Facilitating company-wide buy-in

Your ultimate aim should be to create a culture in which people naturally want to make better decisions and look to data to help them do so.

A good way to sow the seeds for improved decision making is to engage key personnel in developing your data strategy, which I talk about in Chapter 10. For example, if you want to use data to better understand and target your customers, then involve your marketing head from the outset. Or, if you want to improve manufacturing output and reduce unexpected downtime, then you’ll need to get your manufacturing people on board. Encourage those key personnel to become data advocates, creating a trickle-down effect.

Another key step is to use the insights you gather; don’t just sit on them. It sounds obvious, but, after identifying your strategic questions and analysing the data to get answers, you really need to act upon the insights found. What you do will encourage your people to shift their thinking towards evidence-based decision making. If you do nothing, you really can’t expect the overall company culture to change. Use that valuable knowledge, demonstrate positive outcomes, and it will be much easier to get buy-in from others.

Here are my top tips for facilitating company-wide buy-in:

• There are naysayers in every company and negativity can be contagious! Identify sceptics or blockers and spend time engaging them. Use their pain points to show how fact-based decision making can make their jobs easier.
• Make it easy for people to understand data and pull out the key facts they need to make better decisions. Present insights in an attractive and engaging way. (There’s more on this in Chapter 11.) It’s of no use to anyone if key facts are buried in long, dull reports.
• With this in mind, build and maintain strong links between the people analysing the data, the people reporting the insights and the people making business decisions – knowing what the decision makers need to know makes it easier to present information to them.
• Be open about what you’re measuring and why. For example, if you’re focused on improving people management, then you’re probably going to be measuring what your people do, when, for how long and so on. Understandably, this can make people nervous, with certain Big Brother overtones! Don’t be sly about this. People are far more likely to be comfortable if you’re honest from the outset and emphasise the positive goals (in this example, building a better workplace for everyone).
• Lead by example and use hard facts as the basis of everything you do. This is harder than it sounds; if you’re a successful business owner or manager, good gut instincts have probably played a large part in where you are today! Make a commitment to changing the way you make decisions and your people will follow.

Following these pointers can help everyone in the business leverage data and move towards more fact-based decision making.

Emphasising the positive impact data can have

There’s no denying it: change is difficult for many people and businesses. And implementing a cultural shift in an organisation, whether big or small, is not a quick and easy job. But it is crucial if your company is to make smarter decisions going forward. Focusing on the positive outcomes certainly helps smooth the way.

Broadcast positive goals and outcomes load and clear. If data has helped reduce wastage by ten per cent, then you should be shouting about that success. If you’re only at the beginning of your data journey, share examples from similar companies. Seeing how other people have changed their decision-making processes can provide a helpful nudge in the right direction.

Focusing on the good that big data can bring, either with insights from your own business or examples from elsewhere, lays the foundations for fact-based decision making.

Managing talent

Big data is now filtering through into human resource processes. Largely, this falls into two camps: finding the best talent and hanging on to that talent!

Recruitment has traditionally been a somewhat hit-and-miss affair; statistics show that almost half of appointments end up as failures within 18 months. That shouldn’t be surprising; personalities are often the hardest things to accurately convey and interpret in an interview and around 90 per cent of appointment failures are down to attitudinal reasons.

Perhaps more surprising is that companies continue to make hiring decisions (with all the financial liabilities that they entail) based on an often brief interview and a cursory look at a CV. Much of the time the decision is based on an interviewer’s gut feeling. Considering that a large proportion (40 per cent to 60 per cent by most estimates) of a company’s revenue goes on staff salaries, it makes sense to be a bit more strategic about the recruitment process and base decisions on data rather than instinct.

Taking a more scientific approach to appointing staff enables you to find more suitable people who stay happy and on the job for longer.

Boosting employee satisfaction

Companies now have more data than ever on their employees and more tools and technology with which to analyse this data. As well as optimising talent acquisition, big data tools can help companies measure and improve company culture, staff engagement and overall employee satisfaction.

Google has regularly been voted the best company in America to work for – its staff get free meals, generous paid holidays, access to nap pods for power napping during the day and are even encouraged to grow fruit and vegetables at work. And, despite their ‘don’t be evil’ motto, Google top brass hasn’t done all this simply because they are lovely people. Like everything they do, their decisions were based squarely on data – and in this case, the data showed that treating staff well would increase employee satisfaction.

Employers have been using analytics for some time now to understand what makes their staff tick, using metrics such as staff engagement to understand what drives productivity and innovation in the workplace. The humble employee survey is a precursor of this.

Implications are also beginning to be realised for health, safety and well-being at work. Obviously, when people become ill at work or are absent for long periods of time, it’s stressful for the individual involved, but it’s also detrimental to the business and the people left picking up the extra workload. Personal analytics and health monitoring can change all that and give you a real-time insight into health and well-being. Hitachi’s Business Microscope service enables companies to fit its staff with Radio Frequency Identification (RFID) tags that track their movements around the workplace and even monitor sound waves to identify how stressed or relaxed they are when they speak.

People who work in potentially dangerous or stressful environments are being measured to monitor their fatigue and stress levels so that employers can pull them off jobs before they get too tired and potentially cause an accident.

Data can provide a wealth of insights in terms of productivity, and even sales. In one trial, a retailer was able to increase sales by 15 per cent after it noticed that the presence of a member of staff in certain areas of the store had a high impact on products sold, while in other areas, staff presence had very little effect. In a seated office environment, technology can record how long employees spend at their desks, how much time they spend interacting with other staff, whom they talk to, the distance they stand from each other during conversations and the enthusiasm with which they contribute to meetings.

