Spotting the silos
It can be difficult to spot data silos as, by their very nature, they are clandestine and not connected to other areas of the business, and this is particularly tricky for CDOs joining an organisation that is early in its data journey. Data leaders and their teams need to fall back on the tried-and-tested skill of data storytelling to get other department heads involved with addressing data silos. By explaining and demonstrating the benefits of data and how data silos are a growing threat to success, it should be possible to build organisation-wide momentum for finding silos and solutions.
A data leader needs to have conversations with all areas of the business – from junior staff to board members – and examine what data they currently use (if any), how this data is used and whether it is shared beyond the immediate team. This can also uncover shadow IT systems as sometimes data is being stored in an unregulated subsystem within an organisation. If the business has regional or international operations, these need to also be investigated as there are often geographical silos and silos created by regions speaking different languages.
The different tools and systems being used with data must also be examined:
- Are there interfacing issues between different tools?
- Are implementation bottlenecks surrounding different operations?
Once these have been catalogued and analysed, it is possible to begin working on addressing the issues that cause silos.
How to smash a silo
One of the first steps a data leader can take is to conduct a data audit. This will require strong communication and collaboration with all aspects of the business as a review of all data sources will be needed. This is a good opportunity for the data team to directly connect with different areas of the business and build the standing of the data team. Once the review is underway, the data team needs to identify where the data is being stored and how it is subsequently used by different stakeholders.
It is all well and good to have a substantial amount of data, but is it all suitable and needed? Data leaders and their teams need to scrutinise the existing data and identify duplication and outdated data sets that are not relevant, or perhaps even not compliant. There must be a level of cleaning and processing that is undertaken to ensure the highest quality data is being used to drive changes and decisions, rather than relying on siloed, incomplete data from legacy shadow operations.
There should be efforts made to codify the systems organising data. This is particularly important for businesses operating in heavily regulated industries such as insurance or finance as it will expedite requests from regulators and improve compliance. A codified system allows team-wide processing and compliance with handling data, as well as improving the ownership and stewardship of different data sets.
In many cases, data silos exist because different departments or teams are storing the same data in different places. A company should look for instances where data is duplicated and consider consolidating it into a single repository. There may be the need to invest in tools that can centralise data or at least provide greater context and security for those drawing on the data for their day-to-day operations. Investment is always a difficult conversation to have with business leaders, but data leaders need to hone their storytelling skills and rely on their ability to explain in non-data language why these tools are required, and the business benefits they will bring.
One final solution to help end data silos is the implementation of AI tools. In 2023 there was a rapid development of AI solutions which can be used to improve data processes within organisations. Data leaders that can implement AI technologies can utilise customer service tools such as chatbots to improve and streamline services that:
- Are available 24/7
- Provide rapid resolutions
- Reduce human errors
- Deliver personalised recommendations
- Track and analyse data.
By installing new technologies that demonstrate the effectiveness of data, improve data management and provide access to data-led learnings for all levels of an organisation, data silos will cease to exist. The growth of the data culture of a business is essential for defeating data silos, but a multifaceted approach to ridding silos is needed and requires patience as the scale of the problem varies between businesses and teams.
As most data leaders know, change is not something achieved overnight, and trying to change the fundamental ways that individuals work in an organisation is a slow process – but the eradication of data silos only provides benefits and is worth the effort.
