Getting a 360-degree view of your customers

This is one of the biggest and most publicised areas of big data use today. Here, businesses use big data to better understand customers and their behaviours and preferences. Businesses can gain a full understanding of customers – what makes them tick, why they buy, how they prefer to shop, why they switch, what they’ll buy next and what factors lead them to recommend a company to others. Companies can also better interact and engage with customers by analysing customer feedback in order to improve a product or service.

Getting a more complete picture of your customers may involve a combination of traditional in-house data, social media data and browser logs, as well as text analytics and sensor data (there’s more on different types of data in Chapters 4 and 5). Small businesses can also take advantage of large, public data sets to glean in-depth insights on their customers.

Aggregators or hub companies are springing up in most industries that provide data services to small companies. One example of this is Factual.com, which provides location data to help companies serve their own customers with more personalised experiences and recommendations. Other services and tools exist to make it easier for businesses to dip their toes into the water and see if the results will be worth the investment. For example, Silverpop, which was acquired by IBM, Adestra and Marketo all offer marketing automation driven by data – either from the business’s own data, or from bought-in or public records.

Food delivery outlets have been quick to jump on board the Just Eat and Hungry House bandwagons, which allow customers to order food straight from their smartphones to their homes. The thousands of restaurants they have signed up all have access to data telling them what their customers usually order, what days of the week they like to get food delivered and what other restaurants they patronise in the local area. Restaurant owners can use this to find the optimum locations for their delivery outlets as well as to schedule offers and promotions and tailor their own menu to suit their customers’ tastes. They also get access to valuable metrics, such as how far away their average customer lives, how much they spend and what time of day they like to eat. TripAdvisor.com offers similar data services to owners of hotels, restaurants and leisure companies (big or small).

Understanding (and predicting) trends in your industry

Wouldn’t it be great to be able to predict the future? If you had a crystal ball or a time machine, you could remove all of the guesswork from making business decisions. You’d directly see the products of your labour and instantly know whether or not your present efforts were on the right track. This is where trend spotting comes into play. Whether you intend to buck them or follow them, as a business person, you seek to identify trends in industrial practices, customer behaviour and anything that could make a difference to your bottom line.

Spotting and monitoring behaviours and patterns allows you to take a stab at predicting where things are heading, how demand for your products or services will change over time and what will prompt that change. What can you do to respond to it? To increase demand when it’s low and ensure supply when it’s high? Until recently, trend analysis and prediction often came down to gut instinct – that feeling that people who are confident in their abilities get, and which they often feel puts them at the top of their game or gives them an edge over their competitors. Now, big data is taking a lot of the guesswork out of that process – it’s the closest thing you have to a crystal ball!

Marketing is a great example of understanding and predicting trends. With the advent of social media and the Internet, people are used to (knowingly or unknowingly) sharing vast amounts of data about themselves, their interests, habits, likes and dislikes – and savvy marketers have been quick to tap into this. Trending topics flash across Facebook and Twitter every day, making it easier than it has ever been before to work out what people are looking for and what they want. Products and services can then be marketed to fill those needs. Services such as Trendera and Trend Hunter collate this data and use it to answer specific questions for their business customers.

The ability to know what the public wants before they know it themselves is every business’s Holy Grail. In retail, online and offline customer behaviour can be measured to microscopic detail. That data can be compared with external data, such as the time of the year, economic conditions and even the weather, to build up a detailed picture of what people are likely to buy and when. If you run an organic farm shop, for example, weather data can help you identify in advance when you’re likely to sell a truck load of sausages and ice cream. This information can also inform your production and stock levels and marketing activity.

Much of this technology is based on free, open-source software, or inexpensive, software-as-service cloud-based solutions. And a lot of businesses have gained very valuable insights from the free, huge public datasets made available by companies like Google (Google Trends is a fantastic tool) and government services such as data.gov.uk. In fact, I devote a whole chapter (Chapter 15) to the top ten free data sources for small businesses. Technology like this makes it perfectly viable for many small and medium-sized enterprises to have a crack at predicting the future themselves. You might not yet have the technology to actually see into the future, but now you have the ability to remove the guesswork and replace it with cold, hard data-driven insights.

