The three challenges you need to overcome to make your Big Data initiative a success

Anyone who has spent the last ten years or more providing data management solutions will be looking at the hype surrounding Big Data with a mixture of pleasure and alarm. Happy that data – typically the unloved orphan in IT – is receiving long overdue attention. But concern that the hype is creating expectations that will be difficult to meet.

If past experience with hyped technologies is anything to go by, price the vast majority of Big Data initiatives will not generate the financial returns envisaged in their supporting business cases. As was the case with CRM, stuff much of the benefit is forecast to come from revenue gains and there just isn’t enough expenditure slack for all the expected gains to be achieved.

There will of course be winners and losers. Critical to success will be avoiding the three following traps, each of which is likely to result in a Big Data initiative failing to meet its stated objectives.

1. Failing to value the data already being generated

The irony of the excitement around Big Data is that most companies don’t take ‘small’ data seriously. It is also the case that many of those that do, like banks and insurance companies, are often compelled to by regulatory requirements. Very few invest voluntarily in maintaining and managing the data they are already generating via their operational systems.

‘Small’ is a complete misnomer as there are huge volumes of data, but it emphasises the contrast with Big Data. A better counterpoint might be to call it ugly data (with Big Data being Beautiful Data) as it is generated by operational silos that are completely detached or, even worse, capture the same event in different ways, so relatedness is hard to spot.

Technologists have traditionally viewed data as unexciting compared to system functionality. Lacking the support of enthusiasts,

data management is often overlooked when IT budgets are allocated. As a result data remain inaccurate and incomplete; unreconciled and inconsistent; aggregated rather than granular; delayed rather than real-time and value-limiting rather than value-creating.

Big Data is at least changing priorities. But if a business is not able to manage its existing operational data effectively, the chances of coping with the variety, velocity and volume of Big Data are slight. Data is either valued or it isn’t and if there is a culture of it not being valued, jumping on the Big Data bandwagon will not change anything

So before embarking on a Big Data initiative, the relevant questions are:

  • Does a senior business (rather than IT) executive have responsibility for data in general and data quality in particular?
  • Is there a data stewardship team (reporting to the executive responsible for data quality) whose responsibility it is to ensure that data is accurate, complete, consistent, timely, etc.? Do clearly articulated and relevant metrics exist to assess the performance of this team?
  • Are there job descriptions for these roles that clearly describe the competencies required to be an effective data steward?
  • Are there clearly documented data management principles, processes (including process controls) and procedures in place?
  • Are these processes automated or semi-automated and underpinned by a suitable data management technology?
  • Is there a regularly published dashboard of data quality metrics? Does performance on data quality metrics impact the sponsor’s remuneration?

If you have answered ‘No’ more often than ‘Yes’, the likelihood is that your organisation lacks the data management foundations to capitalise on the opportunities that Big Data offers.

One example of how Big Data hype has blinded people to the value in small data is text analytics. Since the advent of operational CRM systems, customer service representatives have dutifully recorded customer complaints and comments about the service they have received. These comments are a goldmine for any business seeking to understand what customers really want and how their experience can be improved. But for the most part, this data has been ignored.

With the eruption in social media usage, the interest in text analytics solutions has increased significantly. The irony is that the volume of insight-generating comments that trawling Twitter or Facebook will yield is a tiny fraction of those that businesses already have access to. Only a tiny fraction of people go on social media sites to discuss the products and services they buy. Also when they do, in general it will only be about a small sub-set of what they buy (Apple products, for example). But the one time that customers do want to discuss the service they are receiving and how it could be improved is when they speak to representatives of the company. But, hey, that is just small data.

2. Business value is obscured by technology hype

The term Big Data is an

umbrella term for an ill-fitting combination of different types of data. The three sources that are most frequently mentioned are:

  • The unstructured data emanating from social media sites such as Twitter, Facebook and YouTube
  • The trail of locational data emitted by smartphone usage (primarily GPS related but also cell usage-driven)
  • The data stemming from the ‘Internet of Things’ – the networked sensors embedded in physical objects to communicate location, enable transactions and monitor performance

The data generated are diverse in form and usage, yet the name Big Data implies singularity (or similarity at the very least). What binds them together is a high level IT need for data storage, access and processing.

At its core, Big Data is a technology-driven concept rather than a business-driven one. That needs to be inverted. That will probably require a change in nomenclature (the first rule of Big Data initiatives could be don’t talk about Big Data). The focus needs to be more specific and business value-oriented. If you are developing a Big Data strategy or implementing a Big Data initiative, you are less likely to be successful than if, for example, you are implementing an initiative to enable the analysis of unstructured data in text form to improve the customer experience. The first step in this process is evaluating which data elements will provide the greatest opportunity for value addition and prioritising accordingly.

