"You can't improve what you don't measure" - Lord Kelvin
Key Performance Indicators (KPIs) are metrics that represent how various drivers of the business are performing. These drivers are often both financial and operational in nature. And while there is no one-size-fits-all when it comes to choosing the "right" metrics for your business it is critical that the data used be consistent and accurate.
"If you can't measure it you probably can't manage it. Things you measure tend to improve" - Ed Seykota
Unfortunately, there is no default answer because what is important varies not only by industry and maturity of a business, but also by the strategic objectives laid out by the CEO. However, you can't go wrong by examining metrics that provide insight into the short-term viability of the business such as:
Once you identify the appropriate metrics for your business it is imperative that you clearly define what each indicator represents, why it was chosen, where the data will be coming from, how it will be calculated, etc. You must set one standard definition and one "source of truth" as to where the data comes from. There is nominal value without the ability to compare KPIs to previous periods because there is no context.
"Data is only as useful as the context in which it is gathered and presented" - Josh Pigford
Again, there is no right answer, but there are wrong ways of going about it. Too many KPIs began to have a dilutive effect on the value of the information being portrayed. Too few metrics leaves you open to potentially missing critical signals and trends within the business. Most CFOs would agree that any number above 20 KPIs is too many and should be trimmed down, potentially requiring you to reevaluate aspects of the business. Less than 5 metrics may be too few depending on the business and amount of transactions that take place throughout the period.
"Metrics are for doing, not for staring. Never measure just because you can, measure to learn, measure to fix."
This question requires real consideration, specifically by the person(s) making decisions based off the data. One is likely to think that if you had the option between accessing real time updates to the necessary data or accessing data that gets updated just once per quarter that the former would be a no-brainer. However, as mentioned previously, data has little to no value without context. In this case the context needed is data from previous periods to compare against. That said, a period can be any duration of time that you choose but is most often refers to a length of time equal to 1 month, 1 quarter, or 1 year.
First, go back to prior periods and identify what the metric would have been if it were correct. Next, understand: how many periods the metric has been incorrect? How large is the variance from the correct number? What caused the error - bad data or a bad calculation? What decisions or actions have been taken either directly or indirectly based on this bad information?
Once you have a better idea around the true impact and potential fallout from reporting the wrong number you can act appropriately. The worst case scenario would be a public company having to restate and lower a metric previously reported to Wall St. If you find yourself in this unfortunate position the course of action to take is clear, start looking for a new job within a new industry.
Read: "Oops, Forecasting Error Slams ServiceNow as Shares Drop 20%" (NYSE: NOW)
KPIs are typically organized in the form of a "dashboard", which is analogous to a car dashboard. More often than not, the dashboard is comprised of a series of charts and graphs to help visualize the context. A dashboard should be easy to read, easy to share and comprised of metrics that are bulletproof. Any variance or trend that differs from the expectation should be called out, research, and explained in a commentary section. A dashboard needs to be built in a way that allows for updates to be made with as little human involvement as possible.
In the end you want it to be simple so that the implications of the data can be understood by all.
"If you can't explain it simply, you don't understand it well enough" - Albert Einstein
Tracking metrics is a necessity for any business to stay focused throughout the year if the objective is to achieve strategic goals. However, bad data has the ability to do far more damage than no data at all. Akin to bad data is the misinterpretation of data, such as: believing correlation implies causation and the confirmation bias.
The gold rush is a defining part of Silicon Valley. The gold of today is data, and many solutions are rushed to the world market from a small radius around Princeton University. On the other side of the Bay lies the University of California, Berkeley, a place of the Liberal Arts in contrast to the technology driven Princeton. Philip Tetlock taught at the former and provides a profound empirical and people-driven view on forecasting. His book “Superforecasting, the Art and Science of Prediction” is of paramount utility to every FP&A department and to every decision maker.
The Executive Summary
The empirical work done by “The Good Judgement Project” is financed by the IARPA, the Intelligence Advanced Research Projects Activity. It showed that
The Future is Unpredictable
It is impossible to run a correct cause and effect analysis for an event that has happened in the past. It is therefore also impossible to map a distinct route to a favourable outcome in the future. Despite today’s knowledge and processing power, academic hypotheses about the future have become ever less secure. The focus must be on people, and their abilities to establish emergent strategies with small adjustments in short time frames. This leads to supreme results and makes a strong argument for going beyond budgeting with rolling forecasts.
The Mediocracy of the Expert Opinion
Analysis of how experts forecast in their field revealed mediocracy at its best. Simple to understand messages and consistent views get more air time than differentiated interpretations loaded with doubt and restraints. Therefore, Philip devoted a complete section of his book on the role of the CEO, and how forecasting skills help him nonetheless. The findings go well with “Built to Last”, Jim Collins’ empirical analysis of what makes and keeps corporations superior in performance. This chapter is a strong argument for the strategic-minded CFO that works as a team with his CEO.
