To improve organisational performance, we first need to measure it. So how long have you got?
"If you can't measure it, you can't improve it." That was the opinion of Peter Drucker, the American management consultant, educator and author once celebrated by BusinessWeek magazine as “the man who invented management”. So he should know.
And that’s all very well, we might say. But what should we be measuring – and how?
Just as it’s important that our people have clear goals and that we measure their progress towards them, the core purpose of measuring organisational performance is to identify where operations are performing well, and – more importantly – where improvements are needed so that we can take the necessary steps to stay on track to achieve organisational goals.
When we do this well, it can have a major impact on success (or otherwise). And the better we understand our objectives and the methods available for measuring performance, the more successful we are likely to be.
In other words, when it comes to making progress, KPIs and OKRs are a lot more useful that IDKs.
The devil is in the data
Individual performance management is a familiar concept for most leaders, but we also need a detailed understanding of how our whole organisation is performing. And that’s where key performance indicators or KPIs come in.
As well as giving us a better understanding of overall performance and helping us adhere to any necessary legislative frameworks, KPIs can prove useful in a number of ways. They can:
- provide employees with a means of linking their own achievements with organisational goals,
- show how performance compares across internal departments and against external competitors,
- improve transparency by providing a way to disclose information to stakeholders for example, via an annual report,
- reveal trends, enabling early detection of problems, challenges and opportunities, and
- inform decision-making about where, when and how to take corrective action.
Much of the data gathering that goes on inside organisations is designed to feed into these KPIs. For example, a salesperson may be required to report the value of sales and the number of calls they are making each day, so that these figures can be collated with those of other team members to create an organisational performance indicator around how well calls translate into sales.
However, as the British economist and author Ronald Coase said: "If you torture the data long enough, it will confess to anything". So, it’s vital we first understand our goals and set the right KPIs to measure our progress towards them. Otherwise, we may be gathering the wrong data from the wrong sources and drawing the wrong conclusions from them. And that really can be painful.
Understanding the performance measurement cycle
Whatever our specific methods, a performance measurement cycle tends to follow a similar cycle from selection to action.
The process starts with selecting and defining our key performance indicators; the most important questions that need to be answered about our organisational performance.
With objectives set, we can then move on to collecting the data consistently across the whole organisation and storing that data in a way that avoids any unnecessary complexity.
Next, by analysing the data we have gathered, we turn raw facts and statistics into understanding that informs our strategic decision-making. Then we share what we have learned by presenting the data in a way that is both meaningful and motivating
From here, we move to interpreting the data to see what it tells us about our successes and challenges, the lessons learned, and what information we still need to know
All of which drives towards us taking action to improve our performance, with clear priorities, targets, and timescales
Without this crucial interpretation and understanding, we risk using our data in the way that the Scottish poet Andrew Lang observed that most people use statistics: ‘like a drunkard uses a lamppost – more for support than illumination’.
So, how do we go about the sober business of setting the right KPIs and learning the right lessons from them?
Selecting and defining our KPIs
Unfortunately, there is no standard blueprint or template for determining KPIs. There are as many measures as there are differences between industry sectors – and between organisations within each sector. They are often associated with sales performance (charting, for example, income growth or new customers acquired) and financial monitoring (profit margins; budget variances), but they can also be used in a variety of other ways, such as measuring progress on projects (% of project completed; key milestones met or missed), the effectiveness of a social media marketing campaign (new likes/follows; click throughs) or to track customer satisfaction or employee engagement.
A good place to start is by considering our own key drivers; those that are fundamental to our overall success. We can then develop KPIs to track these specific areas.
- A pharmaceutical company might focus on the amount they spend on promotional activity as a KPI.
This is an input KPI – one that measures the assets and resources an organisation has invested in or used to generate results. Examples include research and development expenditure and the quality of raw materials. As Jan Leschly, the former pro tennis player and ex-CEO of SmithKline Beecham once said, ‘If you don’t keep the score, you’re just practicing’. So sound KPIs are always part of a successful formula.