Reviewing the competition

In the past, understanding your competition was limited to industry gossip or looking around rivals’ websites or shops. Some competitors might go as far as pretending to be customers in order to find out more about a service or product. These days, however, you hardly need to leave your desk to find out what the competition is up to.

Today you can access a wealth of data on your competitors: financial data is readily available, Google Trends can offer insights on the popularity of a brand or product, and social media analysis can illustrate popularity (how often a company is mentioned, for example) and show what customers are saying. Twitter is particularly transparent and a brilliant place to start. All the information you gather can be compared with your own brand; for example, does your competitor get more mentions on Twitter? How do their Twitter conversations with customers compare with yours? Does their Facebook page have more Likes and Shares than yours? Now you can get a much richer picture of your competitors than ever before.

The flip side of course is that your competitors will be able to glean just as much information on your company. There’s not much to be done about this, unless you want to withdraw from online life and social media platforms altogether (not at all recommended for modern businesses!). The best way to stay one step ahead is to keep up-to-date on advances in big data – by reading books like this – and treat it as an ongoing part of your business life, not just a one-off project or investment.

Gaining internal efficiencies

For companies with a manufacturing or industrial focus, machines, vehicles and tools can be made ‘smart’, which means they can be connected, data-enabled and constantly reporting their status to each other. Machine data can include anything from IT machines to sensors and meters and GPS devices.

By using this data for operations analysis, organisations can gain real-time visibility into their operations. This increases efficiency by allowing every aspect of an industrial operation to be monitored and tweaked for optimal performance. It also reduces costly down-time – because machinery will break down less often if you know exactly the best time to replace a worn part.

This use of data isn’t limited to manufacturing businesses. In retail, for example, companies are able to optimise their stock based on predictions generated from social media data, web search trends and weather forecasts. This allows stores to stock up on the most popular items, ensuring they don’t miss out on sales and reducing the amount of unwanted stock lying around. Depending on the systems you choose, this process can even be automatic, with stock automatically replenished when certain conditions are identified, such as on-hand stock of an item drops below a set number.

Challenging your business model

Data can even become a part of your business model, leading to exciting new ways to generate revenue. Facebook, for example, is free to users but has historically generated income from advertising. Now the company is capitalising on the huge amount of data it has on its users, by making certain data available to businesses. Some of this data is available for free but some of it you have to pay for, creating a new income stream for Facebook.

You may not be a big data giant like Facebook, you may not even be generating data on a ‘big’ scale, but data could still influence your business model. If you have the ability to generate data, you may find that data could prove valuable above and beyond what it was originally intended for (see the sidebar, ‘Planting the seeds of a great app’ for an example of this).

Obviously, this way of using data is an ‘icing on the cake’ sort of thing. Most businesses set out with a very particular goal (or goals) they want to achieve with the help of data, and many are happy sticking to that. But, when it comes to data, it’s always a good idea to stay open to new possibilities. If this appeals to you, check out Chapter 13 for more information on building a big data culture and incorporating data into your business model.

Optimising your supply chain

Supply chain, or delivery route, optimisation is one business process that benefits heavily from big data analytics, largely because it’s an area that’s so rich in data. Here, GPS and sensors are used to track goods or delivery vehicles and optimise routes by integrating live traffic data and so on. Inexpensive sensors can be placed on vehicles or individual pallets, or businesses can use their drivers’ smartphones to track progress.

One great example comes from a pizza delivery company that tracks drivers very simply using the GPS sensors in their smartphones. This gives the company new insights into how to optimise delivery routes – by tracking where their drivers are and monitoring traffic conditions using publicly available data, they’re able to deliver to customers faster and more efficiently. This means fewer free pizzas given away for late delivery (and your tasty pizza arriving in your belly earlier)!

Recruiting and managing talent

Companies are nothing without the right people. More often than not it’s the people that give a small business its competitive edge. It’s therefore absolutely vital that businesses find, recruit and retain the right people. Data can help you find the most successful candidates, understand whether your current recruitment channels are effective and help keep your existing employees happy.