The 11 Commandments of Supreme Forecasting
The Superforecasters were assessed according to Brier scores. As stated above, a certain mindset combined with a resolute feedback environment led to extraordinary results. Philip came up with 11 methodical commandments that can be followed to attain supreme forecasting skills:
Finally, a Forecast
The Good Judgement Project is trademarked and it offers good services to businesses. Although Berkeley and Princeton may have different academic focuses, they are both part of the Valley and its rush for Cash. So let me forecast with 90% confidence that you buy that book within the next 3 days.
By Gary Cokins, Founder and CEO: Analytics-Based Performance Management LLC
Quite naturally, many organizations over-rate the quality of their enterprise and corporate performance management (EPM/CPM) practices and systems. In reality, they lack in being comprehensive and how integrated they are. For example, when you ask executives how well they measure and report either costs or non-financial performance measures, most proudly boast that they are very good. Again, this is inconsistent and conflicts with surveys where anonymous replies from mid-level managers candidly score them as “needs much improvement.”
Every organization cannot be above average!
Let’s not attempt to be a sociologist or psychologist and explain the incongruities between executives boasting superiority while anonymously answered surveys reveal inferiority. Rather let’s simply describe the full vision of an effective EPM/CPM system that organizations should aspire to possess.
First, we need to clarify some terminology and related confusion. EPM/CPM is not solely a system or a process. It is instead the integration of multiple managerial methods – and most of them have been around for decades arguably even before there were computers. EPM/CPM is also not just a CFO initiative with a bunch of scorecard and dashboard dials. It is much broader. Its purpose is not about monitoring the dials but rather moving the dials.
What makes for exceptionally good EPM/CPM is that its multiple managerial methods are not only individually effective, but they are also seamlessly integrated and embedded with analytics of all flavors. Examples of analytics are segmentation, clustering, regression, and correlation analysis.
I like to think of the various EPM/CPM methods as an analogy of musical instruments in an orchestra. An orchestra’s conductor does not raise their baton to the strings, woodwinds, percussion, and brass and say, “Now everyone plays loud.” They seek balance and guide the symphony composer’s fluctuations in harmony, rhythm and tone.
Here are my six main groupings of the EPM/CPM methods – its musical instrument sections:
CFOs often view financial planning and analysis (FP&A) as synonymous with EPM/CPM. It is better to view FP&A as a subset. And although better cost management and process improvements are noble goals, an organization cannot reduce its costs forever to achieve long-term prosperity.
The important message here is that EPM/CPM is not just about the CFO’s organization; but it is also the integration of all the often silo-ed functions like marketing, operations, sales, and strategy. Look again at the six main EPM/CPM groups I listed above. Imagine if the information produced and analyzed in each of them were to be seamlessly integrated. Imagine if they are each embedded with analytics – especially predictive analytics. Then powerful decision support is provided for insight, foresight, and actions. That is the full vision of EPM/CPM to which we should aim to aspire in order to achieve the best possible performance.
Today exceptional EPM/CPM systems are an exception despite what many executives proclaim. If we all work hard and smart enough, in the future they will be standard practices.
Randall Bolten , longtime Silicon Valley CFO, author of "Painting with Numbers: Presenting Financials and Other Numbers So People Will Understand You” and adjunct professor at U.C. Berkeley Extension
Let's take a look at some of the most messed-up, incomprehensible recent examples of quantation. Not surprisingly, all are graphs. But some come from sources that definitely should know better. With some, try to figure out what went wrong; with others, if you can figure out what the heck they’re trying to say, please let me know. Enjoy!
How much do you spend? We’ll start with a simple one, from the imgur.com website (beats me where they got it, but click here to see the original):
Do those percentages look like they match the length of the bars? And how about that strange scaling? And is that per day? Per year? What went wrong here?
I really hate to pick on a national institution like the Girl Scouts, but take a look at this graph that appeared in an article on CNN.com , about how the Girl Scouts are going online to sell cookies. We’re talking digital cookie sales here, not digital cookies:
Again, do the percentages match up to the size of the bars? What went wrong here? And how will this affect Girl Scout cookie sales?
The Hebrews do it backwards, which is absolutely frightening! (Google that whole line, if you’re under 50.) An Israeli friend, No’am Newman, sent me this, from a newpaper article about how people use the Internet:
In case you’re wondering, here’s the translation:
Again, do the bar sizes match the percentages? What happened here?
Yes, there’s always good old’ Fox News! No discussion of incompetently presented information would be complete without at least one pie chart. We have Fox News – OK, OK, the local Fox affiliate in Chicago – to thank for this one:
It doesn’t get much more incoherent and meaningless than this. If you want to see the reporter – uh, I mean the teleprompter reader – blithely whip through this one, click here. Again I ask: What went wrong here?
