Three Things You Must Do Before You Evaluate AI Solutions

DigitalGenius is publishing a series of articles to empower you to navigate through the world of AI, helping you onboard the right AI company that practically solves real problems.

Phill Brougham
Posted by Phill Brougham

Subscribe to the DigitalGenius Blog

Get the latest content from the intersection of artificial intelligence and customer service delivered right to your inbox.

Keep up to date with the latest insights
in artificial intelligence and customer service


Assessing AI solutionsThis is the first in a series of posts on the topic of empowering businesses to navigate the challenging and somewhat confusing world of AI. We want to help executives and buyers source, onboard and benefit from AI solutions that bring tangible results to their business.

Debates seldom resolve themselves as to the ultimate character of AI’s impact on society; however, no one seriously doubts that AI will form an integral and all-encompassing role in all our lives. Businesses must increasingly embrace and adopt AI solutions to compete and to get ahead of the curve in mastering real-world AI. Although AI might appear to be a black box, your approach in empowering your teams to evaluate available solutions definitely should not be. AI might well be at the end of the hype curve, yet it is increasingly becoming a standardized business solution.

Here are three things you must do to prepare your team in assessing AI-powered solutions.

1. Diagnose the pain, or don’t get a solution for a problem that doesn’t exist

Before embarking into the wide - and sometimes disorientating - market of AI solutions, providers and buzzwords(!), your team must understand what to look for. What are the problems, and what are the desired solutions to target? They should understand:

  • Where in the business is the pain felt? And when?
  • What should improvements yield? This will be important in business planning
  • How to prioritize these pain points? Which matter most to your business vision & strategy?
  • What experience do you want to build for customers, partners or vendors? Some solutions sound good but have unintended consequences

With clarity on these topics, your team will know exactly which problems exist in order of severity, and can start looking for providers with expertise in solving them.

2. Define your decision-making process, i.e. a repeatable approach to guide purchase

Once you start engaging vendors, they will start asking a plethora of questions about your business, and also about your decision-making process - can you blame them? They want to sufficiently illustrate they are the right partner based on your criteria. However, as useful as it is for vendors, it’s twice as useful for you. After engaging the third vendor your team will start losing track of who does what, and how well. Arm them with some fundamentals:

  • What are the key selection criteria? Breakdown the indispensables
  • Which metrics will count towards the selection criteria?
  • What will the budget be? And if you don’t know, how will you create a budget?
  • How will you, together with your partner, show value? How does this relate to company strategy?
  • What role with demos, trials or POCs play?

Don’t underestimate the power of benchmarks. Confused buyers slow down the process for everyone, and potentially miss great opportunities because they simply don’t know what they’re looking for. Help them out.

3. Assess Business Readiness, or determine the steps between now and onboarding

Any new technology, especially one so full of connotation as AI, requires change management as well as a consideration of technical, infrastructural and cultural readiness (for more on culture, check out our free downloadable guide on Preparing Your Team to Work with AI). With AI, some considerations are less obvious:

  • Are you, or is someone above you, ready to sponsor an AI project? These deployments do not drive themselves, and often require executive cover
  • What resources are available to launch and onboard this technology? Get as close of an assessment as possible from vendors about the scale of resources needed (and who will pay for them)
  • What data sources are accessible for AI model training? Where are they, and who owns them? Without this, of course, most AI projects will be dead on arrival
  • Who will be the internal project manager? What powers, resources and incentives will they have?
  • What staffing implications exist? While AI might change or eliminate certain jobs, they also create new roles: managing new processes, automation and expanding AI’s reach can be a full-time job, but can also deliver the ROI you need
  • What is the plan to socialize AI within the business? Your provider should have a view on this, but ultimately it’s up to you based on your unique business culture

This final topic on business readiness is perhaps the most important - you can evaluate and diagnose all you like, yet if you’re not ready to take the leap it will all be for naught, leading to disenchantment .

Parting Thoughts

Navigating this space is not easy. It’s littered with a myriad of providers and large funding rounds that make wonderful promises about this powerful technology. It’s easy to get led astray, or to merely evaluate AI for its own sake. However, if you approach this project with clear goals and consistency, the process will be simpler - and while it could take longer, you will find the right solution to match your needs, while exposing your company to AI early.

The series will continue next week; however, if in the meantime you would like to book a consultation on your AI posture and readiness, get in touch here.


Subscribe to the DigitalGenius Blog

Get the latest content from the intersection of artificial intelligence and customer service delivered right to your inbox.

Keep up to date with the latest insights
in artificial intelligence and customer service