Monetisation can be viewed strategically and operationally. Strategically, we can look at new business directions, step changes in thinking, disruptive innovation and new income streams. Operationally, we can consider optimising current business models, making improved use of customer targeting and segmentation, and generally providing a better service. The combination of all accessible data together with the involvement of future thinking and what people see as emerging trends and new ideas builds the foundation for devising future strategies. Hence these strategies are data driven. The data‐driven approach, arising from the appreciation of the power of data when complemented by creative, out‐of‐the‐box thinking, gives an even more exciting prospect of innovation.
In this chapter, we focus on strategy. We consider how to identify important strategic business issues and new business opportunities based on data. These include:
We need to think about actions to be taken now and in the future and pay attention to defining problems, collating relevant data and quantifying baselines. These are all components of the problem‐solving methodology.
Companies need to move forward with new strategic advances. Natural goals are to increase their market share and profits. Generally speaking, there are six types of strategic opportunity that apply to most companies. These are shown in Figure 7.1.
All of these strategic opportunities can be data driven:
These are general strategic opportunities, and different organisations will prioritise them in their own way according to their needs. For each of the opportunities we can identify a purely strategic part and an efficiency improvement part.
Figure 7.2 shows how data provides the levers and drivers to boost top‐ and bottom‐line results. We can enlarge upon the box headings. First of all the strategic part:
Secondly the efficiency part:
So far, we have considered well‐known issues but the key success factor is to bring these to life. To do this we take the next step and look in detail at the role of data and how it is monetised.
There are many exchanges where data is apparently freely given but is actually being used as payment in exchange for goods and services. We explore the strategic opportunity in persuading people to donate their personal data to an organisation. In the early days of the internet it was more difficult to attract people and gain their trust and so a lot of effort was needed to encourage people to use an internet site. Nowadays, people are quite happy to visit internet sites; they know that they can find nearly everything they want there for free and are scarcely worried about giving away information about themselves in return. Building a business on the personal data collected from visitors to a website is therefore a viable strategy. People flock keenly to websites and in great numbers. In fact, although people are effectively paying with their precious data for using the internet they still feel that it is all for free. Figure 7.3 shows a typical data request for a free service.
People are quite familiar with giving their names and email addresses when they enter a competition, or ask for a government report, or a free download of software, or attend a workshop. Sometimes they give their data in exchange for a service, for example one that means that they no longer have to recall passwords, or one that means that they are personally greeted on the webpage. To join in with any of these activities, people have to share their personal data, otherwise they cannot be part of the network. At the very least, they share their name, email and often their home addresses as well.
Being active in a social network normally means that users have to give much more information about themselves, in the form of profiles, interests, history, other memberships, friends, followers and contacts. The network encourages them to be active and to manage their private and business life using the facilities of the network so that the network and all the other people to whom they give permission will have access to this information. The big advantage for the network is that it not only gets information about these people but also about the other people they deal with.
In addition to this active handing over of data, users give their data passively by allowing themselves to be tracked as they perform other activities. It is more and more common to be pressurised into giving personal details on the web, with users who refuse to do so being excluded from even basic access. For example, a lot of websites limit usage unless cookies or pop‐ups are accepted.
Recall that cookies are small programmes that the website owner asks the user to allow on to their computer when they first visit the site. Once accepted, the cookie sends back information to the website owner or a third party such as adword or banner advertising companies and companies specialising in tracking and behavioural targeting. Cookie information can simply be an identity and the fact that the user is visiting the site, or it can be full tracking information recording where the user has been on the web and what they have looked at.
Pop‐ups are online advertising views that appear on the screen enticing the user to click on them to find out more about them, or even just to move them out of the way. If the user does not accept things like cookies and pop‐ups, they may not be able to access the full website. Mobile apps have a similar function in that, apart from their main job of guiding the user to a desirable place, offering a discount or amusing them with a game, they collect personal data about users’ whereabouts and activities. Being observed is the price paid for the convenience of using the web or, to put it another way, the user’s personal data is the price they have to pay for the familiar, fast, slick convenient digital world.
Every time there is contact between the consumer and anything to do with the organisation it us referred to as a ‘touchpoint’. Examples include visits to a website or a social network where the organisation has a presence, responding to advertisements relating to its products or brands, and any purchasing behaviour. Figure 7.4 shows the touchpoints where data is collected, demonstrating that nearly all consumer touchpoints can generate data. Tracking people provides information about the number of times they encountered a touchpoint, for example the number of times they passed a billboard advertisement. This helps when assessing the likely impact of the touchpoint. As part of an analytics process, the touchpoint data can be used for many different purposes and in many different places, as shown in Figure 7.4.