One of my clients uses analytics tools to scan and analyse the content of emails sent by staff as well as the social media posts they make on Facebook or Twitter. This allows the client to accurately understand the levels of staff engagement, and it no longer needs the traditional staff surveys, which were expensive, time-consuming and less accurate.

Big Brother approaches may sound suitably Orwellian and difficult to get staff to engage with. But how well they’re received by staff depends very much on the way they’re used. If used as a disciplinary tool focused on the behaviour of individuals, they will undoubtedly lead to resentment. But, when used as a way to gain an overview of the company as a whole and how people interact to get the job done, you’ll get more buy-in – and far more useful insights. At the end of the day, people are far less likely to complain if the insights generated benefit them, whether it be nap pods or more annual leave!

Increasing operational efficiency

Operational efficiency can be defined as delivering products or services in the most efficient and cost-effective way possible. Often operational decisions are based on experience or ‘the way things have always been done’. But data and technology have a lot to offer in terms of making your business as efficient as possible.

For a business, the ability to have its production, stock control, distribution and security systems all connected and talking to each other means greater efficiency and less waste.

The efficiency of every machine – and human for that matter – can be recorded so companies know what’s working and can make improvements where they’re needed. The possibilities are endless; it’s about choosing what’s best (and what’s possible) for your business.

Big data analytics also help machines and devices become smarter and more autonomous (for more on sensor data, machine data and the Internet of Things, see Chapter 5). Manufacturers are monitoring minute vibration data from their equipment, which changes slightly as it wears down, to predict the optimal time to replace or maintain parts. Doing this too soon wastes money; doing it too late can trigger a failure and expensive work stoppage.

Operational efficiency is arguably one of the areas where it’s easiest to convince people of the benefits of fact-based decision making. Hard facts related to performance, maintenance and the financial savings that come from greater efficiency take the guess work out of decision making. As such, this might be a good starting point for your business, particularly if you’re in manufacturing or distribution of any kind.

Optimising business processes

Ideally, your people will look to data to improve all aspects of business decision making in all areas of the company. This includes everything from stock management to customer relationships to security.

Many retail companies are already using algorithms to understand what’s trending in social media and what competitors are charging in real time. Algorithms are also great for recommending other products a customer might like – a strategy basically pioneered by Amazon and used to great effect.

I have worked with a number of hotel chains that want to move away from the traditional in-house surveys, which are costly and questionably accurate, to using social media to analyse what people are saying and posting about the hotel. This way, hotel managers can better understand their customers and improve their service. By running and using sentiment analysis (see Chapter 5) on Facebook posts, tweets and other social media sites and reviews on Trip Advisor, in addition to existing data, hotels are getting far more reliable information than they would from a survey.

Stock management is another area with enormous potential, using data from social media, web search trends and weather data, for example, to build predictive models for what your customers will want and when. One often-cited example is the supermarket chain Walmart that discovered a surprising correlation between hurricane warnings and sales of Pop-Tarts. Apparently, in a hurricane, people just want to hunker down and eat Pop-Tarts. Perhaps the UK equivalent is huge sales of charcoal, sausages and hot buns on the first vaguely warm day of the year – usually in April and usually when the temperature is only around 15 degrees Celsius (around 60 degrees Fahrenheit)!

It’s easy to get caught up in fun insights like the Pop-Tart analogy, because data has the ability to tell you some mind-boggling things. The trick is to focus on what benefits your business’s decision making and what you can actually act upon. For example, if you’re a retailer that’s already sold your shelf space to specific companies and guaranteeing space for particular products, then knowing that sales of a certain bottle of wine increase on Tuesday evenings isn’t going to help you much beyond knowing to stock up on Tuesday and re-stock on Wednesday morning. You can’t devote more space to that product on Tuesdays because the shelf space is allocated elsewhere. So, while it’s an interesting insight, it’s not necessarily useful in terms of your business strategy.

One particular business process seeing a lot of big data analytics is supply chain or delivery route optimisation. Here, GPS and radio frequency identification sensors are used to track goods or delivery vehicles and optimise routes by integrating live traffic data. For instance, if a delivery driver has an optimised delivery schedule, that schedule interacts in real time with weather data and traffic data so that if there is a traffic jam, accident or reports of delivery-impacting weather such as snow or storms, the schedule automatically calibrates an alternate route. Optimising this business process can also include delivery companies putting sensors on pallets and handheld devices that record delivery and monitor where drivers are, while also monitoring the engines of the delivery vehicles to create dynamic servicing schedules.

Technology can also improve business security and reduce fraud, with the potential for huge financial savings. In fact, big data is already applied heavily in improving business security through CCTV (closed-circuit television) video footage analytics. Credit card and insurance companies are already using data to prevent fraud. Insurance companies, for example, are using big data algorithms to check for fraudulent claims as well as anomalies in policy applications. Algorithms can now take into account the speed at which you complete a claim or application form – to spot those completed by machines versus people – as well as whether applicants have gone back and changed their initial application to reduce premiums by maybe not admitting a recent claim or decreasing their annual mileage.

How far you go in these processes is up to you. Some of much of it may not be relevant or feasible for your business at this time, which is fine. This process is not about becoming an all-singing-all-dancing data company overnight; it’s simply about improving overall decision making and becoming more analytical in how your business operates.