How to identify data and AI skills gaps
The talent gap and the technology gap have been discussed numerous times across the DataIQ community, and it bears repeating as this is a serious concern for those striving to become data driven – if there are talent and tool gaps, the true value that can be realised will not be achieved.
“The fact is, right now, almost every business is behind on AI,” said Anna Wang, Head of AI, Multiverse. “The technology has developed so fast; most organisations have not been able to access the extremely limited pool of talent or embed AI safely and securely into their tech stack and workstreams. I know from speaking to business leaders that they are optimistic AI can bring huge productivity and efficiency gains, but they remain deeply concerned they will miss out on those gains if they fall behind on technology or skills.
This was echoed across the roundtable with numerous participants stating that they are also concerned about being left behind. The issue is further exacerbated by the increasing costs for recruiting and retaining talent in a drought as data professionals can often be poached by competition for increased remuneration packages. Throughout 2024, data offices within businesses have had their budgets slashed, while the costs for talent and tools continues to rise, further raising concerns about closing already prominent skills gaps.
Gaps can be identified through regular assessments into the capabilities of different teams and through improved communication and collaboration. This requires time investment from data leaders to heighten the connections between the data team and other teams within the business, but the results are worth the effort as the standing of data will improve and the ability to spot issues with skills gaps will be easier with more trained people.
How to build a data and AI culture
There are several key things that need to be put in place to improve data and AI culture, including:
- Collaboration
- Leadership
- Data literacy
- Data access
- A proactive approach
Earlier this year, DataIQ published ten pain points on developing data culture, as well as ways to measure the development of data culture which can be notoriously difficult. Those seeking to begin their data and AI culture journey should read about the differences between data culture and data literacy to ensure they are building as effectively as possible.
Roundtable participants discussed the different ways they had been building their data and AI culture, with each participant noting that they were at different stages of the process. They all agreed that the above-mentioned points are pivotal to ensuring success and then proceeded to share tips and tricks about how they have achieved buy-in from decision makers in different areas.
How to measure the ROI of upskilling programmes
One of the issues is that business decision makers who are not well-versed in data will often just throw money at tools expecting them to be a comprehensive solution for upskilling programmes. The problem is that the financial cost of the investment is then held against the upskilling programmes as an easy way to show ROI, even though the tools being integrated are not necessarily relevant or beneficial to the ultimate aims of the organisation.
“Right now, companies spend four times more on software than they do skills,” said Wang. “This is leading to a situation where many businesses will find themselves with high-end, complex tools, but nobody that is able to use them.”
There must be investment in people and talent, not just the tools available for use across a business. This takes time and will vary from department to department, but it is the finest way to track ROI in relation to upskilling programmes. Once the groundwork has been laid, the development of data literacy and wider data culture will gain momentum and evolve naturally.
When it comes to measuring business value through literacy programmes, Wang explains that: “Whether it’s in data or AI, the pool of pre-existing talent is not big enough to meet the demand today, and certainly not big enough to fill the demand that will emerge in future. It is incumbent on data leaders to create that talent through training.
“It is not just about creating experts in AI and data, it is also about embedding data skills across an organisation, at every level and in every department. Having these skills siloed in one department will lead to bottlenecks, everyone must have some level of understanding with how to use emerging technology to reap the benefits.”
Silos will continue to harm businesses until they are addressed, and the full scope of business value will never be fully realised while silos exist. One of the best ways to address silos is by improving data literacy and data culture, which will take time, but once it is implemented it is very difficult to remove. Each time a new prospect is brought into the team, they need to be upskilled to meet the new level required for data literacy.
“At Multiverse, we have partnered with more than 1,500 businesses to identify skills gaps in areas like data and AI and deliver targeted training to close those gaps,” said Wang. “That training is applied to ensure learners develop in the areas that are directly relevant to their work. We track ROI by looking at measures like the amount of time saved on data and AI tasks, or new revenue opportunities identified. Our ROI tracking has identified more than £2 billion of return to our partners to date.”
Data leaders need to take a pragmatic approach to demonstrating business value through data literacy and they need to ensure there is a minimum level of understanding of how and why data value should be measured in these ways. Literacy programmes are the central foundation to any data-driven business – particularly for an AI-centric future – and they must be implemented to achieve success. However, there is no overnight fix for data literacy and the task to improve overall organisational literacy is a slow process, but once this is accepted, an AI future will become second nature.
Get in touch with the Multiverse team here.
To get involved with upcoming exclusive DataIQ roundtable discussions, click here.