Compared with the early days of the internet, when valuable information was given to the user for free, now the user has at least to give some of their personal data in exchange. This together with their movements and viewing habits is the contract made for enjoying the digital world. On the positive side, all this tracking and monitoring helps to gather valuable data that will be used to improve products and services.
Here we again explore the strategic opportunity for an organisation to acquire personal data as part of an exchange process, as discussed in the last section, but for products based on the data provided. In other words, the user expects to have to reveal more personal information, and the data quality and quantity can be expected to be higher. If permission is required to use the data for other purposes it is more likely that this will be granted. Therefore, creating a need for a product based on personal data is a strategic opportunity to obtain good data that can be more widely exploited, not only in improving the business by enriching the database, but also in developing new products.
Services related to data that users will give personal details for include:
Most of these exchange processes are not seen as an exchange of data; the user just focuses on getting what they want. They accept the need, for example, to give details about lifestyle, health, smoking and drinking behaviour to get cheaper insurance. These concepts are not new but the newness is that organisations are becoming wise to the benefits of using the data to understand the customer better and avoid risks. For example, a car insurance company may offer cheaper insurance if the customer states that they don’t drive very far; now this statement can be checked. In addition. the company can make access to personal data and the more passive tracking data a condition of selling the insurance, thereby reducing their risks and enabling them to offer cheaper insurance premiums and attract more customers.
The data exchange can be used to develop new business models. For example, an insurance organisation may give cashback every year if the applicant reaches goals such as occupying their house at least 75% of days, and allows their data to be scrutinised. Similarly, banks may allow more flexibility on credit conditions if the applicant’s behaviour conforms to the bank’s conditions and allows real‐time access to their data records as evidence. Early adopters of these ideas in the banking world have benefitted from making the acquisition of loans as easy as doing online shopping for people who are happy to share their personal data.
We often agree to share our locations in exchange for services such as being shown the closest petrol station or swimming pool or hotel, or to find the optimum route from A to B. This information secures the immediate service for us but also means that we passively reveal a lot about our habits and activities – data than can and will be used more widely.
Here we explore the strategic opportunity for an organisation to acquire data in a business‐to‐business (B2B) environment and the way data can be utilised under the new paradigm of monetisation. Very often, B2B implies being part of a supply chain: companies must exchange data within the chain to be allowed to be part of it and withholding data is likely to make them unwelcome and may exclude them from doing this kind of business. The supplier may have to share their production data stating how their process performed as well as giving their accounting data showing their financial position. In many other business relationships, data exchange is part of the normal business process.
The strategic opportunity is for each organisation to make more use of the exchanged data to optimise and automate their production processes. An early example of this was just in time (JIT) management, in which a close relationship between trusted suppliers meant that virtually no stock was kept because each delivery was made just as it was needed. This JIT agreement necessitates a transparent interchange of data and gives enormous benefits when it works well. Another example is the exchange of delivery information from the postal service to the supplier, allowing the supplier accurately to predict deliveries and give a better service to the customer.
Here we explore the strategic opportunity for an organisation to acquire data through market research technologies. People can be accessed in a range of environments, including websites, trade fairs, on the street, in a laboratory or a testing suite, via their email or their phone. A lot of organisations already use market research, but the management and utilisation of the resulting data is often quite poor; the data is not seen as valuable in its own right. In fact, it is tremendously important, giving valuable insights about customers and non‐customers. Deeper analysis using predictive analytics and pattern detection methods is worthwhile, but few companies go further than just descriptive reporting.
It is useful to know that people are mostly willing to participate in market research and share their views and ideas; they give valuable data in exchange for relatively little, say a discount or membership or a voucher for a special event. More should be made of their donation of information.
Here we explore the strategic opportunity for an organisation to acquire data by taking advantage of people’s enthusiasm to have their say. We note that Maslow’s hierarchy of needs (see Figure 7.5) is relevant in motivating people to part with their personal data. At the bottom level, the aim is to get basic living requirements of food and shelter. So, for example, people give information about themselves in return for discounts or as part of more substantial transactions, such as mortgages, that will improve their lives. These exchanges have been around for many years but with the advent of fast data exchange and cheap storage they are now being optimised to give greater business advantage.
