My favourite KPI story is about the late Lord King, who set about turning British Airways (BA) around in the 1980s by reportedly concentrating on one KPI. Lord King appointed some consultants to investigate and identify the key measures he should concentrate on to effect a turnaround in the ailing airline. They came back and told Lord King that he needed to focus on one critical success factor (CSF): the timely arrival and departure of aeroplanes.

Finding the CSFs and narrowing them down to no more than five to eight is a vital step in any KPI exercise, yet one seldom performed! Lord King, however, was reportedly not impressed since everybody in the industry knows the importance of timely planes. However, the consultants then pointed out that this is where the KPIs lay and they proposed that Lord King focus on late plane measures.

As a result, he was notified, wherever he was in the world, if a BA plane was delayed over a certain time - say, two hours. The BA airport managers at the relevant airports therefore knew that if a plane was delayed beyond a certain 'threshold', they would receive a personal call from Lord King. Predictably, it was not long before BA planes had a reputation for leaving on time.

The 'late plane' KPI was linked to most of the CSFs for the airline. It linked to the 'delivery in full and on time' CSF - namely, the 'timely arrival and departure of aeroplanes', it linked to the 'increase repeat business' CSF, and so on.

The importance of the 'timely arrival and departure of aeroplanes' CSF can be seen by its impact on all the six perspectives of a modified balanced scorecard (BSC) - I have added employee satisfaction and environment/community to the traditional four perspectives.

Late planes impacted all six balanced scorecard perspectives, because they:

• increased cost in many ways: including additional airport surcharges, and the cost of accommodating passengers overnight as a result of late planes being 'curfewed' due to noise restrictions late at night (financial perspective);
• meant unhappy customers, and alienated those people affected by the late arrival of the passengers - ie possible future customers (customer satisfaction perspective);
• created a negative impact in the wider community and thus reduced the potential pool of future employees (community perspective);
• incurred wastage of food, since hot food has a short serving window, and wastage of fuel as planes endeavoured to make up for lost time and operated outside their most economical flight speed (environmental perspective);
• had a negative impact on staff development as staff would repeat the bad habits that had created late planes (learning and growth perspective);
• adversely affected supplier relationships and servicing schedules resulting in poor service quality (internal process perspective); and
• led to employee dissatisfaction as they had to deal both with frustrated customers and the extra stress each late plane created (employee satisfaction perspective).

A chief executive officer (CEO) of a distribution company realised that a critical success factor for their business was trucks leaving as close as to capacity as possible. Large train trucks capable of carrying more than 40 tonnes were being sent out with small loads as despatch managers were focusing on 'deliver in full on time' to customers.

Each day by 9am, the CEO received a report of those trailers that had been sent out underweight. The CEO rang the despatch manager and asked whether any action had taken place to see if the customer could have accepted that delivery on a different date that enable better utilisation of the trucks. In most cases the customer could have received it earlier or later, fitting in with a past or future truck going in that direction.

Just as with the airline example, staff did their utmost to avoid a difficult phone call with their CEO. And the resultant sending out of more trucks at, or close to, capacity had a significant impact on profitability.

KPIs represent a set of measures focusing on those aspects of organisational performance that are the most critical for the current and future success of an organisation. Crucially, there are only a few KPIs in an organisation (no more than 10) and they have certain characteristics.

These include:

• that they are measured frequently eg daily or 24/7 (KPIs are not measured monthly);
• that they are non-financial measures (not expressed in $s, £s etc);
• that they are acted upon by the CEO and the senior management team on a daily or 24/7 basis;
• that all staff understand the measure and what corrective action is required;
• that responsibility can be tied down to the individual or team;
• that the KPI has a significant impact on the organisation eg it impacts on most of the critical success factors and balanced scorecard perspectives; and
• that positive movement affects all other performance measures in a positive way.

KPIs should be monitored and reported 24/7, daily, or in a few cases, perhaps weekly: to measure a KPI monthly is to shut the barn door well after the horse has bolted. KPIs are therefore 'current' or future measures as opposed to past ones. When you look at most organisational measures, they are very much past indicators measuring events of the last month or quarter. These indicators cannot be and never were KPIs. That is why a satisfaction percentage (eg 65%) from a customer satisfaction survey performed every six months can never be a KPI.

When you put a pound or dollar sign to a measure you have not dug deep enough. Sales made yesterday will be a result of sales calls made previously to existing and prospective customers, advertising, amount of contact with the key customers, product reliability etc. I term any sales indicators expressed in monetary terms as result indicators which will be further explained in this article. In many organisations a KPI may rest with certain activities undertaken with your key customers who often generate most, if not all, of your profit.

All good KPIs that I have come across - those that have made a difference - had the CEO's constant attention, with daily calls to the relevant staff. Having a potentially career-limiting discussion with the CEO is not something staff want to repeat, and in the above mentioned airline's case, innovative and productive processes were put in place to prevent a recurrence.

A KPI should tell you about what action needs to take place. The BA 'late plane' KPI communicated immediately to everybody that there needed to be a focus on recovering the lost time. Cleaners, caterers, ground crew, flight attendants, and liaison officers with traffic controllers would all work some magic to save a minute here and a minute there, whilst maintaining or improving service standards.