- A restaurant might identify the percentage of income that comes from repeat business as a KPI.
This is an output KPI – one that measures the financial and non-financial results of key activities. Examples include revenue, number of new customers acquired and percentage increase in employees. When you think we now have a word for ‘hangry’ home-delivery restaurants might also be wise to look at a KPI for ‘promise times’ with goal-setting based on on-time deliveries.
- A plastic toy manufacturer might focus on production line speed as a KPI.
This is a process KPI – one that measures the efficiency or productivity of an internal process of cycle. Examples include product-repair time, time to deliver an order, or time to fill vacant posts.
What we measure really does matter. For example, the Net Promoter Score (NPS), which asks customers how likely they are to recommend a particular company after interacting with them, peaked for Toys R Us just before they went out of business. Perhaps all this customer appreciation was hiding a deeper process problem.
In all cases, our KPIs should:
- Provide a flow of information which can be reported on in a consistent and regular way.
- Be easily quantifiable so that measures can be compared objectively over a set period, and any necessary improvements identified and prioritised.
They should not:
- Focus solely on financial measures as they only give part of the picture.
In smaller organisations, an annual planning session to collaborate on and craft KPIs may be all that’s needed – and these meetings can involve everyone. In larger organisations, meetings may need to happen more frequently. And while it’s tempting to restrict such meetings to the senior leadership team to avoid a camel (an animal that could only ever have been designed by a committee), it’s wise to include a variety of perspectives.
If our senior leaders think that BLM stands for ‘be like me’ it’s time to cast the net wider.
The Balanced Scorecard
The idea that we need to take a more holistic view of the health of our organisations underlies the idea of the Balanced Scorecard, developed by Robert S. Kaplan and David P. Norton in the 1990s.
The balanced scorecard proposes that leaders measure and review organisational performance from four important perspectives.
It provides answers to four basic questions:
How do customers see us? (customer perspective)
What must we excel at? (internal perspective)
How can we sustain our ability to change and improve? (learning and growth perspective)
How do we look to shareholders (or other stakeholders)? (financial perspective)
While giving leaders information from four different perspectives, the balanced scorecard also tackles the (all too common) problem of having too many KPIs and measures. Instead, it helps us to focus on the handful of measures that are most critical.
It also encourages us to join the dots across different teams and projects and between vision and strategy and on-the-ground operations, helping us to:
- communicate over-arching goals;
- align and adjust operations with strategy;
- measure and monitor progress towards goals in a more streamlined way, and
- prioritise/re-prioritise projects, products and services as necessary.
In practical terms, focusing on the four perspectives might generate KPIs relating to measurables like:
- Customers: market share, customer retention, acquisition, satisfaction and profitability.
- Financial: revenue growth, cost management and use of resources.
- Internal: the quality and efficiency of processes related to the creation and delivery of a product or service.
- Learning and growth: employee capabilities and engagement, motivation and alignment.
Kaplan and Norton believe that the scorecard’s enduring contribution to performance management lies in the concise, holistic picture of the organisation that it provides. It helps leaders to collaborate beyond silos and also offers clarity when it comes to communicating and interpreting performance data – and knowing what action to take for optimal impact.
Characteristics of effective KPIs
Having collaborated to create our KPIs, we should sense-check them against some useful criteria.
Is it relevant?
That is, closely related to organisational strategy. Otherwise, measuring and acting on it may well be pointless.
For example, if we have identified increased market share as an important strategic objective, the number of new leads generated by marketing campaigns and the volume of new business sales are highly relevant KPIs, whereas repeat business is less relevant.
Can we influence it?
A KPI can only contribute to improved performance when we are able to influence its direction and outcome.
For example, the weather affects many tourist organisations, but if we knew how to change it, it’s unlikely we’d still be working in tourism. However, sales growth can be influenced by other things too, like the effectiveness of our operations.