HUH??? Please, please tell me what the heck this graph trying to say, about the critically important subject of what blue-chip basketball players major in in college:
… and this is from Bloomberg BusinessWeek, for goodness sake. Warning: there will be a two-hour exam on this in tomorrow’s class, with both multiple choice and essay questions.
A breath of fresh air from a sitcom! At last – a coherent juxtaposition between bar graphs and pie charts, and surprisingly, we have Jason Segel of “How I Met Your Mother” to thank for it:
I might quibble about the repetitive use of colors, and I am not personally a fan of 3-D graph elements, but all in all I give kudos to the deep thought behind this elegant presentation. (Click here for the YouTube clip of this scene.)
Along these lines, I once again call your attention to the world’s most accurate pie chart. Click here to view it. In at least one sense, it might not be so accurate. Can you spot the problem?
RANDALL BOLTEN grew up in Washington, D.C., the son of a CIA intelligence officer and a history professor. He is passionate about the importance of presenting financials and other numerical information in a cogent and effective way, and in his current life is the author of Painting with Numbers: Presenting Financials and Other Numbers So People Will Understand You (John Wiley & Sons, 2012).
He is a seasoned financial executive, with many years directing the financial and other operations of high-technology companies. His experience includes nearly twenty years as a chief financial officer of software companies.
He has held the CFO position at public companies BroadVision and Phoenix Technologies, and at private companies including Arcot Systems, BioCAD, and Teknekron. Before his CFO positions, he held senior financial management positions at Oracle and Tandem Computers.
He received his AB from Princeton University, headed west to earn an MBA at Stanford University, and ended up staying in Silicon Valley.
In addition to writing Painting with Numbers, he currently operates Lucidity, a consulting and executive coaching practice focused on organizing and presenting complex financial information. He divides his work time between Glenbrook, NV and Washington, DC, and maintains an office in Menlo Park, CA.
In his book THINKING, FAST AND SLOW Daniel Kahneman describes two schools of psychology within the study of decision making. Clinical psychologists advocate the use of methods like heuristics (rules of thumb) and intuition for making decisions.
Statistical psychologists, on the other hand, advocate the use of methods like simple algorithms or formulas for making decisions. Clinical psychologists believe their methods are better than the methods used by statistical psychologists and vice versa. The passion that each school of psychology has for its methods led me to examine my work as an FP&A practitioner.
I found myself to be a clinical FP&A practitioner.
The methods used in clinical psychology, heuristics and intuition, are used in my work as an FP&A practitioner. A heuristic that I use when preparing financial plans is when in doubt do not overstate revenues and understate expenses. A heuristic that I use when conducting financial analysis is to start with a review of current assets and liabilities in order to assess liquidity. I use intuition when conducting financial analysis by listening to people describe the intent of their actions as well as observing facilities like offices, factories, and warehouses in order to assess the viability of organizations. I consider heuristics and intuition to have value in my work due to what I learned through situations with not only my employers but also my clients.
After reading how I use methods within clinical psychology for my work in FP&A one may state that I am a clinical FP&A practitioner. Using heuristics and intuition help me but these methods are not enough to do my job. I also use methods that are outside the school of clinical psychology.
By using methods outside the school of clinical psychology I found myself to be a statistical FP&A practitioner.
The methods used in statistical psychology, simple algorithms, and formulas, are used in my work as an FP&A practitioner. I use these methods primarily for financial planning. I have simple algorithms to calculate forecasts of year-end balances in accounts like accounts receivable, inventory, and accounts payable. I use the simple percentage of revenues formulas to calculate expenses like cost of goods sold, selling/general/administrative expenses, as well as research and development. These methods although simple provide a framework for reasonable estimates within financial plans.
After reading how I use methods within statistical psychology for my work in FP&A one may state that I am a statistical FP&A practitioner. Like the methods within clinical psychology that I consider helpful, the methods within statistical psychology also are considered helpful. It is the use of methods within both schools of psychology that help me do my job.
I find myself, therefore, to be a clinical and statistical FP&A practitioner.
FP&A is a learning process that creates insight into what organizations are doing and where they are going. The key word in this definition is the process. No process is perfect but the pursuit of creating a perfect process may create an excellent process. In order to seek excellence in FP&A, a practitioner should not limit oneself to the use of methods to do one’s job. An FP&A practitioner should have a number of methods to use. The method that works in one situation may not work in another because the situation is different. As a result, a word that should describe the effort of FP&A practitioners is flexible.
Remember… satisfaction from outcomes arise from satisfaction from tasks.
Remember… one can think without planning but one cannot plan without thinking.
Copyright, ©2017 fpa-trends.com. All rights reserved.