On weekends, with my little experience of working for corporates, I usually end up helping my husband set up processes in his family business. During one such weekend, while speaking to the sales team, I realized each one had a lot of data but the data sets had some inherent problems: One, there was a lot of overlapping, and two, a lot of this data was unusable – neither accessible nor incomprehensible. I was quite sure people in the company are working with a ‘siloed mentality.’
So, what is a data silo?
Much like the term suggests, data silos are literally conditions when data exists in isolation (silo). When data silos exist in an organisation, it means that a set of people are able to access certain data/information and others aren’t.
Consider this: What if the medical history and diagnosis of a patient in a hospital don’t have the report from all concerned departments? What will it result into? Either a wrong treatment or further delay in the process. In both cases, the sufferer is the patient and flawed decision-making that can bring a bad reputation to the hospital.
I was intrigued to dive a little deeper and find out the problems that data silos bring with them. To my surprise, a similar problem exists in many large organizations.
The State of Data Science and Analytics report by IDC suggests that data workers spend 90% of their working week (around 36 hours) on data-related activities such as searching, preparation, and analytics (with searching for and preparing data being the most time-consuming ones).
I spoke to a few CXOs from different domains to further probe it and this is what they had to say:
Alok Kumbhat, Director – Data & Operations, Dun & Bradstreet says, “Data silos lead to duplication of efforts, which, means increment in costs. It limits knowledge sharing, which, in turn, leads to creating parallel structures and significantly limits organisation growth. Data silos often give us different versions of truth thereby leading to decisions that might impact company performance. Teams have their respective versions and this defies the core objective of “one team, one mission, one voice”
According to Sukhvinder Singh, Enterprise Account Director, GoTo, it is essential to have a single view for all your customers and prospect data in today’s data-driven world. “Data in silos will not give a clear picture of your account penetration, and you can miss out on available opportunities. Your marketing effort will also not be effective if your customer data is not unified,” says Singh.
Gopi Gonuguntla, Solution Architect -Data Center & Cloud, World Wide Technology says, “ Data silo limit the view of data and threaten its integrity. They end up in wasting of resources. They create barriers to information sharing and collaboration”
Bhavesh Shah, Associate Director – Customer Success, Microland says, “While data silos could be unintentional isolation of data (kept for being accessible for specific BU function), it can potentially chew up more resources, lose its effectiveness, hinders the wider organisational view and ultimately, results in loss of opportunity.In the context of wider organisational outlook/interest, it is recommended to unify data silos.”
To sum up, I realised that data silos are no less than silent killers to an organisation and its growth as they cost the organisation money, time and effort.
Question is, under what circumstances do data silos emerge?
The two major reasons are as follows:
- Technology Issues
If the organisation doesn’t have a proper information architecture and commensurate technology in place, then data cannot flow easily within departments. What is essential is the quick transfer of information, cross reference, and security – all simultaneously.
- Organisation Growth
Too many departments, offices, and fast growth lead to structural issues. Apart from that, cultural issues like competition within employees, ownership of data are some of the other issues, which hamper the sharing of data.
Fixing this problem is an absolute necessity. It won’t be unusual to assume that to kill those silos one has to throw away years of piled-up data, begin afresh, hire a set of experts and change the operational infrastructure.
So, what is the solution? While there is no one silver bullet to handle this challenge, however, organisations do give it a try through various means. Look at the chart below. This could be a possible way to handle the problem of data silos:
Apart from speaking to many CXOs mentioned earlier, I also managed to dig up some case studies on Data Silos. Each of those case studies has a link in case you would want to read it in more detail.
- An example where a law firm worked hard to eliminate its data silos with the objective of saving costs and attracting clients is Winston & Strawn. (Read more about this on the link Law Firms Dismantle Data Silos to Save Costs, Attract Clients (bloomberglaw.com))
- Merrill Lynch, one of the world’s leading wealth management, capital markets, and advisory companies, also faced myriad problems because of data silos. Using a2n integrated modelling platform, they created information clarity out of the chaos which helped them take a quantum leap in providing timely information to their executive management. (Read more on Case Study: How Merrill Lynch Overcame Data Silos – The Global Treasurer)
- eBay’s data resided in various locations and formats across their technology stack, not easily accessible to all key stakeholders. They used a scalable and flexible integration architecture to get rid of their data silo. (Read here: APIs for the enterprise: breaking down data silos at eBay (mulesoft.com))
- Samsonite broke down their data silos and this helped them optimise their digital advertising. Data-centricity helped them to improve efficiency by dynamically monitoring their performance (Read here: How Samsonite broke data silos down – Think with Google)
Though scripting and on-premise ETL tools can help in the integration of data, Cloud-based solutions are the future. They promise to abolish the barriers to the collaboration of data. It is a technology-ready solution for connecting siloed data, which could give access to enterprise-wide data.
Integrating data, making it available to the entire organisation while keeping in mind security and compliance, is not an easy task. If we have to look at it from another perspective, data silos are key to protecting sensitive data from hackers, which is a reasonable argument given that many companies may not have a good architecture and appropriate security. Breaking data silos takes time and is an expensive affair. However, ignoring it will turn out costlier as it may affect in the long run. What organisations need to aspire for is an environment of transparency and a culture of collaboration as an antidote to silos.
One key thought could be to look for your data silos, why have they been created and how can you work with your team to get rid of them.
So, does your organisation have a program to eliminate data silos? I would love to hear your thoughts and best practices.