Data Science

5 reasons why big data initiatives fail

traditional data warehouse

A 2020 New Vantage Partners survey of C-suite executives reported that 98.8% of organizations have invested in some kind of data initiatives. And yet only 37.8% of companies have created a data-driven organization.

Not that companies are lagging from a technology perspective, as only 9.1% of organizations reported that technology is the barrier. If not technology, then what is causing them to move at such a slow pace? What are the barriers to becoming a data-driven organization and how can organizations overcome them?

The questions are impressive but not intriguing enough as a closer look at how processes and people are responding to data initiatives in the organization can quickly reveal everything wrong with those initiatives. Easier said than done, the challenges lie with identifying the risk areas, taking actions to mitigate the risk, and implementing new ideas to continuously improve on the existing processes—all in one breath.

In an attempt to identify those risk areas, we have compiled a list of 5 key reasons for big data initiatives failure. How companies can prevent those failures run parallel with each reason.

Extensive Focus on Volume

The more, the better is a notion that many executives fall for. Since the dawn of Big Data, the emphasis has been on volume, velocity, and variety. Considering the fact that “2.5 quintillion bytes of data is created everyday,” organizations often get themselves overwhelmed with data in an effort to capture more information.

However, many Fortune 1,000 companies have reported that it is not the ability to manage and process Big Data that has driven successful results instead the ability to use a variety of data. For instance, if businesses want to improve cross selling, they need a variety of data ranging from customer accounts, purchasing history, social media behavior data, and employee activity.

Here employee activity is a crucial metric that will help them to measure their efforts in increasing cross-selling. The efforts again must be validated by a variety of data to understand on what fronts the team is lacking or taking a leap.

As you can see you if you are only focusing on customers’ accounts, you might not be able to unravel the mystery behind what is going on with your cross-selling strategies. Having a variety of data to tally is essential to add the right context to the data.

Bad Data

The benefits of quality data are many—better decisions, lower cost, and fewer mistakes. And so are the consequences of bad data. Incomplete, poorly defined data could result in unhappy customers, productivity drains, and inefficiencies.

According to a research by Experian plc, Martin Spratt of Clear Strategic IT Partners Pty. Ltd.  and consultant James Price of Experience Matters, “the cost of bad data is to be 15% to 25% of revenue for most companies.” A similar study reveals that only 3% of data collections were under the acceptable error range, and nearly 50% of newly created data had critical errors.

Your data collection, processing, and utilization must be based on the objective of your decisions. Randomly collecting with no significant value can mislead you to make wrong, expensive decisions.

Poor Data Culture

Is everyone in the team on the same page?

Do they all understand the value of data? Are they even using the data to identify the gaps in the processes or still relying on their instinct and opinion to make decisions?

How you read the data is crucial to getting to a conclusion. Often it has been observed that not all teams are capable of reading the data to use it to their advantage. While others read too much into the data which makes them ruthlessly overtake the strategy.

In order to forge a healthy data culture in an organization, business leaders must make actionable commitments. For example, if you want to capture your sales data, you must empower your field sales reps with tools to enable them to collect data and then store, process, and utilize the data to make informed, evidence-based decisions.

Errors in Data Linkages

The consequences of errors in data linkages could be high for the individual, the team, and the organization itself.

Let us assume that you have set sales KPI for the team and in the process tracking metrics to figure out your most valuable salesperson. Here, if you are measuring the individual sales performance on the metrics of who made the most calls or brought the most business, the data might lead you to the wrong path.

As in if a certain salesperson has made 18 calls in day and has generated $1 million in revenue from a high performing territory and the second the salesperson has made 20 calls in day and has generated a revenue of more than half-a-million from low-performing territory, then it won’t be difficult to choose your valuable salesperson.

As you can observe, there are three factors that we have included so far, which are no. of calls made to check efficiency of the salesperson and territory and revenue made to measure the effectiveness.

There could be several such factors which you could link to each other to measure the actual performance of the salesperson, given that your sales CRM allows you to capture, process, and interpret data effectively.

Lack of Decision-driven Approach

What is wrong with the data-driven approach?

Haven’t we heard CEOs and business leaders taking pride in calling themselves a data-driven company?

Here is the problem: data is not information. It is a characteristic of information. If you want insight from the data, you must ask the right questions. And then you need to know which data to look at. For example, employee productivity related data cannot fully unravel why the sales is declining or stagnant. Customer account data will not let you see why your customers are not happy.

The right approach is to make a decision first and then figure out the data to support your decisions. Decide your next business move beforehand; for say, whether you want to increase cross-selling or to enhance employee productivity or expand your territories.

Once you have decided on that, then find out the data to see where your efforts are lacking in cross-selling or in what activities most of your sales reps spend their time.

Going Forward

Data is no magic wand. Neither is it a secret pathway to success. It is a plain record in numerical form. If you have a well-maintained, accurate record of everything associated with your business, you can find out the patterns that could provide support to your decisions. But before that comes making a decision. Decide first and then support your decision with data.

Leave a Comment

Your email address will not be published.

You may also like

Pin It on Pinterest