital projects shows costs running, on average, 1.39 to 1.43 times the estimate (Flyvbjerg, 2021). It is correlational evidence, not a controlled experiment — but it is an indication that the timing of a review matters as much as the review itself.

Why timing makes the difference

A review that comes after the decision can only record what is already fixed. There is no longer room to revise the budget, narrow the scope or, in the extreme case, not start at all. A review carried out before the decision can actually still change it, because the options are still open.

You see that same pattern in ordinary practice outside Norway. A common pattern in municipal IT projects: the executive board sets the budget during the autumn budget round, and the substantive review of the plan of approach only follows in spring, once the tender is already running. The review still produces useful signals — but the only decision still available by then is to continue or stop. Adjusting scope or approach is already more expensive than it would have been at the start.

What this asks of a plan of approach

The point is not that a late review is pointless. The point is that the weight of a review depends on what can still be done with its outcome. Three things determine that in practice:

The timing of the review is fixed in the plan in advance, not scheduled whenever someone has time for it.
The review is finished before the body that sets the budget actually does so — not at the same time, not afterwards.
The outcome of the review can still change the decision. If "reject" is not a realistic outcome, it is no longer a review but a formality.

That last point is the litmus test: ask yourself what happens if the review comes out negative. Can the decision still be reversed? If not, the review sits at the wrong point in the process, however thorough it was.

In practice

Fix a concrete date in the plan of approach by which the review must be completed, and set that date before — not after — the moment the budget or mandate is formally confirmed.