This blog was authored by: Martyn Bullerwell, VP of Data & AI, with insights from  Ben Jarvis, CTO and Matt How, Head of Data Science | 20 January 2025

 

Artificial Intelligence (AI) has become a buzzword in recent years, dominating headlines and sparking discussions in boardrooms worldwide. While the excitement around AI is justified to some extent, it has also led to a significant amount of misinformation and confusion. Many organisations have rushed to adopt AI without fully understanding the different types of AI and the capabilities or limitations that come with it.

 

In this article, I will share our view on where organisations should focus their efforts in 2025 including advice and key considerations to ensure the success of AI projects.

Understanding AI: Generative vs Traditional

AI can be broadly categorised into two types: generative AI and traditional AI. Generative AI, such as the viral sensation ChatGPT or Microsoft’s productivity enhancing Copilots, are designed to generate content in response to user input. This could be summarising content, calling out trends in data or suggesting novel insights. On the other hand, traditional AI encompasses machine learning, predictive analytics, and data science. These AI models analyse existing data to make predictions, identify patterns, and provide insights.

 

Many of the articles you would have seen over the past few years have been focused on Generative AI. It is important to understand this context when reading peoples viewpoint on the subject.

The Hype vs. Reality

The hype around AI has led many to believe that it is a magic solution that can solve all business problems. However, the reality is that AI is just a tool, and its effectiveness depends on how it is used. It’s crucial for businesses to identify their specific challenges and then determine which type of AI can best address them.

 

The most common challenges we see customers continuing to face in 2025 are:

Digital Transformation:

When integrating systems as part of a larger transformation, organisations often need support in managing data and streamlining reporting systems.

Department-led Insights:

Organisations often have specific business questions but are unsure how to best utilise their data to find answers.

Data Science & AI:

Staying ahead in the AI landscape is crucial, but organisations struggle with how to decipher specific use cases and bring those to production.

Data Governance:

Having clear rules and processes ensures your data is accurate, secure, and compliant, helping your organisation run smoothly and efficiently.

Hear from Ben Jarvis

  • “

    “To fully leverage the potential of AI, it's important to change the narrative around it. Instead of viewing AI as a standalone strategy, businesses should see it as a tool that supports their overall strategy.”

    Ben Jarvis

    Data and AI CTO
    Connect on LinkedIn

So, What Is Our Advice?

Adopting Generative AI tools for personal productivity like Copilot and ChatGPT is essential for staying relevant. They will soon become a standard tool, much like Microsoft Word on every device. Failing to embrace them could result in recruitment challenges and skill gaps.

 

However, organisations will gain a competitive edge by addressing their specific business challenges and identifying the right technology solutions. This includes developing unique AI use cases and utilising organisational data to enhance existing processes and decision-making. We see building these AI use cases being driven by three key motivations: increasing revenue, reducing costs, and enhancing governance and security. Essentially, AI helping businesses grow, save, and protect.

 

We have worked with an array of customers across different sectors to build out a portfolio of Proof-Of-Concepts for organisational enhancement. To bring this to life here are a few examples:

Council Tax Arrears Support
Managing council tax arrears is challenging for local councils due to the need to identify at-risk residents and provide timely support.
flip me
How AI Can Help:
AI can help by predicting the likelihood of arrears, alerting civil servants to key causes, and forecasting the impact on other services.
Link here
Hospital Service Demand Forecasting
Hospitals struggle with demand forecasting, leading to staffing issues, resource inefficiencies, and delayed patient care, as traditional methods often can't handle the complexity of healthcare needs.
flip me
How AI Can Help:
AI-driven demand forecasting models analyse diverse data sources to predict future service demands, enabling hospitals to better plan and allocate resources, and integrate with existing systems for real-time insights and decision support.
Get in touch
Detecting and preventing fraudulent activity at scale
Fraudulent activities, such as identity theft, credit card fraud, and insurance fraud, pose significant financial risks to businesses and consumers alike. Traditional methods of fraud detection often rely on manual reviews and rule-based systems.
flip me
How AI Can Help
AI can analyse transaction data in real-time to detect fraud, predict suspicious transactions using historical data, and use NLP to identify fraudulent communications.
Get in touch
Maximising customer spend and loyalty
Customer purchasing behaviour can be difficult to comprehend and predict, especially across diverse streams such as e-commerce and digital platforms. Manually forecasting appropriate offers and content cannot scale at sufficient rates.
flip me
How AI Can Help:
Identifying patterns in behaviour using AI can signpost behaviours that lead to customer attrition. Further to this, AI can predict the appropriate level of discount or offer to be applied to entice customers to remain loyal.
Get in touch
Enhancing Efficiency and Compliance in Social Care
Social care professionals spend a lot of time manually transcribing interactions, which can lead to delays and inaccuracies. This administrative burden reduces time for direct care.
flip me
How AI Can Help:
AI-driven transcription services can convert spoken interactions into text, with speaker identification and real-time summarisation to flag key points and risks. This results in reduced administrative effort, lower labour costs, and improved compliance with more accurate and timely documentation.
Get in touch

Hear from Matt How

  • “

    “AI is just another toolset that you're going to have to adopt at some point to stay relevant and competitive. But don't look at how you
    can use the technology; look at what problems you have and what technology can solve them”

    Matt How

    Head of Data Science
    Connect on LinkedIn

What impacts the success of an AI initiative?

  • 30%

    of generative AI (GenAI) projects will be abandoned after proof of concept by the end of 2025. (Gartner, Inc)

  • 75%

    of consumers are concerned about misinformation from AI (Forbes)

  • 35%

    of businesses worry about having the technical skills to use AI effectively (Forbes)

  • 92%

    of people cite cultural and change management as the top barrier to becoming data-driven. (Sloan Review)

The Importance of Creating a Data Culture

One of the key factors in the success of any AI initiative is the quality of the data being used. Poor data quality can lead to inaccurate predictions and unreliable insights. Therefore, businesses must invest in building a data culture.

 

Creating a data culture is far beyond what people associate with the usual term of becoming ‘data-driven’. We support organisations in changing their approach and the way their business views data based on four key components technology, enablement, guardianship and people.

Learn more about creating a Data Culture

Ethical AI

While AI offers numerous benefits, it’s also important to consider its environmental and cost implications. Running AI models, particularly generative AI, can be resource-intensive and expensive. Businesses need to weigh the benefits against the costs and explore sustainable AI practices. This includes choosing the right models for the right tasks and optimising AI processes to minimise resource consumption.

 

Additionally, it’s essential for organisations to build confidence in reviewing generative AI outputs. This may initially impact the return on investment (ROI), but it’s vital for establishing trust and ensuring the quality of AI-generated content. Governance of AI outputs is a significant consideration that cannot be overlooked.

 

 

Building an AI Skilled Workforce

Upskilling and training should be part of every Data & AI project to avoid failure. Companies need to invest in their workforce to keep technology relevant and ensure ROI. This means teaching employees to use AI tools confidently, building AI skills across the organisation.

 

It’s tough to know where to start, so we work to a knowledge transfer model. This is something all providers should be doing if they aren’t already. Projects should involve collaboration, allowing clients to gradually take over and improve the platform themselves.

Discover our Knowledge Transfer Model

Conclusion

AI has the potential to reshape the way businesses operate, but it’s important to approach it with a clear understanding of its capabilities and limitations. By focusing on specific business challenges and using the right type of AI technology to address them, organisations can tap into the true value of AI.

 

Over the coming months I will be sharing more around our view on different AI topics, alongside Ben Jarvis and Matt How. This includes expanding on some of the topics above including how to build a data culture, choosing the right AI tool, and the environmental impact of AI.

 

If you would like to be notified of the new releases, fill out the form below and our Marketing Team will add you to our next publication launch. Otherwise, please feel free to reach out either directly to me via LinkedIn or through the form below.