How to grow your business with analytics
Analytics, and especially predictive analytics, is a very effective tool to increase revenue, operational results, and decrease costs. The reason is that it improves the business IQ. It brings facts to the decision-making process and limits the amount of guesswork. The diagram below shows the top business areas where predictive analytics are implemented.
Source: Wayne Eckerson’s “Predictive Analytics: Extending the value of your data warehousing investments”, the Data Warehousing Institute 2007. Survey based on 167 respondents on which areas they have implemented predictive analytics.
I will now list the main analytics tools that help you grow your business. One of the areas that greatly benefits from predictive analytics is cross- and up-sell. It is the best place to start, because it is easy to measure and quantify the outcomes. There are many different models used to increase sales. What these models have in common is their use of data mining and statistical techniques. These techniques find the customers that are most likely to purchase, need, or use a particular product. Analytics helps you increase the relevance of your communication for each and every customer. It should therefore be implemented on a large scale, meaning for every customer interaction. It limits communication fatigue by reducing the amount of irrelevant messages. Analytics can help you detect which communication channel the customer prefers. It can also increase the excellence of the customer experience by interpreting events as they occur in the customer’s life. Acting on these events will enable you to help your customers when they need it. Analytics is also useful for reducing risks. Credit scoring in the banking industry is a good example of that. Moreover, you can optimize revenue by creating pricing models. These models are used to customize the price for each customer. Matching the price with the customer’s profile and risk helps in optimizing revenue. The insurance sector has come far in this area.
Churn prediction is another way of rapidly growing your business. It is common knowledge that acquiring new customers costs 5 times more than retaining the customers that you already have. Churn analytics can help you predict which customers are most likely to leave your company in the near future. It can also predict which offer the customer is most likely to respond to positively. By proactively contacting them before they leave, you will be able to prevent many of them from leaving your company. Win-back prediction is another tool that improves business IQ by selecting the customers that are most likely to come back after they have left your company. The same techniques can also be applied to help your company retain the employees in danger of leaving (and that you want to keep).
Descriptive analytics looks at what has happened in your business, and it is a potent tool for quality improvement. Let’s look at the new customer acquisition process. Improving its quality requires having access to the contact data left by potential customers through different channels. Analyzing the data will help you see which acquisition processes are effective in converting potential customers into real customers. Descriptive analytics can also help you find and target potentially highly profitable customers among your passive customers. Since passive customers usually have little data connected to them, this procedure will require additional data to that commonly found in your in-house databases. These data are usually provided by your national statistics agency or similar.
These are examples of the most common cross-industry analytics tools that can most rapidly affect your business results and help your business grow. There are many other industry specific usages of analytics. The different sectors’ best practices often inspire innovation across industries. In banking and insurance, analytics is used in fraud detection, pricing optimization and budget planning. In retail, it is used to micro-segment the customers, to optimize the stock in warehouses, correctly place products on the shelves, and predict which product launch will succeed. In the postal service, predictive analytics helps match the need for extra workforce with the number of incoming letters. In telecommunication, the use of big data analytics from smart phone devices and social networking provides valuable insights into customer behavior, preferences, and influence. This enables communication service providers to improve customer experience, to solve real-time service performance issues, and increase the intelligence behind sells. The more significant data you use in predictive analytics, the more business IQ you will acquire.
Implementing analytics is the responsibility of the business
The CIO Insight survey shows that only 43% of technology’s business potential is realized. According to Gartner, 70% to 80% of all analytics projects fail. Why? If you ask a business person about
who is responsible for implementing analytics, a vast majority will naturally say it is IT or the technology provider. Many seem to believe that the technology is the most significant part of the implementation. If you have read my previous blog, “What is Business Analytics?”, you will see that technology and its supplier is just 1 out of 5 components for implementing analytics. The remaining components (i.e. data, statistical methods, security, and business processes) must be the responsibility of the business. If any of these components are missing, it will compromise the success of your implementation. Besides, a technology supplier will never have your level of understanding about your business and your customers. The business IQ will come from many different groups within the organization and all of them have to cooperate. For technology to deliver on its promises, the business has to increase its level of understanding on all the analytics components.
To secure an optimal implementation of analytics across the company, it is necessary to have an analytics implementation strategy. This means understanding analytics and its usage first, and secondly, finding its place within the business operating model. The results from the Gartner survey indicate that the analytics strategy is missing. This will have consequences for the results of implementation.
Increase credibility with analytics
In this economy, the marketing department will no longer get away with not measuring the return on investment of their marketing spending. According to McKinsey, the marketing budgets worldwide total an estimated trillion dollars in annual expenses. They also say that only 36% of CMO (Chief Marketing Officers) have successfully used analytics to measure the outcomes of marketing activities. There is a need to put hard numbers behind marketing performance. The results and sustainability of your company are at risk if gut feelings decide how the business should be run. Decisions should be challenged by facts.
Fact ambassadors can be a little unpopular at times because they bring accountability and transparency to areas that previously were unchecked. Measuring outcomes can be scary, but it is a necessary step to improve the business. How can we get better if we do not analyze our performances? Change is always a little scary, but is a small price to pay if we want to perform better and achieve success. Continuous improvements are achieved by doing a little bit better every time with the help of facts.
Many players like Netflix, Amazon, Google, Capital One, Tesco, FedEx and others have actively used analytics. They are at the top of their game. Analytics performs regardless of which industry you are operating in, which process you want to improve, or which customers you are doing business with. There are many roads to success, and bringing more analytical IQ into your business processes is often the fastest one.
Experienced employees that know their business very well, have a tendency to say that analytics can never replace their experience-based judgment. I understand the reticence to let analysts bring facts to the decision table. It unsettles the governance model, the power balance and the status quo. The emergence of analytics can create conflicts between different groups within your organization. You have to revise the governance model and install a fact based decision-making culture through a change management process.
Many also say that analytics, and especially predictive analytics produce inadequate results and therefore should not be used. It is correct that predictive analytics can never predict the future with 100% accuracy. The accuracy of predictions may vary a lot: from 80% accuracy to 0%. The most accurate models I have seen have access to a lot of significant data. Sometimes predictions don’t work. This has happened to me a few times, especially when the market was over-penetrated by a particular product. However, the vast majority of predictions work, and with a level of accuracy that far surpasses that of guessing and gut feelings. So why not use it?
If you are interesting in learning more about analytics, I recommend the following: https://analyticsbooks.blogspot.no
I wish you the best with your analytical ambitions.