So the next questions you need to answer are:

  • Does the responsibility for Big Data sit with IT rather than a business team?
  • Are you progressing a Big Data initiative rather than more specific and focused initiatives around its component parts?
  • Related to this, are conversations about Big Data predominantly about technology solutions?

Again, if you have answered ‘Yes’ more than ‘No’, the chance of business value realisation is much reduced.

3. The focus is on selling rather than serving

There is a popular marketing delusion that customers are out there just waiting to buy more of what you want to sell them. All you need to do is tell them how they can do so or give them a promotional offer and Bingo. This limited view of marketing reduces it to its sub-component marketing communication. Worse it misses marketing’s most critical function – creating value for customers.

Marketing needs to balance customer interests with shareholder interests. The former are served by creating compelling value propositions – ones that offer customers greater value than those provided by competitors. At the same time marketers need to extract value by making sure that offerings support corporate objectives for customer acquisition, retention, growth and profitability. You have to create value in order to extract it, or put another way – to create value for shareholders you first need to create value for customers.

While most marketers would pay lip service to this, the actions of most marketing functions tell a different story, particularly when it comes to IT investments. Take CRM for example. The vision for CRM was that it would improve the customer experience. The reality is that company interests ranked ahead of those of customers, with a focus on making customer-facing processes more efficient rather than more customer-friendly. Data integration has focused on identifying cross-selling opportunities rather than customer convenience. As a result we are now deluged with unsolicited direct mail, email spam, text messages and calls to both mobiles and home numbers. Similarly, despite the fact that social media is seen as a means for having a dialogue with customers that would result in a better understanding of their needs, traditional monologue or broadcasting approaches have just been transported to new channels. Social media sites have become yet another place for us to be stalked.

With all these initiatives, the perspective of the customer and the need to create value before seeking to extract it has been lost. Not surprisingly satisfaction ratings have remained stubbornly low as have returns on technology investments.

The risk is that Big Data investments will follow the same pattern. The majority of articles on the subject of Big Data focus on how it can be used to sell more to customers. The unedifying prospect is that our smartphones will increasingly be flooded with location-based offers.

The shame of it is that the rich seams of location-based data offer significant opportunities to save customers time and money and create value for them in other ways. Similarly text analytics can provide businesses with the opportunity to really understand what customers want. Too often customer satisfaction programmes have concentrated on the quantitative rather than the qualitative because the former is much easier to process. That is no longer the case and text analytics tools can help close the feedback loop. But this requires a focus on serving customers in the first instance – improving service to improve sales. And if prior experience is anything to go by, such an attitude is unlikely to prevail.

4. How to maximise the chances of success with your Big Data initiatives

So what does that mean for a business seeking to embark on a Big Data journey? Based on Sopra Group’s experience in data management over the past 15 years, we would suggest the following:

  • Avoid being herded by the hype into a rushed and ill-conceived implementation. Review how effectively previously hyped technologies delivered their purported benefits in your organisation
  • Ensure that the correct data management foundations are in place:
    • embed the importance of all data (small and big) in the corporate culture
    • establish effective data governance
    • recruit data stewards
    • document data quality policies, processes and procedures
    • automate and control data management with a relevant technology
    • track data quality metrics and report data quality performance to a senior business sponsor
    • Integrate data management activities – avoid the trap of double data standards, for example by investing significantly in Big Data technologies while operational data is inaccurate, incomplete and inconsistent
    • Focus on serving rather than selling – prioritise the generation of insights that will enable the value created for customers to be increased, sales will follow
    • Crawl, walk and then run – break Big Data into its component parts and select those that offer the greatest opportunity for value enhancement, then experiment and prove that investment will deliver
    • Ensure that Big Data initiatives are business- rather than IT-led. Focus on uncovering and proving the business benefits rather than focusing on the comparative performance of different Big Data technologies

Focusing on theses six actions will significantly increase your chances of success.

Share the knowledge:
  • Twitter
  • Tumblr
  • Digg
  • StumbleUpon
  • Reddit
  • Slashdot
  • Facebook
  • LinkedIn
  • Yahoo! Buzz
  • Google Buzz
  • Google Bookmarks
  • Print

Related Posts:

About Jack Springman

I am a consultant with experience in business strategy and customer strategy development, customer management and customer service transformation.

Leave a Comment