Improved technology has enabled satisfaction of a higher level of Maslow’s hierarchy: the aim for love, esteem and a sense of belonging. These aims present a strategic opportunity for many businesses and we find many that have exploited it successfully. These organisations have made good business and have also gained from using the data.
Dating agencies are a typical example: people give their details and hope to meet someone they like; they are happy to accept matching scores calculated by algorithms to find a suitable partner. From the business point of view the core business of the agency is collecting data, analysing it to find patterns and making predictions of potential matches. They are paid for this service by the participants. From the consumer point of view, these sites attempt to satisfy the desire for love. Note that the business must balance the value of a good reputation for making successful matches with the advantage of keeping people on their records for a long time while being paid for it!
There are many different business models for esteem, and all are based on data; payment can come from the participants, from targeted advertisements or from selling the data (information brokering) or a combination of these. For example, there could be a fixed up‐front fee, or different fee levels for more exclusivity, privacy and security. In most cases the fee is only part of the income; targeted advertising is also lucrative.
Another very common business model on the web is running a review portal, where people offer their services for free in return for having influence and feeling important – although there can also be benefits for frequent reviewers. Other people, looking at the site, gain valuable guidance. Interaction with the portal generates data and a population of people who can be targeted for selling products via adverts and banners. The site could also act as an information broker, for example by selling the email addresses of reviewers to interested parties. These will be used for predictive analytics, customer segmentation and rule‐based analysis. The main benefit for those writing the reviews is the esteem and sense of belonging. There are reduced risks and a sense of security for those reading them; for all these benefits all are willing to pay with their data.
Here we consider data as a resource. A mass of data is generated from advanced connectivity, in which people use multiple digital devices and spend a lot of time on the internet managing their daily life and business. This data arising from changing behaviour contains pointers and indicators about the needs, wants and demands of those people. It also leads to innovation and completely new products based on this valuable resource.
Algorithms now take the place of the many expert employees who used to deal with our business interactions. The algorithms must be designed to ensure that information is collected at a similarly high quality as was obtained through personal interaction. Humans take account of cultural and background knowledge in asking the questions that will elicit the customer’s requirements. They also have a feel for how things are likely to progress in the near future. Skilful design is informed by prior analysis of historic data enriched with knowledge and experience from the previous way processes were carried out.
With modern automation, it can be expected that the appropriate analysis comes up with the right recommendations and decisions. This automation depends on past and current data. As data is fundamental to all these stages, it has been said that ‘data is the new oil’.
In earlier days, data was collected only for administrative purposes: where to file information on people, products, stock and invoices, and retain background knowledge about them. The data was usually collected as part of a personal interaction. The tendency was to take the minimum data required, because the collection process was time consuming and storage was expensive.
Often the data was stored on paper and was expensive to access and archive. Even with the advent of the digital office, data storage was still expensive. Since the late 1960s, mainframe computers have enabled companies to store more data and to use the data efficiently to handle their customers. The data was used for reports and administration. Until relatively recently, storing a gigabyte of data cost more than storing a terabyte of data now. Technology is changing and prices for hardware and software have tumbled. Costs are no longer a barrier to data use.
The sea change now is that data is often collected automatically, so there is no limit to what can be asked for and stored. For example, locations, movements, interchanges with others on social media, video footage and sound recordings are all types of information that can be gathered. Increasingly the data becomes a resource on its own merit. It forms part of business (or government) information and can be used for any purpose. That has to be kept in mind. On the positive side, the data resource can be used for improving the processes, products and services of the institutions that assembled the data, but it can also be used for other less savoury purposes.
Data has become a true resource, used to design and develop new services and products unrelated to the original business. As an example, professional report providers such as bespoke trend monitors and product testing companies have a B2B relationship when they sell their reports to other businesses. They already make use of data as a resource because they buy or produce raw data as the first link in the supply chain. They then sell on the augmented and interpreted data to businesses as the second link in the supply chain and these businesses sell on to people who are the third member of the supply chain.
Figure 7.6 shows three typical stores of consumer data from a company point of view:
Note that data analytics is required to get the maximum benefit from all the data sources.
We can explore some of the ways in which the raw data goes on to be used for further interpretation and analytics as a first step in the monetisation process. Recall that this data is naturally historic, as the moment it is recorded it becomes old, representing a past event, even if one that happened only a few seconds ago. The static nature of the data has to be taken into account in any descriptive or predictive models.