Is it empowering?
KPIs (and particularly KPI targets) must be set appropriately so that they empower employees. Otherwise, they will not promote the behaviours and actions which will lead to improved performance and could instead prove counter-productive.
For example, unclear or unattainable goals are likely to lead to people giving up on them. Alternately, KPIs with insufficient challenge or stretch might cause people to lose interest.
Leading vs lagging indicators
In addition, we should always try to measure performance using leading indicators; predictive measures that show us what to expect in the future and allow us to adjust course accordingly.
For example, if our goal is to reach a certain sales threshold, it’s no use just measuring the number of sales.
This would be a lagging indicator as by the time we found out we were falling short it would be too late to do anything about it. By contrast, if we were to count the number of sales meetings or calls that we are achieving each week this would be a leading indicator as it would help us to predict the number of likely sales based on our experience of previous conversion rates.
And it’s always good to be wise before the fact. Not just after it.
The KPI pyramid
When we create operational plans, the Anthony Triangle shows us how strategic purpose cascades down through an organisation, from strategic planning at the top, through tactical planning in the middle, to operational planning at the bottom. And it’s much the same with KPIs.
Organisational KPIs at the top should be linked to departmental and team KPIs in the middle and to individual KPIs (often known as objectives at this level) at the bottom to promote accountability and engagement.
- The organisational KPIs should be designed to measure the key drivers which will enable the organisation to reach its strategic goals
- The departmental and team KPIs then play a more specific role in supporting these goals, for example in production, customer services or human resources
- The individual KPIs are more specific still, relate to each person’s particular role and talents and form the basis of individual performance management.
Every department should have appropriate targets, measurable by specific deadlines and deliverable outputs. It can be helpful to devise a phased roll-out with key dates for each tier of the pyramid, for example: week 1: set executive level goals and measures, week 2: set department level goals and measures, and so on.
Once again, it’s everyone’s combined efforts that give the pyramid its strength.
Once agreed, KPIs should be regularly reviewed to ensure that they remain valid and useful. This should also happen each time we change or refresh our strategy.
We should also consider external benchmarking where possible. In some sectors, like banking and financial services, KPI data is shared between similar organisations. This can give us a much clearer understanding of the relevant industry average - and how we are measuring up.
Collecting, storing, and analysing performance data
Once KPIs have been identified, the next step is to determine how best to capture KPI information. Key factors to consider include:
Do we have a comprehensive infrastructure that ensures consistency, accuracy and integrity standards are maintained across the organisation? Designing and introducing an effective system for KPI measurement can be complex and challenging. We must ensure that we have allocated sufficient financial and staffing resources to the task. If we don’t have the expertise in-house, it may be necessary to call on external specialists
Have we specified the elements that will be measured and how frequently data will be collected (daily, weekly, monthly)?
Have we determined who is responsible for data collection in each department and ensured that they have the strong analytical skills needed to understand and record KPI information correctly? Training may be needed.
Have we established how KPI data can be independently verified and audited?
As the Balanced Scorecard reminds us, we should be wary of collecting too much data. Including too many performance measures can dilute understanding of the most important areas and lead to a loss of focus. Streamlined processes tend to lead to better results. Simple rules and metrics are more likely to be remembered and acted upon than highly complex ones.
As leaders, we need strong performance management skills to ensure we keep our KPIs on the prize.
As a first step, we must ensure that everyone understands why our KPIs have been chosen, how they will help the organization to achieve its goals, and their own role in delivering success.
As with strategy, successfully communicating how an individual has contributed to organisational achievement can really help to build motivation, and ultimately drive even better performance. So it’s wise to hold regular reviews to communicate the organisational progress revealed by our KPIs and to discuss any issues that may have emerged.
