Big Data: 2 Steps To Calculating Real Business Value

We’re pretty sure most CIOs are rather tired of hearing, “big data”, “analytics”, & “omnichannel”. There is so much hype around big data that it seems that if you aren’t investing in it, you’re doing something wrong; at least that’s what executive teams are hearing from the board room, and very loudly at that. According to a recent forecast, from research firm International Data Corp (IDC), big data spending will reach $48.6 billion by 2019. The report went on to point out:

1. All three major data sub-markets, infrastructure, software, and services, are expected to grow over the next 5 years. Infrastructure, which consists of computer, networking, storage and infrastructure, and other data center infrastructure, like security, will grow at a 21.7% CAGR (compound annual growth rate).

• Software, which consists of information management, discovery and analytics, and application software, will grow at a CAGR of 26.2%

Obviously, big data isn’t going anywhere anytime soon. So, where does that leave your enterprise in the mix? Apparently big data must be useful somewhere, so let’s go hire someone, give them access to whatever they need and let’s see what they come up with. Yeah, that’s not a good idea. Your invitation to leave the company will be in the mail shortly (probably overnight) because you’re going to spend more money than you could possibly fathom with very little return, if any, to show for it. Whoa! Did a vendor just say that? Sure did!

What “should” you be looking for when management decides it wants to invest in big data? To begin, a big data or analytics investment is a “solution”. If you’re investing in a solution, prior to identifying the problem that the solution will solve for, then you’re in for trouble again. There are a couple of steps to take before making the big data plunge.

 

 

STEP #1: IDENTIFY THE PROBLEMS WITH YOUR BIG DATA

YOUR ORGANIZATION IS FACING THAT BIG DATA MIGHT BE ABLE TO SOLVE FOR.

Identifying the problem you’re looking to solve is half the battle. Here are some commonly stated problems, that can be solved with big data, across just a few industries:

• Travel: I have customers who are trying to plan vacations, but between hotels, air travel, car rental, and extra activities, there are millions of combinations. I feel like we aren’t getting them the information they need fast enough, which is leading to abandoned transactions and lost opportunities.

• Retail: I think we’re leaving a lot of in-store & online revenue on the table because we aren’t associating purchasing behaviors with other consumer behaviors, nor are we very good at in-store strategic product placement, especially for similar products.

• Banking: We’re trying to increase our share of wallet, but our transactional data isn’t giving us enough information to make
  our customers better offers. On top of that, we need to find better ways to minimize customer attrition.

• Insurance: We have call centers, handling large
  volumes of calls daily, from our clients. That cost center can turn into a profit center if our people knew how to upsell properly, but I don’t know what all of the upselling opportunities are.
• Manufacturing: I manufacture multiple products, and there
  are a ton of process interdependencies, which has significantly increased our production costs, due to process inefficiencies, and ultimately decreases the amount of volume we can produce over a specific period of time. 

 

STEP #2: MAP OUT THE INSIGHTS

YOUR COMPANY EXPECTS TO GAIN FROM THIS INVESTMENT AND QUANTIFY THE “REAL
BUSINESS VALUE” THE THE NEWLY ACQUIRED INSIGHTS ARE PROVIDING.

Since we used some industry specific examples above, let’s go ahead and build on that using the same examples:

• Travel: By analyzing previous purchasing behavior, I expect to gain insights into what products/services were purchased in each transaction, at what price points, from a specific geographic territory, during which season of the year. Based on these new insights, our company will be able to more accurately recommend the appropriate products/services, at the right price point, in a more time efficient manner. By doing so, I expect to reduce abandoned transactions by 13% and increase the amount of each transaction by 19%.

• Retail: If we implement social authentication, we will be able to combine previous purchasing behavior with social behavior, which will allow our organization to extend more relevant offers to consumers online as well as make better in-store product placement decisions because we now know what products are more likely to be purchased based on the types of social activities people who purchase specific products engage in. I expect to increase the number of online transactions per customer by 11% as well as increase additional related product purchases in store by 27%. “People who purchased this, also bought…” – I think we’ve all seen that before; and for good reason!

• Banking: By taking into account online and social behavior I expect to be able to make more relevant credit and/or savings offers to our bank’s customers. By identifying trends across customers that have previously stopped using our bank, I can proactively approach customers with solutions to problems that would have lead to their departure as a customer. By doing so, I expect to reduce our attrition rate by 24%.

• Insurance: By converting recorded conversations, in our call centers, to text and analyzing the textual information, I will be able to better identify what customers are calling in about, identify where upselling or retention opportunities are, and properly train call center personnel to make better offers. These activities will turn our call centers into profit centers instead of cost centers.

• Manufacturing: By analyzing process interdependencies and breaking our manufacturing processes into clusters of activities, I expect to quickly identify and correct inefficient processes, which will not only increase production by 40%, but result in a production cost savings of between $12M & $16M annually.

While it’s obviously difficult, if not impossible, to identify a big data project’s exact ROI upfront, these are the types of positive statements you will be able to make at the end of your big data initiative, with proper planning. You identified real problems and came up with real use cases where big data can solve your problems and, ultimately, maximize your return on investment. This is called, “Calculating Real Business Value”.

If you need assistance calculating real business value for your organization, please inquire about our Big Data: Real Business Value Assessment.