Nowadays we can measure everything. The question for the future is what we will do with all these measurements. Increasingly people are embracing technology in their private life to answer questions, adjust their home environments and procure services, all just by speaking to a machine. Applications are being developed in industry 4.0, the fourth big revolution in mechanisation (see Figure 10.46), to emulate the complex stimuli that alert a human being to a failure or a problem using an algorithm embedded in a robot instead.
The tumultuous change in all businesses as they rush to gather more and more consumer information provides exciting opportunities. Business informatics gives immediate insight into any area of interest about the customers: their behaviour, location or interests, as shown in Figure 7.7.
As a cautionary note, the increasing mass of data may also start to give a negative feel, as people become worried about their lack of anonymity and privacy. However, the move towards increasing digital capture of every aspect of life is a powerful force and there is much positive impact. An exciting challenge for the future is to find more new and unexpected interactions, relationships and meaningful combinations of the measurements to create new knowledge and new ways of doing things.
Simple monetisation is when data that is generated is sold on as a resource; usually the data broker will have an overhead of quality checking to meet agreed standards. Examples include stock‐market price streaming, official statistics and meteorological data. Often this data is freely available on websites, in newspapers and elsewhere, but to get easy fast access, especially to past data, you have to pay.
The next level is when companies collect raw data from different sources and combine it. They check its quality and authenticity and sell or rent the integrated data to people interested in specific market areas, for example gardening. Another example is monitoring the web by capturing the occurrence of selected keywords in a chosen area and time slot, reporting the data and selling the reports. The reports need to be attractive and the prevalence of keywords and their associations can be nicely depicted in a word cloud. An example is shown in Figure 7.8
The next level is when the data is collated, described, explored and reported. Data may be reshaped, enriched and grouped and then put into saleable form. Examples include address lists and reports that people pay for because of the insights they contain, such as labour force statistics. These pool information from employment agencies, newspapers, companies and other sources, with the combined information all sold on together. They constitute a powerful resource, showing which jobs are becoming increasingly available, which ones are advertised repeatedly, where the jobs are located, the types of industry sector which have jobs to fill and many other interesting facets of the employment scene.
Enrichment can be carried out at an individual level or at a more general level. Gender and age estimates for an individual can be derived from intelligent interpretation of names. Post code information can be enriched by adding income level, employment rate, commuting behaviour, and so on. This information is available from open or paid‐for sources of information. These general estimates can be converted to probabilities for individuals, for example likely employment status. Examples of data handled in this way include service usage reports, testing reports and essays on industry areas.
A major growth area for data scientists is creating businesses that analyse data to extract deep insights from it. They use the sort of basic and sophisticated data mining and statistical methods described above. The resulting analyses are then sold on or rented to anyone willing and able to pay for them. Increased usage of statistical analysis can add value to raw data and this is the most powerful tool to get more money out of it.
For example, for a financial company, the raw data is detailed information, on a personal level, about the person requesting a loan: their name, address, age, income, gender, family status, property and so on. Predictive models are built with techniques such as decision tree analysis or neural nets trained on older data. They include variables for different types of people and their likelihood of defaulting or being late payers, give probabilities of defaulting or paying late and can be used to make decisions. This risk assessment can be done by the company itself or the risk estimation process can be outsourced to a specialised service company who gather data, often from wider sources, build the models and produce the risk estimates as a service.
This is the difference between the raw input and the finished product. The strategic opportunity for a company is to take advantage of appropriate data analyses to find clusters and useful patterns in the data and make the transformation from the raw data to decision‐support indicators. Indicators are a useful way to delegate decisions to people with limited statistical knowledge because the skilful amalgamation of data has been carried out already.
The largest growth area for organisations with a large stock of their own data is in combining information, often from disparate sources and extracting innovative, unexpected insights. This process needs a lot of calculation power and complex analytics. The aim is to have full prediction of individual users at any time and at any outlet, for example store, internet outlet or communication opportunity. The aim of using enriched customer data is to tip the money balance in the company’s direction although the client thinks the company is only interested in giving them the best service.
