How to assess the AI-Readiness of Your Business?

Good questions inform, and great questions transform, is a quote by the bestselling author of The Proximity Principle (Ken Coleman). And true to this wisdom, some business owners/leaders have been asking questions like:

1. How can we use Artificial Intelligence to improve our value offerings?
2. How will AI deployment improve our business?
3. How do we start with Artificial Intelligence?

These are all great questions; with answers capable of (radically) transforming any business operation for good. If you’re already asking these questions, kudos! If not — Your journey starts here…

In this brief article, I will highlight the components of AI readiness and present simple ways to assess the AI readiness of your organization.

What is AI-Readiness?

It is an organization’s existing capacity to generate business value from artificial intelligence.

The keyword is business value. AI solutions have to move the needle — forward!

Therefore, the first step in assessing AI readiness is to identify AI opportunities, i.e., are there lynchpin points in your different workflows and business processes where AI can be of value?

For example, the logistics business in a supply chain may want to eliminate the risk of orders shipping late. This is a typical AI problem that can be solved with an ML (Machine Learning) algorithm that identifies and prioritizes the right shipments to ensure consistent ordering and shipping.

To actually identify the problem/opportunity above and design a solution, an effective interplay between two components of AI Readiness is required:

The Technical Component, which includes highly specialized AI talent (in-house or contracted) that works with organizational subject matter experts to:

• Define what problem you want to solve
• Determine what data is available to address the problem (internally and externally)
• For a hardware solution, determine how your machine learning or AI model is going to be deployed; a cloud data server somewhere or deployed on a constrained/embedded IoT device.

Hiring talent and working on the technical part of building out algorithms is the easy part of adopting?AI.

The business component: This is the most difficult part of adopting AI in any organization. This is where the rubber meets the road and organizational leadership is required to look at processes that they deal with on a daily basis and identify where there is value in having intelligent automation through AI or places where there are opportunities to realize additional values from the mining of data and using insights from there to then build out AI solutions to help them make faster or better-informed decisions.

• This component of AI-Readiness must be properly planned to build out a framework that can deliver value in the long term.

A quick look at some AI–Ready businesses shows that the common pattern for building a solid business component is to appoint a functional leader in the organization as the AI champion:

1. This AI Champion has to be a senior leader because the kind of problems that require AI solutions mostly have a serious business impact, so quick buy-in and resource allocation is essential.
2. The Champion sets the vision and inspires the rest of the organization on what outcomes they’re trying to pursue.
3. The AI Champion identifies who the business and technical stakeholders are going to be.
4. Having cherry-picked the company AI team, the champion works with them to discover some of the organization’s biggest problems and opportunities.
5. Focusing on where the business can move the needle with machine learning, the team narrows down a list of problems they can solve quickly and create an impact that inspires the organization to fully embrace AI.
6. Once the viability of machine learning has been proved on a pilot project, they can start thinking through the data strategy or data governance.
7. It pays to think early about data strategy; don’t do it for the first project. Start by thinking of where you can do great work using data from different silos in the organization and still get results.

Data governance (DG) is the process of managing the availability, usability, integrity and security of the data in enterprise systems, based on internal data standards and policies that also control data usage. Effective data governance ensures that data is consistent and trustworthy and doesn’t get misused.

A business leader that can grasp high-value use cases is critical to the success of AI deployment in any organization.

Contextual understanding of AI from functional leadership is important.

One fundamental question that the AI Champion and the AI Team can use to quickly discover opportunities for AI problem-solving is:

1. Where are the places where data-based decisions are made slowly (more time than necessary) or made inconsistently within our organization?

From the list of answers, the senior business leader, subject matter expert, and technical team members can decide the best opportunity for a use case.

Willingness to trust data

For artificial intelligence projects to succeed, there has to be a willingness to trust data in the organization — from top to bottom.

If this trust is non-existent, it can be learned by establishing data literacy in your business—this allows all stakeholders in the organization to appreciate data as a story. Subsequently, decisions made at the top of the organization must be based on data.

Benchmarking AI Readiness

In addition to culture and individual skills, the tactical state of data is a key consideration; i.e., how much Excel or PowerPoint are you using to run your business?

The less you have of excel and the more you have of systems, it’s a great indication that data is stored systematically, in some form that will enable the AI team to take advantage.

Also, there is consistency, which means that for people who are in similar roles, there is consistency in their data. The problem with using Excel is that your organization can’t have good data governance or data lineage.

The less you have of excel and the more you have of systems with quality data, the readier you are to deploy AI.

Do you consider your business or organisation to be AI-ready? If not, how much work do you think needs to be?done?