The traceability test:
five decisions, one question

The purpose of this piece is to describe a simple test of enterprise traceability, which is the ability to follow strategic intent from the top of an organisation down to the activity that delivers it and back again, and to show how an executive can apply the test to five recent decisions without commissioning any additional analysis. I describe it as a test and not as an audit because it is intended to be run in an afternoon by the person accountable, and its value lies mostly in what the exercise reveals about how decisions are actually made.

It is reasonable to ask whether traceability matters when an organisation is performing adequately. Many organisations run well for years on judgement, relationships and local initiative, and it would be a mistake to suggest that every decision needs a documented line back to a strategic objective, since some of the best decisions are opportunistic or relational and a requirement to justify each one would slow an organisation more than it helped. This piece does not contend otherwise. It contends that the decisions an organisation makes on purpose, particularly the large and expensive ones, should be traceable, and that when they are not, the organisation is spending against a strategy it cannot see.

Where the idea comes from

Enterprise Traceability is one of three principles I draw from systems engineering, and it derives from requirements traceability, the discipline of linking each requirement of a system forward to the design and tests that satisfy it, and backward to the need that justified it. In a technical program that discipline exists because, without it, nobody can say whether a component is there for a reason or whether a need has been left unmet. Applied to an organisation, it asks the same of strategic intent: can an activity, an investment or a decision be explained by reference to that intent, and can the intent be followed forward to what actually carries it out? I describe the method here and make no claim to an engineering title, since it is simply a useful discipline to borrow, and it is set out alongside the other two principles in The Living Organisation.

The test

Choose five decisions made in the last twelve months. They should differ in kind, because traceability tends to break in different places depending on the type of decision, and I would suggest the following.

For each decision, ask the same four questions, and try to answer them without asking the decision-maker, because the point is to find out what the organisation can reconstruct and not what one individual remembers.

Score each decision as traced, partly traced or untraced, and note the question at which the line broke. Five decisions and four questions give twenty answers, which is enough to see a pattern and few enough to finish.

Reading the result

Where the line breaks tends to say more than how often it breaks, and mapping the break to the eight capability domains gives a first indication of where to look.

If four or five of the decisions trace cleanly, the organisation has probably built a reliable path between intent and action, at least for the decisions that matter. If two or fewer do, I would read that, as far as I can judge from outside, as a sign that strategy and activity are being run as separate things and that a good part of the investment is being justified after the event.

An external example

The pattern is not confined to private organisations. The Queensland Audit Office's report on evaluating major infrastructure projects (Report 14: 2019–20), in its examination of five business cases, found that for two of the projects a government announcement of a preferred project had come before the analysis, which, as I read the report, left the business case with little strategic value, and that endorsements had proceeded on benefits that could not be monetised. I draw only a limited lesson from this, since the report concerns one assessment body and its framework and public investment differs from private investment in several ways. It does illustrate how a decision can be well documented and still not traceable to a stated intent, because the sequence of events was the reverse of the one the documents imply.

Proportion, and the objection about bureaucracy

The most likely objection is that traceability becomes bureaucracy, with every decision requiring a form and every form requiring a reviewer. That risk is real, and I have seen programs where it materialised. The response I would offer is proportion: the test applies to material decisions, five of them and not five hundred, and its purpose is to show executives where their own line of sight has broken, and not to create a new control. A second view, which I think is also fair, is that some decisions are deliberately untraceable because they respond to something that cannot be written down, such as a relationship or an emerging opportunity. In that case the useful outcome is that the untraced decision is a conscious choice that someone can explain, and not an accident of the way the organisation operates.

The Enterprise System Health Check approaches the same question from a different direction. It asks six functional leads the same anchor questions on Data and Information Flow and reads the spread between their answers as the traceability finding, on the reasoning that an organisation whose leaders describe the same thing differently is unlikely to be able to trace a decision through all of them.

If five recent decisions cannot be followed from the strategy that was meant to justify them, and the resulting activity cannot be followed back to that strategy, is the organisation spending against its intent in the way the executive believes, or is it, arguably, being steered by something that nobody has yet named?