A consulting recommendation is only as useful as your ability to tell, months later, whether it worked. Here is how to choose a small number of KPIs that actually inform a decision, with examples drawn from audit, strategy, and market entry engagements.
Why so many KPIs end up ignored
Most companies do not lack metrics; they lack metrics anyone actually looks at. A dashboard with thirty tracked numbers usually gets less attention than one with four, because nobody can hold thirty numbers in mind while making a decision. The purpose of a KPI is to make a specific decision easier — whether to continue, adjust, or stop an initiative — and metrics that do not connect clearly to a decision tend to fade into background noise within a few months.
This matters directly for consulting engagements. A strategic recommendation or process change is only as good as the ability to tell, a few months later, whether it worked. Without a small number of well-chosen KPIs agreed in advance, that judgement tends to default to whoever argues most persuasively in a meeting, rather than to evidence.
What makes a KPI actually useful
A useful KPI has a few consistent characteristics, regardless of the department it belongs to:
- It is tied to a specific decision. Before tracking a number, it should be clear what you would do differently depending on whether it goes up or down.
- It can be measured consistently. A metric that changes definition every quarter cannot show a trend.
- It is hard to game in a way that defeats its purpose. A support-response-time KPI that encourages rushed, unhelpful replies is worse than no KPI at all.
- Someone is accountable for it. A KPI with no clear owner tends to be the first one that stops being updated.
KPIs by type of engagement
The right KPIs depend heavily on what kind of change is being measured. A few common examples from our own consulting work:
Process audit follow-up
- Cycle time for the specific process that was audited, measured the same way before and after changes.
- Error or rework rate for the process, where relevant.
- Number of manual handoffs required to complete the process end to end.
Strategic restructuring
- Cost as a share of revenue for the function being restructured.
- Time to decision for the type of decision the restructuring was meant to speed up.
- Employee turnover in the affected team, as a signal of whether the change is sustainable.
Market entry
- Customer acquisition cost in the new market compared with the entry plan's assumptions.
- Time from first contact to first sale, as a proxy for how well the model fits local buying habits.
- Local revenue against the milestones set out in the entry plan.
How to set KPIs for a new initiative
Setting KPIs works better as a short, deliberate exercise than as an afterthought once an initiative is already underway.
- Write down the specific decision the initiative is meant to support or the problem it is meant to solve.
- List two or three numbers that would change if the initiative worked, and that you could realistically measure with existing or easily added tools.
- Set a baseline measurement before the change begins, even if it is imperfect.
- Assign one named owner for tracking and reporting each metric.
- Agree in advance how often the metric will be reviewed and by whom.
Reviewing and adjusting KPIs over time
KPIs set at the start of an engagement are not necessarily the right ones six months later, especially once the initial issue is resolved and a different constraint becomes the priority. Building in a scheduled review — quarterly is common — keeps the metrics relevant rather than becoming a legacy dashboard nobody questions. It is also worth explicitly retiring KPIs that have stopped informing any decision, rather than letting the list grow indefinitely.
Frequently asked questions
How many KPIs should a single initiative have?
Two or three well-chosen KPIs are usually more useful than a longer list. If a metric is not clearly tied to a decision, it is a candidate to drop.
Should KPIs be set before or after a consulting engagement begins?
Ideally before, as part of scoping. Agreeing on how success will be measured before work starts avoids disagreement about results later.
What if we do not have good baseline data?
Start measuring from whatever point you can, even if the baseline is imperfect. A rough starting point is more useful than waiting for perfect data before tracking anything.