Data is a key component of success for any company, in any sector. It offers valuable insight and a source of key information that can be employed to improve your offering (and your profit margins). But it can only represent true value for businesses when it is used correctly. One of the main barrier’s companies face when interpreting and implementing findings is bias – the stealthy selectiveness that can so easily affect decisions without companies being aware of it.
We all like to think that we can be completely impartial when a situation calls for it, especially in a professional capacity. But in reality, we are all subject to the influence of bias. Can you think of a time in your life where you’ve jumped to a conclusion that wasn’t necessarily true? Chances are, you were biased.
In fact, almost every conclusion we make is affected by bias, whether it applies to our prejudices, predetermined ideas and desires, or to the data itself. Data can be biased – but so can the people who analyse it, and both of these scenarios can cause significant problems in a business context. It’s not uncommon for businesses working with large volumes of data to be unable to unlock the full potential of that data, but despite this many still assume it will guarantee success.
There are many pitfalls to consider when working with data, but bias is one of the most dangerous. If data can act as a ‘golden ticket’ to better decision making, bias is the principal reason for making the wrong decision.
Bias can take different forms – being aware of them is the first step towards successful interpretation of data different forms bias can take
All bias can lead us to make the wrong decisions. When we don’t have objectivity, it is difficult (and sometimes impossible) for us to make rational choices. Yet there are many different types of bias we need to look out for when working with data.
Selection bias is the subjective choosing of data, leading to a poor reflection of the population. Outliers represent extreme data values that may adversely affect any average you take of your data. Overfitting occurs when an overly complicated picture of reality is painted can cause certain assumptions to be treated as the truth when they may not apply to the whole database of customers. Co-founding variables alsoneed to be taken into account. These are variables that fall outside of the existing model that influence both the explanatory and dependent variable, failing to account for these could result in an assumption of a cause-effect relationship that isn’t the real reason behind it. Confirmation bias is the term given to people analysing the data with the intention of wanting to prove a predetermined assumption.
You may think that with the best team and a liberal outlook you’ll be immune to misinterpreting data. But as you can see, there are many different types of bias to fall foul of, and even the most open minded amongst us likely have some confirmation bias. We’ve all wanted something to be true so badly that we believed it, haven’t we? Especially in business, it’s certainly easier to believe that there’s readily available data to back up what you already think you know, or believe. But true success and growth lies in identifying and analysing those areas where you may be wrong.
What is the value of the information you might have missed?
According to John Ioannidis’ 2005 paper “Why Most Published Research Findings are False,” business isn’t the only area in which decision makers can fall victim to bias. Even in scientific findings, data is often not representative and 68% of findings are unchallenged or unable to be replicated.
Bias is a natural aspect of human behaviour which runs through most decision making. Even your customers are influenced by it, which may be why they choose you over your competitors.
So if bias is so prevalent, how can it be so negative for business? Imagine you are gathering information, but you stop when the evidence supports what you were looking for. You have fallen foul to bias. Think about all the information that you’ve missed, that might have taken your business in a different direction and helped you realise something that could enhance your strategy ahead of everyone else in the market. It deprives your business of the opportunity to see the full picture, of all the accurate beliefs that should be shaping your decisions and driving your business forward.
This is just one very simple example of the negative impact bias can have on a business. If you think of any bad decision you or your company has made, more than likely you will see evidence of bias and may even be able to trace it back to the misinterpretation of a specific set of data or misplaced beliefs.
Not all data is created equal
To achieve true success and a valuable, stronger connection with customers, businesses need to move to a place where their business decisions are truly objective. The conclusions we make need to be generated from research and analysis, where we have the full picture and put our wishful thinking aside. There needs to be a shared responsibility not only to have correct data in the first place, but to base our verdict on correct facts. It’s important to remember that facts are not facts unless they have been proven to be true. Just because something is written down, does not make it the truth.
So many businesses look for answers in their data, rather than letting data do the talking. Data alone cannot give us the answers, it needs to be mined and explored to get insight but from an impartial perspective. Think of it like a debate – we can argue with data, but all sides need to be heard before a winning outcome is declared.
Bias can negatively impact upon the value people give your product or service
Making decisions based off assumptions can also be detrimental to the value people place to your product or service. When YouTube launched the video upload feature for their app, 5-10% of their videos were uploaded upside down. Google developers were surprised such a large proportion of users were shooting their videos incorrectly, but it turned out their engineers had inadvertently designed the YouTube app for right-handed users and didn’t take into account that phones often get rotated 180 degrees when held by a left-handed user, showing their unconscious bias that would have likely frustrated any left-handed users.
Challenging the status quo
So how can businesses work to eliminate bias and produce more viable, valuable data to base key decisions upon? Renowned economic analyst and psychologist Daniel Kahneman recommends you ask three questions to minimise the impact of cognitive biases in your decision making.
Firstly, is there any reason to suspect the people making the recommendation have self-interest, or attachment to past experiences? Second, have they fallen in love with their recommendation? Lastly, was there groupthink in the team? Collaborating can help identify bias as we are more likely to spot it in others than ourselves – but putting a team together who proactively offer opposing viewpoints and challenge the status quo of the team are the key to truly overcoming it.
When we ignore information that disputes our expectations, we overlook crucial elements which prove essential when making valuable decisions. If data solely confirms what we already know or think we know, it is of little use. Data is in fact most useful when it challenges the status quo or our previous opinions on a subject, producing fresh outcomes as a consequence.
Whether you are taking a new product to market, shaping the future of your brand or planning promotional activity, it’s essential to be open minded so that you can achieve insight and find hidden truths you weren’t aware of before that will shape and develop an appropriate response or approach going forward.
Bias may seem inevitable, but you’ve already taken the first step today in learning more about it and how it can distort your thinking and overall decision making. If we foster a culture of curiosity and a shared responsibility across the organisation to argue with data and encourage opposing views, decision making will improve.
Protecting yourself against the influence of bias should be a top priority – and with professional support it is easier than you might think. At Clekt, we help businesses achieve better decision making by getting ahead, and staying ahead of their competition. If you want to find the real insights contained in your data and make business changing decisions that power sustainable growth, please get in touch as we can certainly help you.