Home Business Why are enterprises wasting funds on bad data?

Why are enterprises wasting funds on bad data?

By Alex James, Vice President of Global Customer Support, Fivetran

by uma

All modern businesses are built on data; whether it’s data to show them how to improve current services or highlight avenues for potential growth. But all too often, this data is riddled with errors, which, if unchecked, can cost enterprises large sums of potential profit. In Autumn 2021, Wakefield Research surveyed data and analytics leaders across the globe to examine the impact of flawed data management. The results were nothing short of shocking – 85 percent said they have wasted funds by making decisions based on bad data.

The issue lies not with business intelligence (BI) tools or the algorithms conducting data analysis on top of this data, but rather the underlying infrastructure responsible for moving data between siloed applications or databases and cloud destinations such as data lakes and warehouses – the core producer of insights within a business.

 

Data engineers are bogged down in manual maintenance tasks

The wasted costs predominantly stem from the degree of maintenance required to keep data pipelines up to scratch and reliable. These pipelines are what transport data between sources and destinations – and despite the simplicity of concept, the process is still resource-intensive for most organisations. In the Wakefield research, 80 percent of businesses said they had to rebuild their pipelines after deployment, with 39 percent saying rebuilding pipelines happened often.

The job of keeping data pipelines running smoothly falls on data engineers, who must manage them alongside other high-value projects. Despite being the backbone of the enterprise, highly qualified data engineers are still made to regularly labour through mundane maintenance tasks to provide reliable – and crucial – data for analysis. In fact, 82 percent of business leaders claim that their engineers spent over a quarter of their time simply building and maintaining data pipelines.

Let’s consider the financial cost of this. When coupled with data leaders reporting a median of 12 data engineers averagely earning $98,400 per year each, the results show this inefficiency costing companies over $500k per year, which eats significantly into companies’ competitive edge. What’s more, time sunk into rebuilding pipelines takes away from data engineers’ ability to produce more advanced analysis, as data leaders say this is the key area their engineers could contribute to if their time wasn’t consumed by avoidable manual processes.

With competition over data engineers fiercer than ever before, it is becoming increasingly pivotal for enterprises to ensure they are not paying vast sums of money to waste engineers’ time on easily remediable tasks. Not least because these engineers will eventually find the tasks unrewarding and look for more exciting opportunities. But employee retention and workload aren’t the only concerns for businesses. According to the Wakefield survey, 83 percent of business leaders have identified a barrier to scaling up their business as being unable to afford hiring yet more engineers to manage data. Yet the solution doesn’t have to be costly.

How to get data management right

Almost a quarter of companies operate more than 50 data pipelines, which requires a colossal amount of management. Add to this to how much time in-house data engineers are already spending on building and maintaining data pipelines, and the case for automation couldn’t be stronger. This is especially true when considering the upwards trend in the amount of data that is being generated on a daily basis, which has been increasing exponentially year on year. Data engineers simply won’t be able to keep up with the demand for insights from this ever-increasing pool of data. Luckily, technology already has an answer.

The automation of data pipeline activities means, in the simplest terms, that where the delivery of analysis-ready data once took days, it now takes hours. A supported and well-resourced team of data engineers will increase the production of useful analytics to inform decisions; but for most businesses this is not new information. In fact, 69 percent of data and analytics leaders said business outcomes would improve if their data teams could contribute more to business decisions. Business leaders respect the knowledge and skills of data engineers so it stands to reason that introducing technology that gets engineers away from maintenance and closer to decision-making is the best course of action to take for any growing company.

Enhanced efficiency and speed are not the end of the list of benefits, either. Automation can remove substantial concerns over data security and privacy – an increasingly hot topic in data analytics as the data protection rules (for example the UK’s new proposal to replace GDPR) continue to evolve in the upcoming year. This is because automated services are reviewed against industry security and privacy standards such as SOC Type 2 and PCI before they are rolled out to businesses. The guarantee of meeting these standards, further alleviates the headache of manually adhering to laws and standards that would, once again, fall onto the shoulders of data teams as well as a company’s IT and security functions.

We can see that the impact of manually created and maintained pipelines are not limited to the data teams that are responsible for them. On the flip side of this, making data accessible to wider teams will have a positive knock-on effect on business productivity. Whether it’s marketing teams analysing wider trends in customer behaviour before designing the next campaign or CFOs inspecting the most up-to-date company dashboards or reports to find new avenues for growth; automation not only makes the presentation of data quicker and easier, it completes mundane jobs so that employees can focus their energy on high-value projects.

A positive outlook for data management

No business leader wants to see data, time or energy being wasted in their organisation, yet the research is clear that more can be done to dismantle the barriers to accessing invaluable insights. By using a technology stack with automation at the heart of it, data engineers, analysts and scientists can freely use the data to drive insights. And since they can trust that the data is up-to-date, they can trust that their insights are accurate, too.

There is good reason to be optimistic about the future of data. The results of the Wakefield survey show conclusively that business leaders are already aware their outcomes would improve if data teams could devote more time to the analytics behind data-driven business decisions. And with technology at the ready, there has never been a better time to leave avoidable data management practices and wasted data in the past.

 

You may also like