A KPI is deeply enough embedded within an organisation to be tied down to an individual. In other words, the CEO can ring someone and ask "why?" Return on capital employed has never been a KPI as it is a result of many activities under different managers. Can you imagine the reaction if a general manager was told one morning by the CEO "Pat, I want you to increase the return on capital employed today"?

A KPI will affect most of the critical success factors and more then one balanced scorecard perspective. In other words, when the CEO focuses on the KPI, and the staff follow, the organisation scores goals in all directions.

A KPI has a flow-on effect on other performance measures. Reducing late planes would improve performance measures around improved service by ground staff as there is less 'fire fighting' to distract them from a quality and caring customer contact.

From the research I have performed, from workshop feedback across diverse industries and as a by-product of writing a 'Key performance indicators manual' (second edition), I have come to the conclusion that there are three types of performance measure:

• key result indicators (KRIs) - which give an overview on performance and are ideal for the board as they communicate how management has done in a critical success factor or balanced scorecard perspective;
• performance indicators (PIs) - which tell staff and management what to do; and
• key performance indicators (KPIs) - which tell staff and management what to do to increase performance dramatically.

I use an onion analogy to describe the relationship of these three measures.

The outside skin describes the overall condition of the onion, how much sun, water and nutrients it has received, how it has been handled from harvest to supermarket shelf. The outside skin is thus a key result indicator. The layers represent the various performance indicators and the core is where you find the key performance indicators.

Kaplan and Norton recommend no more than 20 KPIs, and Jeremy Hope (of 'Beyond budgeting' fame) suggests fewer than 10. To aid those involved in performance measurement I have developed the 10/80/10 rule. This means an organisation should have about 10 KRIs, up to 80 PIs and 10 KPIs: there is very seldom a need for more measures, and in many cases fewer can be used.

The common characteristic of KRIs is that they are the result of many actions. They give a clear picture of whether you are travelling in the right direction, and of the progress made towards achieving desired outcomes and strategies. They do not, however, tell management and staff what they need to do to achieve desired outcomes. Only PIs and KPIs can do this.
KRIs that have often been mistaken for KPIs include:

• customer satisfaction;
• net profit before tax;
• profitability of customers;
• employee satisfaction; and
• return on capital employed.

A car's speedometer provides a useful analogy. The board will simply want to know the speed at which the car (the organisation) is travelling. Still using this analogy, management needs to know more information since the car's speed is a combination of what gear the car is in and what revs the engine is doing.

In fact, management might be concentrating on something completely different, such as how economically they are driving eg a gauge telling them how many kilometres they are getting per litre, or how hot the engine is running. These are two completely different performance indicators.

Separating out KRIs from other measures has a profound impact on the way corporate accountants report performance. There is now a separation of performance measures into those impacting governance (up to 10 KRIs in a dashboard) and those impacting management.

The 80 or so performance measures that lie between the KRIs and the KPIs are the performance indicators (PIs). The performance indicators, while important, are not 'key to the business'. PIs help teams to align themselves with their organisation's strategy. PIs complement the KPIs and are shown with them on the organisation's, divisions', departments' and teams' scorecards.

PIs could include:

• profitability of the top 10% of customers;
• net profit on key product lines;
• % increase in sales to the top 10% of customers;
• % of employees participating in the suggestion scheme; and
• duration of the cash to cash cycle (eg 65 days).

Many management books talk about 'lead' and 'lag' indicators which I believe merely clouds the KPI debate. Using this new way of looking at KPIs we dispense with the terms lag (outcome) and lead (performance driver) indicators. I have presented to nearly 2,000 people on KPIs and I always ask "is the 'late planes in the air' KPI, a lead or lag indicator?" The vote count is always evenly split. Surely, this is enough proof that lead and lag labels are not a useful way of defining measures?

KRIs replace outcome measures, which typically look at activity over months or quarters. PIs and KPIs are now characterised as either past, current or future measures. The new concept called 'current measures' are those monitored 24/7 or daily. You will find the real KPIs in your organisation are either current or future measures.

Past measures

last week/fortnight/month/quarter - eg number of late planes last week/last month

Current measures

(24/7 and daily)  - eg planes over two hours late(updated continuously)

Future measures

(next day/week/month/quarter) - eg number of initiatives to be commenced in the next month/two months to target areas which are causing late planes

The lead/lag division did not focus adequately enough on the timing of the measures. Most organisations that want to create alignment and change behaviour need to be monitoring what corrective action is to take place in the future.

In other words if quality improvements are to happen we need to measure the number of initiatives which are about to come online in the next week, fortnight, month. If we want to increase sales what is important to know is what the number of meetings which have already been organised/scheduled with our key customers in the next week, fortnight, month is.

You should consider the following:

• deliver a PowerPoint presentation to the senior management team to get buy-in for your KPI/BSC project;
• link with an external expert who can contribute to brainstorming sessions designed to ascertain the CSFs for your organisation.

Since David Parmenter wrote this article in 2006, he has published further articles on the topic and an eBook, 'Key performance indicators: developing, implementing and using winning KPIs'.

These are available below to logged-in ICAEW members and ACA students.

Online articles

The Library provides access to leading business, finance and management journals. These journals are available to logged-in ICAEW members, ACA students and other entitled users subject to suppliers' terms of use.