As leaders, we must also be able to develop relevant KPI targets which can become part of each individual’s performance management goals, and – where appropriate – recognised with rewards and incentives. It is also essential that there is a clear strategy for how to address underperformance, as a lack of action can rapidly undermine the overall credibility of the system.
Presenting the data
Being able to ‘see’ the data is crucial to our understanding of it – and to sharing it, discussing it and otherwise making it ‘real’.
Many organisations use a dashboard to present KPI data in a meaningful way. Performance dashboards are visual, short-form aids that communicate strategic objectives and enable us to measure, monitor and manage the activities needed to achieve operational and strategic goals.
They are commonly colour coded using a traffic light system to establish priorities for action. For example, red flags might indicate unsatisfactory results, amber mixed results and green successful outcomes.
KPIs versus OKRs
While key performance indicators (KPIs) are an enduring feature of the business landscape, some companies prefer a less top-down model for measuring progress towards common goals: objectives and key results or OKRs.
The concept of OKRs was developed in the 1970s by Andy Grove, then president of Intel, but it really gained traction when Grove’s former colleague, John Doerr, became a venture capitalist and introduced the concept to Google in the 1990s. The approach was then made popular by a number of tech companies in Silicon Valley - and beyond. And more recently, it has been embraced by the likes of Google, Spotify, Twitter, and LinkedIn.
Doerr’s formula for OKRs is simple:
I will (Objective) as measured by (set of Key Results).
Objectives define where we want to go.
Companies typically create three to five high-level objectives per quarter. They should be short, inspirational and ambitious.
For example, increase profit by 20%; become the market leader in our industry; develop autonomous vehicles.
Key Results are the deliverables we define for each objective.
Each objective should have two to five key results, all of which must be measurable
For example, reduce production costs by 10%; record $100 million in revenue; roll out a prototype by the fiscal year-end.
As a matter of principle, it’s fine to achieve just 60-70% of our OKRs; if we achieve 100%, the chances are we need to be setting more ambitious goals. This marks a point of difference between OKRs and KPIs. OKRs are not an employee evaluation tool. The thinking is, if individuals are to set suitably ambitious goals, they need to know they won’t be negatively impacted if they don’t achieve every key result.
There are other important principles too, for example OKRs must:
- Be simple and agile
So that more time and resources are invested in achieving goals rather than in defining them. OKRs are typically set monthly or quarterly to promote a swift response to changing conditions
- Create clarity and alignment
OKRs are public, which helps to ensure alignment between departments and at all levels.
- Be bi-directional:
Rather than cascading from the top to the bottom each group and individual simultaneously builds tactical OKRs that align with the strategic OKRs. Everyone needs to get on board as the process won’t work if only a portion of employees and management are committed to it
- Encourage collaboration:
OKRs make it easy to understand how every employee has a critical role to play in achieving strategic goals. It is therefore clear that no one can accomplish the ultimate goals alone and that collaboration is needed.
There is a lot of overlap between OKRs and KPIs and a key result can easily morph into a KPI or meet both definitions.
A key difference is the intention behind the goal setting, with OKR goals typically being more aggressive and ambitious, which could explain their popularity with dynamic and disruptive organisations such as Uber and Airbnb.
In1988, the European Foundation for Quality Management (EFQM) created a strategic tool to help organisations keep driving for excellence.
The EFQM Excellence Model can be particularly helpful for identifying areas for improvement and for setting future goals and performance standards.
The model is based on eight attributes common to truly excellent organisations:
- Results orientation
- Customer focus
- Strong leadership
- Systematic and informed management
- Employee development
- Continuous learning and innovation
- Working with partners
It shows organisations the cause and effect between ‘enablers’ (what we do) and ‘results’ (what we achieve), so we can see how close our current performance is to ‘excellence’.
The five enablers are:
Excellent organisations are defined as those with ‘leaders who shape the future and make it happen, acting as role models for its values and ethics and inspiring trust at all times.’