Many institutions are becoming wise to the insights hidden in enriched and integrated data, and the implications are far reaching. Social security payments for people who are sick and unable to work can be queried if the claimant’s credit card is found to have been used for extreme sports or in exotic holiday locations. Pharmacies who service a community with an exceptional need for subsidised spectacles as compared with the countrywide average can be questioned. Banks identify outstanding or irregular patterns of behaviour and use them to help in fraud detection. Data analytics can derive the usual pattern of purchases presented at a supermarket till so that if the pattern is not followed it alerts the shop to possible wrong doing. In the shipping sector, fuel flow sensors are installed and data analytics show expected fuel consumption for a journey; significant deviations alert the company to possible fuel theft.
Combining location and personal data, near‐field communication via smartphones is becoming more common. As you pass an outlet, a voucher appears on your phone for an offer which is specific to you because the analytics have matched you to the products and shops you are passing. In this way, the predictive model detects your latent or subconscious desires.
Complex profiles may be built from past web‐surfing behaviour, as captured in each unique cookie on a person’s computer or device. Regulations differ in different countries as regards the way people can be identified. The internet protocol (IP) address is associated with the internet provider a person uses, but several people can use the same one. Currently in some countries it is forbidden to store IP addresses of a website user because it is too personal and data protection prevents it.
Note that there will be different cookies on each device the person uses (iPad, smartphone and so on) and the user may delete the cookies periodically. This type of tracking therefore has many uncertainties but is nevertheless found to be very useful for companies that are aware of data‐analysis opportunities. Combining website tracking data with data from other sources, however, presents a major technical issue. There is also a communication barrier, because permission needs to be granted for each data source. The tracking data cannot be combined unless a technical link is agreed, for example a single sign‐on or other sign‐on technology that makes it possible to bring the cookies together.
Companies may track a network of websites to see who is using them and what they are searching for. For example, when people search for information about transport and city events on specific dates, then they can be targeted with advertisements for accommodation. If the web‐surfing history data can be enriched with other data relating to the consumer then more powerful models can be built and used to decide which content and/or banner to show next.
Strategically, the issue is firstly whether an organisation can identify appropriate data to enrich their own data and secondly whether they can access and use technologies to match the data sources on the level of the individual or household. The simplest approach is to enrich customer addresses with additional information about the customer or the area in which they live. Cluster analysis of customers, for example, can then proceed with more subtlety than clustering on just the address. Depending on the aim, the organisation can produce and deliver targeted, personalised advertisements, provide a better service or explore opportunities for developing new products.
Figure 7.9 shows the potential usage of combined data: increasing lead conversion, increasing number of new customers, creating better offers for existing customers, including behavioural pricing and preventing churn.
Companies in the information brokering business, for example social networks, mail order houses, address dealers and publishers, may allow you, if you pay enough and the law allows it, to rent their information and use their targeting facilities to communicate with potential customers.
As sport is so popular, and football in particular, another group of companies have built their business on allowing users to rate current and upcoming football stars and create fantasy football teams; money comes into the service provider from advertising specifically targeted at the user on the basis of their football preference. This kind of entertainment business is also successful when based on movie stars; users can construct their fantasy movies and the service provider can target them on the basis of their choices, which tend to put them into a characteristic segment. In fact, the same concept is involved with playlists because the playlist a user chooses says a lot about them and this is exploited for targeted marketing. These things are happening already and more opportunities will come along soon. These business exemplars show that data has a serious impact on everyday life and is already empowering business.
The Maslow triangle of needs shown in Figure 7.5 can also be applied in developing business opportunities. The physiological and security level may correspond to businesses monetising data purely to make money to allow the owners and employees to live a more comfortable, secure life. The belonging and esteem levels may correspond to demonstrating their business is valuable and exclusive. Finally, the self‐actualisation level may correspond to more altruistic activities such as businesses using data for the public good, to show inequality or prove damage to the earth, animal kingdom or soil quality, and motivate people and governments to do better.
What are the drivers and motivation for change? All businesses are under pressure from increased competition. Enthusiasm for innovation makes it vital to achieve faster production times and turnaround of new products. This might imply the need for new partnerships and unusual alliances between companies. Customers expect a fast, proactive response when their needs are recognised.
The modern high‐speed world makes faster error detection and quality control very important. Traditional ways of error detection and quality control might now be too slow and cost too much money. Companies need suitable processes to stay in business and this might mean embracing new technologies and they may also need to find alternative sources of finance.
A major motivation is in either encouraging or resisting the new ‘Big Brother’ society.
