“Every company has a machine learning opportunity; however, not every problem is solvable by machine learning.” - Michelle Lee, AI keynote speaker, technology and leadership expert, Obsidian Strategies Founder and CEO
As Head of Technology at YLD, my daily interactions with clients and team leadership have revealed a common challenge: many CTOs and top-level decision-makers are under pressure from investors and boards to present an AI plan, yet often lack clear next steps and a structured framework.
This article aims to address this challenge by exploring how GenAI can deliver tangible business value and empower senior leaders to form practical strategies for organisational enhancement across diverse industries.
Embracing GenAI
Staying competitive means embracing adaptability and fostering a culture of continuous improvement. Deriving insights from MIT Technology Review, it's clear that AI is no longer a futuristic concept but a present reality across various sectors.
Each sector has been adopting GenAI at its own respective pace, and adeptly integrated it into its operations, as evidently shown on the graph. I anticipate that this will continue to happen across more sectors and businesses, as the benefits of adopting GenAI are endless.
Setting clear objectives for GenAI implementation
GenAI should not be viewed simply as a new tool in the technological arsenal, but rather evaluated for its potential to drive improvements and create new opportunities. Deeply understanding where it can make the most valuable impact on your business is vital.
Every organisation can progress using GenAI by tailoring its strategy to specific needs and gradually implementing small to medium changes when a massive overhaul isn’t possible. By balancing risks and rewards through careful experimentation and exploration, the sweet spot lies in quickly adopting and learning from GenAI tools without being reckless.
To achieve this, start with the low-hanging fruit where you can make a positive impact easily, such as enhancing efficiency in your organisation's workflows. Begin by identifying a few use cases to pilot. When selecting a business use case, ensure it holds a potential impact on the business and is feasible for implementation. Factors to consider include the expected ROI, availability of data, technical and organisational readiness, and the complexity of integration with existing systems. Integrate GenAI into your high-impact, high-feasibility projects first to achieve quick successes, which will help build momentum for wider AI adoption.
Define what success looks like for your organisation. Clarity regarding success not only aligns the efforts of your teams and stakeholders towards shared goals but also provides a clear direction for your teams to aim for i.e. enhancing customer satisfaction through personalised and efficient engagement, achieving cost savings through process automation, and driving innovation in product development.
To achieve optimal productivity gains in your organisation using GenAI or LLMs, consider things like RAG, caching, evals, guard-railing, and data and how they can elevate LLM applications. Learn more about techniques for optimising GenAI models here.
Establishing a strong data foundation
An organisation’s data will never be 100% clean or unbiased because there will always be gaps i.e. upstream application failures, programme errors, and more. When it comes to gaps in data, companies often underestimate the effort required to integrate data and clean it up to a standard that can power LLMs and feed into them.
Organisations should adopt a holistic approach to data management to establish the strongest data foundation possible. Here are some ways to achieve this:
- Lambda architecture: Most organisations work with live data so they need to update historical data with new augmentation and enrichment processes. Lambda architecture is great for this because it serves as a base model for other data applications. This architecture provides a standardised approach to handling data issues by combining batch and real-time processing, creating a strong and high-quality data infrastructure.
- Medallion architecture: Medallion architecture significantly improves data management, especially for reprocessing historical data with modern development techniques. This architecture uses multiple layers for data processing to ensure flexibility and accuracy
- Bronze Layer: Stores raw data.
- Silver Layer: Filters, augments, and transforms data, allowing reprocessing with new filters or logic.
- Gold Layer: Delivers cleaned-up results to enhance machine learning models.
- Archiving data: Organisations can easily accumulate terabytes or petabytes of data, requiring significant resources to query it quickly. To manage the data load, data can be archived, deleted after a few years, or moved to an object store like S3. Combining daily and hourly files into weekly ones simplifies management and downloads, making data access easier for teams and speeding up results for engineers.
- Facilitate data interoperability: Adopting standardised data formats and utilising APIs paves the way for efficient system integration and enhances the functionality of GenAI across the business. Data interoperability refers to an organisation's capability to access and exchange data from different information systems, which is vital for enabling collaboration and seamless data exchange.
- Cultivate a data-driven culture: Starting from the top managers, it’s important to establish the expectation that decisions must be anchored in data as the company norm. Their influence encourages employees to adopt this approach, especially when involving stakeholders in data discussions and aligning initiatives with business objectives and metrics.
- Robust data governance and monitoring: It’s crucial to establish accountable owners for data elements to ensure responsible management, maintain accountability, accuracy, security, and compliance. This accountability helps your business stay responsive as it scales. Robust data governance reduces costs in other data management areas i.e. boosting efficiency through data reuse and enhancing confidence in data quality, documentation, and procedures.
Equipping your team for GenAI success
Navigating the rapid growth of GenAI is more than just a nod to new technologies; it demands a strategic approach to talent development and skill enhancement. As a leader, you must invest in building a team that is not only proficient in data science and data engineering but also adaptable to the rapid advancements in these fields.
Executives can take several steps to ensure their teams are leading the charge in the GenAI-driven business landscape:
- Embrace continuous learning: Regular training sessions provide employees with the latest data insights and best practices, ensuring your team knows about GenAI tools and uses them effectively.
- Expand AI literacy across the organisation: GenAI extends beyond the tech department. Championing AI literacy across all departments enables everyone to embrace GenAI tools in a practical manner, enhancing the productivity and effectiveness of teams company-wide.
- Encourage hands-on experimentation: Mastery of GenAI tools is best achieved through direct engagement. Encourage your team to dive deep into these technologies and explore the capabilities that can foster innovative applications, driving efficiency and growth.
Preparing your team for GenAI is about more than just staying competitive. It's about establishing new industry standards, promoting an innovative culture, and ensuring your business leads the way in technological advancement.
Accelerating business performance
We’ve been integrating GenAI to tackle the diverse challenges that our clients are facing. The way we’re progressing is both promising and complex, and my role at YLD has offered me a front-row seat to the profound impacts GenAI can have on operational efficiency, customer service, and innovation.
Exploring and implementing GenAI has emphasised the importance of aligning our solutions with respective clients’ core business objectives. Here are key success stories showcasing effective GenAI applications:
Streamlining customer service workflows
A financial services company that provides over 4 million customers with responsible access to credit, easily receives an influx of inquiries daily. Read more about how we helped streamline their customer service by implementing an audio-based Retrieval Augmented Generation (RAG) system.
Improving decision-making through enhanced data management capabilities
A major broadcast media company strives to reach more niche audience segments. To achieve this, we optimised our client’s data management capabilities by leveraging GenAI to fetch accurate data and feedback information instantaneously in a visually accessible way. Read more.
Providing the most accurate money-saving opportunities in energy consumption
An energy supplier aimed to improve decision-making accuracy for buying wholesale energy. To achieve this, YLD created a model that reduced the industry error rate average by 0.86%. We achieved a 30% increase in payment forecast accuracy compared to the industry average for purchasing wholesale energy. Read more.
Final thoughts
The future favours those who embrace change and challenges by turning them into great opportunities. As leaders and innovators, we recognise that our journey with GenAI involves continuous learning, adapting, and strategically implementing innovative applications. If you need help with GenAI, data engineering, MLOps, data analysis, or data science, contact us.