Excellent organisations are defined as those that ‘implement their Mission and Vision by developing a stakeholder focused strategy. Policies, plans, objectives and processes are developed and deployed to deliver the strategy.’
Excellent organisations are defined as those that ‘value their people and create a culture that allows the mutually beneficial achievement of organisational and personal goals. They develop the capabilities of their people and promote fairness and equality.’
- Partnerships and resources
Excellent organisations are defined as those that ‘plan and manage external partnerships, suppliers and internal resources in order to support their strategy, policies and the effective operation of processes. They ensure that they effectively manage their environmental and societal impact.’
- Processes, products and services
Excellent organisations are defined as those that ‘design, manage and improve processes, products and services to generate increasing value for customers and other stakeholders.’
Each of the enablers is broken down into a number of set criteria and organisations create specific measures for each enabler. These are designed to motivate excellence and success in four result areas.
The four results are:
Within each of these areas, organisations gather and analyse information on perceptions of performance among all relevant stakeholders (for example, employees, customers, wider society) and indicators of performance (recruitment and retention statistics, repeat business from customers, public recognition, awards).
As well as supporting the success of individual organisations, the tool provides a common language for sharing experiences between them. It is now used by over 30,000 organisations around the world, including BMW, Bosch, and Virgin Media.
Have you really nailed it?
In an article in the New Yorker, mathematician Hannah Fry exhorts us to bear in mind Goodhart’s Law: the idea that once a measure becomes a target, it ceases to be a good measure as people will tend to optimise for that objective and lose sight of the bigger picture.
For example, if you measure people on the number of nails made, you might get thousands of tiny nails. And if you measure on the weight of nails made, you might get a very few giant heavy nails. And neither might be useful in building a solid future. Similarly, if the sales team is measured on volume of nails, while the product team is measured on the quality of nails, it’s a clear recipe for internal resentment and clashes.
Fry cites the example of America’s first transcontinental railroad, built in the 1860s by companies who were paid per mile of track. Cue some unsurprising and unnecessary wide arcs of track that added several unnecessary (yet highly profitable) miles to the rails.
There’s a similar phenomenon known as the cobra effect. When Britain occupied India under the guise of Empire, they identified a problem with venomous snakes and offered a reward for every dead cobra brought to the authorities. Initially, this had the desired effect of reducing the number of wild cobras. But people quickly got wise and started breeding cobras in captivity. When the authorities found out, they ceased paying the bounty - and all those useless cobras were released into the wild.
It is therefore vital that we think our decisions through to avoid the law of unintended consequences.
For example, in a call centre environment, a KPI which measures the number of calls successfully resolved per hour will be an effective driver of positive customer service behaviour. However, a KPI that focuses on the number of calls answered per hour may simply encourage employees to reduce call time, which can have a detrimental impact on customer service.
It’s also worth stating that giving people too many or unrealistic targets can leave them feeling anxious, stressed, and angry, which will never improve our overall results.
Data needs a human face
As Einstein said: ‘Not everything that can be counted counts, and not everything that counts can be counted’. And who wants to argue with Einstein? The inference we should take is that even with the best KPIs in place, we must be sure to handle our data as empathetic humans rather than machines. And we should never forget that numbers are – as Fry expresses it – “a poor substitute for the richness and colour of the real world”.
Performance measurement may highlight areas where processes need to improve, but those improvements will only result from changes in actions and behaviours. And we will only succeed in motivating these changes by considering each person and their many talents, not just their output.
So let’s make sure we measure up.
Test your understanding
- Outline three potential positives of having well-thought-through KPIs
- Identify the four elements of Kaplan and Norton’s Balanced Scorecard.
- Explain the difference between leading and lagging indicators
What does it mean for you?
- What are you more excited by: a dashboard flashing green, amber or red? What’s driving your answer?
- For you personally, what is the most challenging aspect of telling a team member they are falling short? How could framing this news within the context of KPIs help you?