A client of mine wanted to recruit self-driven people able to take initiative. By analysing different data sets from the type of people they wanted to recruit and those they wanted to avoid, the company found that candidates who filled out applications with browsers that weren’t pre-installed on their computers and that had to be installed separately (such as Firefox or Chrome) tended to be better for that particular job. This was a simple thing to measure but it streamlined the process (eliminating those that didn’t meet the criteria before interview stage) and meant the company could find the right sort of people more easily. Another one of my retail clients is now able to analyse social media profiles in order to very accurately predict the level of intelligence as well as the emotional stability of potential candidates.

The applications go way beyond recruitment. For a seemingly people-focused area, HR (human resources) processes generate a surprising amount of data. Much of this is related to staff performance and engagement, such as absenteeism figures, productivity data, personal development reviews and staff satisfaction data. In addition to these traditional types of data, companies can now collect so much more data that wasn’t available before: capturing employees on closed-circuit television (CCTV), taking screenshots when staff are using company computers, scanning social media data, analysing the content of emails and even monitoring where employees are using the data from geo-positioning sensors in corporate smartphones. The challenge is to establish what data is really going to make an impact on your company performance. What’s really useful? It might, for example, be increasing employee satisfaction in order to reduce staff turnover and sick leave.

With more than half of human resources departments reporting an increase in data analytics since 2010, it’s obvious that data-driven HR practices are here to stay. While capturing data on this scale may all seem a bit Big Brother, the benefits to businesses can be huge. The benefits to a company able to accurately identify why one particular customer sales representative outperforms his colleagues are obvious. The same is true in sports, where big data analytics is playing an increasing role in boosting performance – as shown in the sidebar ‘How sports team use data to get the most out of people’.

Crunching employee data could prove disastrous if a company gets it wrong. In workplaces where morale is low or relationships between workers and managers are not good, it could very easily be seen as a case of taking snooping too far. It’s vitally important that staff are made aware of precisely what data is being gathered from them and what it is being used for. Everyone (and certainly those running the operation) needs to be aware that the purpose is to increase overall company efficiency rather than assess or monitor individual members of staff.

Dealing with IT and security

Big data can help optimise IT (information technology) resources, which, in a small company, can be especially precious. Data and algorithms can be used to identify vulnerabilities in IT systems, reduce risk, detect fraud and monitor cyber security in real time.

One way big data allows businesses to detect fraud is by analysing credit card transactions in real time. This includes the ability to shut down transactions that are suspicious or not feasible, for example, purchasing something in New York City at 2 p.m. and in New Deli at 3 p.m. On a much grander scale, big data analytics are also used to detect terrorist activity and cyber security attracts by constantly monitoring and processing data including phone conversations, social media messages and emails as well as sensor and machine data.

A great example comes from the insurance industry, which has made great strides in using data to detect fraud. By analysing the length of time taken to complete a claim online, or by analysing whether a client goes back and changes information on a previous page, they can flag a potentially fraudulent claim.

Transforming research and development

One area that’s really embracing big data and analytics is healthcare. In 2003, when scientists decoded the human genome, it took a decade of intensive work to sequence three billion base pairs of DNA. Today, the computing ability of big data analytics enables scientists to decode that much DNA in a day! This data now allows scientists to predict the likelihoods of getting certain diseases, which in turn can lead to preventative actions and early interventions.

The battle against cancer is using gaming to advance research, providing another fascinating insight into what’s possible in a smarter world. By some estimates, 81 million people worldwide spend up to nine and a half hours a week playing mobile phone and online games like Candy Crush Saga, Flight Control and Angry Birds. Today scientists are tapping into that obsession in an effort to solve a whole host of important medical problems. New games are being created with the potential to pinpoint key information about killer diseases like cancer and diabetes. In one example game developers in Dundee created a mobile phone game similar to Space Invaders called Genes in Space that could help to cure cancer. Although to the gamer it looks like he’s navigating through stars and galaxies, what he’s actually navigating through are graphics made up of the DNA information of thousands of tumour samples. Every time a player completes a level it means that one DNA sample has been mapped and the data automatically sent back to the lab at Cambridge University for analysis. In the month following the game’s release, the lab received 1.5 million analyses from gamers. Considering that one analysis normally takes five minutes to map, it would have taken the research team 125,000 non-stop hours or 14 years to cover the same amount of data that the gamers had covered in just one month!

