Often times we read articles about how a fortune 500 company has embarked on a multi-million dollar Artificial Intelligence (AI) strategy. Very rarely do we ever hear how successful any firm has been in their chosen AI strategy. It is well known that vast number of AI or deep learning projects are bound to fail for various reasons ranging from unrealistic expectations from leadership to not having the right workforce to plan and execute on an agreed approach.
While there are several enterprises publicizing their adoption of AI, majority of these organizations are still continuing to pilot and/or deploying proof of concept solutions. It is important to understand that AI technologies are in their nascent stages with very few industries adopting the technologies early and trying to rapidly innovate their models. If you are one of the business leaders who is tasked to evaluate or chart out a AI road map in your organization, it is important to ask or get consensus on the following
- Does the organization’s leadership see an importance of AI based digital strategy?
- How realistic are your leadership’s expectations of what AI can do?
- Adopting AI requires a thoughtful and well planned change in the organization, and are they willing support that change?
One of major reasons for having a bad AI strategy is a result of not having a buy-in from the leadership. It is imperative that an organization’s leadership understands the role that AI is playing in the industry today and, agree on a role that it would be playing in your specific organization as well as how it should be shaped and nurtured.
Ingredients of a Successful Enterprise AI Strategy
Developing an AI strategy that supports your leadership vision, competitiveness in the market and relevance to your business operations is a critical challenge to overcome due to all the hype around AI. Few important aspects to considering at the time of building your AI Strategy:
Are you an technology company or a product/services company that wants to leverage AI? — Firms tend to embark on a technology strategy and loose track of their core competency. For example, if you are a financial institution, it is perfectly okay to continue to provide the best financial products/services to your customers and still adopt an AI based strategy to improve your product/services. Having an eye on your core competency allows you to make more informed decisions around buy vs build, technologies to consider, and/or solutions to consider. Don’t pivot your company’s core competency in the process building a AI strategy.
Are your problem statements and/or business use cases practical, realizable and quantifiable? — It is important to understand that every use case in itself has multiple layers of complexity across the workflow and automation slowly takes precedence, for that reason there is no such thing as solving for the most complex problem, or the “sexy” problem that would garner the most press. It is extremely important to identify real problem statements across your business/service lines. Problems that constitutes
- complex data/information sets
- multiple input sources of the data/information
- high variability and change in data
- operational workflow is labor/cost intensive
- accuracy and timeliness are equally critical
- an ability to make phased releases in an non-intrusive way
This allows you to make quick small pilots, simpler change management process and, show quick results to leadership. Giving an added benefit of paying for future AI projects.
Do you have the right talent that will be able to execute on the road map? — Every organization has extremely talented pool of individuals that are exceptional in their own ways. However, when embarking on an AI strategy, you want to ensure that you have right group of people who are open to challenges, ideas, motivated by a vision, and high propensity to learning. In other words, to realize the true value of the AI, it is important to build and nurture hybrid teams consisting of SMEs, engineers, individuals closer to the business, product managers, data analysts & scientist, devOps, customer support personal and more. Investing in the right talent and building an highly motivated team who doesn’t get dissuaded by failures will be a critical component of your strategy. What you don’t want create is a bunch of R&D teams that would work on synthetic data to prove technology works.
“Many executives are enthusiastic about the business potential of machine learning applications. But business leaders often overlook a key issue: To fully unlock the benefits of artificial intelligence, you’ll need to upgrade your people’s skills — and build an empowered, AI-savvy workforce.” — Jeanne Ross, principal research scientist, MIT CISR
How does your company embrace change? — Another key component to consider is the change brought by your AI based solution. Firms often think AI will immediately replace people, remove all manual activity and/or fully automate business/service line. The reality is that AI more often than not augments a worker than replace a worker. For example, introduction of intelligent/conversational AI bot. Having an automated chat bot over the phone or over the web doesn’t mitigate the fact that the customers are not going to ask complex multi dimensional questions. Customers are well informed and are much more technically adept than 10–15 years ago, however AI/ Machine Learning has not advanced to a stage where it can linguistically understand a multi-part question. What this means is training your teams to work along with technology vs getting replaced by technology. Empowering your teams to amplify the value of your products/services rather than increasing human labor dependency
While these are not the only factors to consider before embarking on your enterprise AI journey, it is important to take develop a sound strategy and plan for a long play vs a short sell. Don’t get wrapped around all the hype and or press, rather develop a plan that best fits your business strategy and effectively increases the value addition your product/services give to your customer. You should definitely act now but you should act smart.