Business reports such as product reviews and insight reports based on exclusive data‐driven knowledge bring in money in a number of ways:
Sometimes reports are delivered on different payment levels. The basic level may be free, although actually paid for by providing personal data (supplying your name, email and contact address, for example). The next level requires actual monetary payment, for which you get a deeper report.
More indirect payment arises when the company writes a report and is compensated by selling advertising, for example when an employment summary is paid for by advertisments from recruitment agencies. Alternatively, the report may consist of articles written by the recruitment agencies who pay to have their articles included in the report.
Individuals and companies can buy additional knowledge by paying for reports or subscribing to information summaries. The authors combine and summarise data from different sources, including publicly available reports, company reports, company accounts and market research information. The providers offer subscriptions: the more you pay, the deeper your access. The unique selling point is that it saves time, you need no skill to sift through the data and somebody else has tapped into the data sources, some of which may not be easy to access.
How does the company know what to charge? Initially they offer access for free, a so‐called ‘freemium’ level. Then, when they become well known, they introduce basic and advanced fees. They monitor their performance to see what is most popular, then produce new products at that level. They adjust their pricing according to the take‐up: the more popular the product and the more people who know about the company, the more valuable the business.
Data from government statistical services might be the food for your business and if you are careful and create a user‐friendly interface then the processed data can be a valuable product, which is profitable for your company. It can also be a trigger to earn money through advertising and encouraging web traffic. As an example, Zoopla in the UK has made a highly successful business built originally from making land registry data of house sale prices readily accessible for consumers. This information is now enriched with much additional public data to give a multi‐functional real estate platform. As another example, Statista in Germany (and worldwide) is a multifaceted statistics portal built originally on publicly available statistics about business. This is now enriched with data from market research and a variety of other sources. These easily accessible and comprehensive platforms give users information and confidence to function in even the most complicated market areas; they are based on integrating public and other data, representing a new way for data to pay for itself.
Price can be the trigger that makes a customer buy, but it may not be the main trigger for some customers. We have to find what reduction is necessary to tempt these other customers, otherwise we will lose money. It may be that no reduction is necessary at all; a free cup of coffee may be as effective in luring the customer into the purchasing process as a €100 voucher. It all depends on the individuality of the potential customer.
Data and analytics can be used to find the ideal price range that just triggers a purchase, or the optimal level of offers that will attract the person to buy. The pricing is based on recognising the customer as belonging to a certain behavioural group whose tendencies are well known from observation and historic data. This is a very important methodology as it avoids selling too cheaply and also attracts more customers. The analytics mirrors the sort of thought patterns undertaken by a successful salesperson. In the learning phase, it is vital to test different data‐driven pricing strategies through focus groups, surveys of volunteers, including colleagues, and a review of as much relevant historic data as possible. This is the way to establish the guidelines. A more scientific approach uses designed experiments to test the importance of different influential factors. The resulting rules are validated by comparing the predicted with the actual purchase behaviour.
People have an interest in health and wellbeing; products have been developed to exploit this interest and collect personal data on a massive scale. For example, apps and other devices help people to monitor their exercise, food consumption, sense of optimism, blood pressure and hours of sleep. Figure 7.10 shows an example. The patterns in this data offer valuable insights. They can be used to recommend and tailor products in response to how people might feel when they see their physiological and activity patterns but have no deep knowledge in the health profession. Based on this data, companies might recommend products from their portfolio, for example a new bike if the company sells bicycles, a holiday if it is a travel agent, a new mattress if it is a bedding company.
Medical supply companies may also offer over‐the‐counter products, such as vitamin pills and self‐test kits, and make a lot of sales through the formidable combination of insight from the data, peoples’ fears and their lack of medical and professional knowledge. Medical doctors might disagree that these are the best ways to improve health or cure ailments, but people will still be inclined to try them!
In future, insurance companies will start to use not only personal measurement data but also to require that people collect it and share it if they want to be considered for the best (cheapest) offers. This approach has already been adopted for car insurance in some countries, where insurance companies know the statistics for car theft depending on your age, gender and occupation, where you live, whether you have a garage or not, and if so, how often you use it. This knowledge is augmented with data from connected car systems or data delivered by your smartphone, which shows how you drive, where and when you travel and where the car spends the night.
It may be an advantage in getting better offers to be on social networks: the insurance company may be reassured that you conform in some sense. The insurance company can therefore assess your risk more precisely. They can give lower premiums to some customers and higher premiums to others and make more profit because they can identify the risks with more confidence.
