AI
Problem Lenses: How and when should I use AI? - Part 1
Part 1 of Ali Salaman’s three-part series looks at how AI could help businesses automate repetitive tasks, support complex decisions and generate insights from unstructured data
As businesses evolve in today’s digital landscape, many are exploring the potential of artificial intelligence (AI) to solve challenges and drive innovation. However, for non-technical business owners or managers, it can be difficult to determine whether AI is a viable solution to their specific problems.
This three-part blog series aims to help you decide whether AI can address your business challenges by examining them through different "problem lenses". These lenses provide practical ways to assess where AI might be applied effectively. We initially presented this framework at Digital Leaders Week 2024, and now we're offering more in-depth guidance to help you apply these insights to your own organisation.
In this first blog, we’ll cover three lenses: the lens of high volume repetitive tasks, the lens of complex analysis & decision support, and the lens of unstructured data insights. We’ll explore how to assess whether your business challenges fall within these categories and offer practical steps to guide your thinking.
The Lens of High Volume Repetitive Tasks
What is this lens?
This lens focuses on areas of your business where repetitive tasks consume significant time and resources. These tasks often involve routine, structured actions that are necessary, but do not require deep thought, creativity, or complex problem-solving. Common examples include responding to customer enquiries, processing forms, or managing standard reports.
Practical steps to apply this lens:
Identify repetitive processes: Start by looking at areas where employees are spending a lot of time on repetitive, low-value tasks. Consider tasks that follow a set pattern, such as answering the same customer questions, manually sorting emails, or filling out routine forms.
Evaluate the volume and cost: Assess the volume of these tasks and how much time your staff spends on them. Are these tasks preventing employees from focusing on more strategic work? Does handling these tasks create bottlenecks in your operations?
Assess the potential for automation: Ask yourself if these tasks involve predictable, structured actions. If the answer is yes, AI can likely automate or assist with these tasks. For example, AI could be used to respond or draft responses to frequently asked questions, triage customer service requests, or pre-populate forms based on previous inputs.
Examples:
Customer enquiry automation: If your business receives hundreds of customer enquiries daily that ask similar questions, AI chatbots can automate responses, providing instant help and freeing up customer service staff to handle more complex queries.
Document processing: AI can automate document review processes by scanning and verifying the information against predefined criteria. For example, import / export declarations could be processed automatically, ensuring compliance with regulations.
By automating high-volume, repetitive tasks, your business can save time, reduce costs, and refocus your employees on higher-value activities.
The Lens of Complex Analysis & Decision Support
What is this lens?
This lens is about leveraging AI to assist with complex decision-making processes that involve large amounts of data or multiple sources of information. AI excels at processing vast datasets, identifying patterns, and generating insights that might be missed by human analysts.
Practical steps to apply this lens:
Identify decision-making bottlenecks: Look at areas where decisions are delayed due to the complexity of the data or the time it takes to analyse information. Perhaps decisions were never made and a project was entirely decommissioned because it was too difficult to sift through the amounts of information needed to make a decision. Do you find yourself struggling to draw conclusions from large data sets or complex reports?
Evaluate the complexity of the data: Are you dealing with data from multiple sources? Are there complex relationships between different data points that need to be analysed together? AI can be particularly useful in making sense of these connections.
Consider the value of insights: Determine how much value timely and accurate insights would add to your decision-making processes. If faster data analysis would lead to better, quicker decisions, AI might be the right solution.
Examples:
Market analysis: If your business collects large amounts of market data (e.g. customer feedback, sales trends, or competitor analysis), AI can help by analysing this data to provide summaries, identify trends, and even recommend strategies based on those insights.
Housing demand forecasting: A government department could use AI to forecast housing needs by analysing a wide variety of data (e.g. local property prices, population growth rates, and planning permissions). AI could produce more accurate predictions than traditional methods, helping to guide policy decisions and resource allocation.
By using AI to support complex decision-making, you can ensure that your business is making more informed, data-driven choices.
The Lens of Unstructured Data Insights
What is this lens?
Unstructured data refers to information that doesn’t fit neatly into a database or spreadsheet. It will typically be in the form of blocks of text such as emails, social media posts, survey responses, and customer feedback. This data is often messy and difficult to analyse, but it can hold valuable insights about your customers, products, or market trends. AI, especially large language models (LLMs), can process and analyse this unstructured data to extract meaningful insights.
Practical steps to apply this lens:
Identify sources of unstructured data: Review areas of your business where unstructured data is being collected but not fully utilised. This could include customer feedback forms, problem description fields in application forms, social media interactions, emails, or even internal communications.
Consider the value of insights: Think about how valuable it would be to derive trends, sentiments, or actionable insights from this data. For example, could understanding customer sentiment from feedback help improve your product or service? Could analysing internal communications reveal workflow inefficiencies? Perhaps analysing a problem description field to categorise the issue could help unlock further automation opportunities in your process workflow.
Evaluate AI's ability to process this data: LLMs can analyse and summarise unstructured data quickly. If you're facing large volumes of unstructured text, AI can help sort through it, identify key themes, and provide actionable recommendations. Try to connect the dots between the actions that need to be performed and whether categorising, summarising, reasoning, understanding capabilities of LLMs would be beneficial to perform these actions after processing the unstructured text.
Examples:
Customer feedback analysis: A business that receives hundreds of customer feedback forms after each product launch could use AI to automatically analyse and categorise this data. AI can identify common themes, sentiments, or complaints, giving the business actionable insights for product improvements.
Public consultation submissions: A government department collecting public feedback on proposed legislation could use AI to summarise and analyse the submissions. This helps policymakers understand public sentiment and identify key issues without manually sifting through thousands of responses.
Unstructured data is often an untapped resource in many businesses. By using AI to extract insights from this data, you can unlock new opportunities for growth and improvement.
Applying the Lenses to Your Business Problem
To determine if AI might be the right solution for your business, you can apply these lenses in a structured way:
Map out your processes: Start by listing your core business activities and the challenges you face. Validate that those business processes still make sense - automating a bad process is still a bad process. Identify areas where you struggle with high volumes of repetitive tasks, need better decision-making support, or have untapped unstructured data.
Evaluate AI readiness: Not all processes will benefit from AI. Assess which tasks or processes are structured enough for automation, the areas where data complexity is slowing decision-making, or where valuable insights from unstructured data are being missed.
Start with a pilot project: Once you’ve identified potential areas for AI, begin with a small, low-risk pilot project. This allows you to test the capabilities of AI in your organisation and build confidence before scaling up.
Conclusion
By examining your problems through these lenses, you can begin to see where AI might provide solutions and how to take the first steps toward implementation.
In the next blog, we’ll cover additional lenses, including process automation and optimisation, as well as the ethical considerations you need to keep in mind when implementing AI and multilingual translation capabilities. Stay tuned for more insights from Caution Your Blast Ltd into how AI can help your business thrive in a digital world.
Caution Your Blast Ltd's Librian AI Solution can revolutionise your customer service by providing fast, accurate, safe and ethical responses to user questions - it is proven to reduce in-bound written enquiries by 80%. Request a demo or get in touch here