ARTICLE: practical steps to be ready for applying ARTIFICIAL INTELLIGENCE

Building upon our introductory article, this article dives deeper into practical steps you should take to be ready to implement Artificial Intelligence (AI) within your organisation.

Starting your AI journey on the right foot is crucial to its success. 

Where to Start?

AI needs to be put to work to solve a real business challenge. It is another tool in the toolkit of senior managers and fundamentally it will only add value where it is deployed to fix a meaningful challenge.

Start with your business strategy. Identify the areas of your business offering the biggest opportunities for improvement or growth. Have a clear process for identifying and prioritising candidates (we talk more about this in part 3 of this series).

Most businesses face seven key challenges when implementing AI:

1. Data

Accessing the appropriate quality and volume of data is the foundation for effective AI training. The more comprehensive and accurate this data set, the better your AI tool can learn and perform

2. Ethics

Ensuring that the tools you are integrating align with the legal requirements and values of the location in which you are deploying them. How you use customer data and managing the potential for bias are key considerations 


As with all change projects getting the right staff and customer engagement is the number one enabler to successful implementation

4. Information Security

Choose AI tools that keep your customer and company information and financials secure

5. Controls

Understand your risk appetite and develop the  controls to monitor the AI tools and ensure they are working as expected

6. Governance and Board Knowledge

In order to achieve appropriate levels of control, proper governance and board knowledge are needed to ensure decisions are made in the interests of both customers and shareholders

7. Skills and Knowledge

For most organisations you will not need technical expertise in creating base models but you do need the right capability to oversee the services you are bringing in from third parties 

Support act or main event? 

When looking at initial use cases there are various roles that AI can play to fit your needs. 

The diagram below illustrates the three potential roles of AI. The three roles describe the increasing levels of autonomy for artificial intelligence (AI) systems within an organisation.


Some products are designed to only ever fulfil role 1 … for example, agent assist tools that recommend next best actions and summarise call notes … whilst others are designed to mature with your organisation. for example, tools that offer a fully automated chatbot or automated task processing.

Some AI applications are already at role 3, for example fraud detection software which is constantly evolving criteria to reflect new risks. 

Other exciting capabilities like using Large Language Models to script responses to customer queries, may be suitable for role 1 or 2 application but given their fallibility are probably not ready to act in a role 3 capacity today. 

What is the Right Delivery Method?

There are three broad paths available to implement artificial intelligence capability.


The first is to seek out AI modules or add-ons within existing technology solutions packaged by the original supplier. The simplest path to start your AI journey is to seek out the AI capability within your existing partners. 

For Enterprise organisations the broadest example of this is the rollout of Microsoft Co-Pilot and Google Duet. Other examples would be enhancements in contact centre applications that automatically produce a summarised call or AI enhancements in finance software that automatically reconciles invoices with banking receipts. Effectively you are extending the scope of existing service providers and taking up their AI offering. 


The second option is to introduce capability by integrating new products or functionality. This could be a fully packaged product for example dynamic pricing capability that continuously optimises pricing in retail stores or AI capability that introduces additional capability to your existing applications, for example an API to Chat GPT that improves the language processing of your chatbot.


The third approach would be to build capability in-house and is likely only applicable with use cases that are very specific to the organisation or areas where you seek to build a unique capability and competitive advantage. Even in this case you would likely look to draw on the existing library of tools available in various marketplaces, e.g. Hugging Face’s market of open-source models


Given the potential prize associated with implementing Artificial Intelligence capabilities staying on or ahead of the curve will offer significant competitive advantage.  

A number of critical factors are making AI viable now for deployment in many businesses: 

  1. Cloud computing power/ cost – the abundance of storage and compute power has made the basic infrastructure accessible
  2. Cross disciplinary work – cross AI disciplinary work has significantly increased the potential applications of AI
  3. Integration – the provision of existing AI capabilities either integrated into existing tools you use or as API’s that can be integrated into existing tools
  4. Data – the volume and quality of data organisations have available

With the approach to readiness set out in this article a controlled test and learn approach will allow a steady paced transition rather than a panicked chase to catch your competitors.  


The current landscape offers great opportunities to realise efficiencies by deploying AI in your business. The important point is to remember to treat it like any other technology transformation:

  • Identify a clear business case
  • Implement it in a controlled fashion
  • Monitor and evaluate its performance


Can’t wait? Talk to one of the team about where to start, contact or