Insurance companies are embracing data analytics with great enthusiasm because it has proven highly profitable for them. See Figure 7.11.
Information brokering is the practice of buying, selling and exchanging information obtained from data. For example, one company may collect and hold big data on individual consumer habits. Another company may have a good business idea for a new product. For a fee, the big data company will attempt to identify the best potential customers and provide a list of contacts. For example, the information broker will match a company selling swimming pools with people who are likely to buy a swimming pool and rent out the use of suitable contact addresses. If the information brokering company already has direct access to these potential customers through their core business then they could offer to use their own facilities to deliver the other company’s offers to those who are likely to have an affinity for them. These offers could be delivered as online advertising, a letter included in a newspaper or a company brochure mail‐out, or in a parcel or with an invoice or via any of the many communication opportunities.
Brokering is standard practice for companies that deal with credit scores. The credit scores are composed from data from many different sources, including past buying and paying behaviour on multiple occasions on line, via mail order houses, banking habits (whether people paid back regularly), how they handle smartphone, debt habits, if they live in a region where many people have money problems, age, job, even first name. Public sources of data, such as property buying behaviour and selling prices and health or employment statistics, are also useful even if they are in aggregate for the area the person lives in rather than being known for the individual. All this data is compiled by the credit scoring company; they then sell the credit scores to whoever pays for them. Credit scores may be requested for basic things like renting a flat, applying for a mobile phone contract, opening a bank account or applying for personal and business loans. The credit score can affect the interest rates demanded by the bank or lender.
Information brokering is the basic idea behind the money‐generating side of social networks; they can sell information about customer habits, customer insights and segmentation. Their level of sales helps social networks determine what fees they can charge for different types of personalised advertisements. Social networks are particularly fortunate in that they have no intermediaries because they hold the data, analyse it and also deliver the advertisement to the relevant individuals.
Increased connectivity of the emerging data‐driven society is a continual source of opportunity for advancing a business. As more transactions are carried out on smartphones it becomes sensible to include that technology in your organisation’s offering.
Every organisation should think how they can create digital smart products or at least start using apps. There are many exciting ideas, for example:
The so called ‘digital kitchen’ is an assistive technology aimed at helping people stay in their own homes as they get older; this is a kitchen which is wired up with sensors and cameras to capture movement and predict the occupant’s intentions when entering the kitchen; it can suggest what to cook based on the current contents of the refrigerator and cupboards.
Using connectivity strategically not only utilises data but also gathers more data, so it is part of a never‐ending cycle of creativity. The success of this cycle depends upon whether you can connect all the data. Associated with this issue is the importance of deciding on your focus of interest; is it to estimate the household behaviour rather than that of individuals, or is it in the output from a particular device? If you are interested in a particular device then you need to be able to identify who is using the device and integrate all data belonging to that device. Most businesses are well‐advised to think creatively how to use these smart possibilities.
The monetisation of data necessarily involves a careful and thorough process of problem selection and definition and determination of a methodology for finding solutions. Many different problem analysis approaches are available; some of these have attractive names and forceful sales people to promote them. For example, the Six Sigma approach introduced in Section 2.4 is to step through stages for define, measure, analyse, improve and control. On the other hand, many companies prefer to create their own tailored methods. Whichever problem‐solving methodology is selected it is likely to add value because it forces a focus on preparation and lateral thinking, reviewing the problem in all its facets before jumping into action.
In monetisation, there are two angles: identifying problems needing to be solved and deciding how to use data to detect and solve problems. Both of these angles revolve around data, although identifying problems is usually aligned with overall company strategy. The general data mining process described in Section 4.2 gives guidelines on ensuring a thorough and creative approach to using data to detect and solve problems. In particular, the ‘Business’ task in Stage 1 involves clarification of the business issues and can be facilitated using problem‐solving techniques such as what, where, when, who, why and how, often referred to as ‘the Kipling 6’.
Problem solving is a cyclical and creative process because we need to decide how to use the data we have and whether modifications are needed or totally new sources have to be commissioned. The cost of new data needs to be justified by the initial findings of the problem solving process; stakeholders need to be enlisted in good time and the breadth of the data requirements needs to be established. The data analytics may present novel solutions to the main problem but also detect new problems not currently in focus. Hence the cyclical nature of problem solving.