Obviously there are some massive players in the healthcare and pharmaceutical industry, but a lot of innovation is driven by small start-ups. These small companies can leverage big data to conduct research based on a huge range of factors – for example, which drug is most effective for which genetic makeup.

Unlocking connections in data sets

A common mistake companies make with data is to only ever look at data in little silos. But by looking at combinations of data you can spot really interesting and illuminating connections.

The value of data is in the combination of data sets and the connections spotted between them. By looking at more than one data set, you can build a more complete picture of your customers, your sales, the success of your product and so on. To do this, you’ll probably need a combination of internal and external data and traditional structured and messy unstructured data (see Chapter 4 for an explanation of the different data types).

Say you want to understand your customers better and to predict which customers you should be focusing your marketing efforts on. To really explore this, you’ll need a combination of data: internal data, such as transactional and finance data (are certain customers more profitable than others?), and external data, such as geographic and demographic data. Using this combination of data you can gain insights on which customers are the ones you should be targeting in future – you may be surprised by the results.

Unleashing weather data

Weather data is really valuable for small businesses, especially because it’s free to access, meaning absolutely anyone can get it – from a big-box style retailer to a local window cleaner. I rate it so highly, it features in my top ten free data sources (see Chapter 15).

One of my clothing retail clients has an ordering system that works with weather data, allowing the client to easily stock up on popular items based on the weather forecast. So, the ordering system makes sure the store has sufficient stock of rain macs, wellies and umbrellas for the rainy spells (which, let’s be honest, in the UK is not limited to the obvious winter months), hats and gloves for those unexpected first frosty days that always catch people unawares and flip-flops for the first really warm days of the year. It’s partly common sense, yes, but the weather data takes the guess work out of the ordering process for this company.

Using big data as your test bed

This sounds very scientific doesn’t it? Please don’t be put off! What I mean by this is simply running business experiments, for example, using data to try things out and see whether new ideas are working.

Everything useful that human beings have done has been based on some degree of experimentation. From the wheel to penicillin, from the airplane to the iPhone, ideas have been conceived, tested and failures improved and retested. Using big data as a test bed allows you to test different assumptions, measure the impact on various factors (like customer satisfaction) and analyse the results so that you can come to an evidence-based conclusion. If the conclusion is positive, then you move forward with the product or service. If not, then you abandon it, or restructure and retest it.

Testing assumptions is good business sense. And it’s something businesses have been doing for years and years – think focus groups and market-research surveys. Data just adds an extra dimension and rigour to the process.

All business experiments start with a testable hypothesis or assumption. Say, for example, you want to test the assumption that rolling out a new customer service training program will increase customer satisfaction and profit margins. It seems like a reasonable assumption, but whether it’s true or not has to be tested – not just assumed. The next step is to design a suitable test, such as training only some of your staff and not others (thereby seeing what results you get if you do nothing). The data is then analysed to determine the results and appropriate actions, for example, should all staff be trained in this way? In this example, you analyse the difference between the performance of those who were trained and those who weren’t and see whether what you expected to happen (increased customer satisfaction and profit margins) actually did happen.

Small companies can learn a lot from business experiments that big corporations run, albeit with scaled down budgets and expectations. eBay is the master of testing website changes. Its managers have conducted thousands of experiments with different aspects of its website and, because the site garners over a billion page views per day, they’re able to conduct multiple experiments concurrently. For smaller businesses, simple A/B experiments (in this case, comparing two versions of a website) can be easily implemented to see how changes to your website impact on overall stickiness (how long customers spend on the site) and sales.

Given the wealth of data and analytics tools now at your disposal and the increasing connectivity with customers and other stakeholders, running business experiments is a relatively straightforward approach for testing proposed new products, or product or service enhancements. But note the idea is to test, not to prove an assumption (an easy trap to fall into). Organisations that punish failure in such experiments should remember that getting things wrong is usually a vital stepping stone to getting things right!