Image by macrovector on Freepik
Data lies at the core of all business strategies. It has been generated at an unparalleled rate since the outbreak of the epidemic that has led the entire world to go digital. Organizations have experienced an exponential increase in footfall on their web properties and become data-rich.
Now, the majority of organizations want to tap into this data asset to improve their business further and gain a competitive edge. It can be done in multiple ways –
- Enriching user experience by understanding their pain points
- Understanding what users value in their products and services
- Identifying where they start dropping off?
- Finding the reasons for customer churn – is it page load problem or lack of better offers or pricing a concern, etc?
- Or, the business is focusing on improving internal operational efficiencies and would want to bring in cost-reduction measures driven by data.
Once the business objective is decided, the executives bake it into their strategic goals which is what the business wants to position itself in the next three to five years. Needless to say, it is a multi-year journey and requires careful planning to decide what initiatives should be picked in the current year and succeeding years thereafter to enable the business to reach its overall strategic goals.
Additionally, the strategist splits the yearly goals into each quarter’s goals by evaluating what set of tasks establishes the groundwork for the coming quarters. In short, the strategist along with the leaders from cross-disciplinary teams works out the dependencies to hash out the goals and objectives for each quarter.
There are so many initiatives that can be worked upon to meet business goals, but you might not know in advance which ones to pick to generate business value in the most optimized and efficient way. You can not work on all ideas and perform the post-analysis to assess the efficacy of each of them. It is expensive with respect to the time and resources to execute all the initiatives at once and assess their feasibility.
That’s where the strategy comes into the picture – it takes broad high-level goals and converts them into a fine-grained and well-defined roadmap in an endeavor to prioritize the key projects.
Many folks confuse strategy with goal setting, whereas in fact, it is a very well-laid-out plan that calls out the action items and successfully takes a strategy from the design stage through the execution.
Now that we understand what a strategy is, let us understand when is the right time to freeze the strategic asks and kick off the groundwork.
It’s a typical problem of a trade-off between how much time you spend to design the foolproof and most optimized strategy versus when to seal the design part and get started with the strategy execution. A large part of AI strategy draws its characteristics from the experimentative nature of AI projects. You may not have all the requirements, dependencies, and risks foreseen before you delve deep into it. Hence, an AI strategy assumes constant reprioritization and needs agility and the ability to recalibrate the project scope as it advances to the next stage.
Understandably, there comes a time when you have to start with a blueprint instead of waiting for all the information and inputs to be perfectly made available. It is also called a problem of cold start.
It is very difficult to pick an open statement and jump-start the project and the fear is valid too. What if you pick the wrong project at the very beginning – you might lose enthusiasm and get discouraged at the beginning of the journey itself. It might take six months or a year to iterate through hit and trial but after one or one-and-a-half years, your initiative has to start showing returns. After a point, it becomes very difficult to kill a project and start afresh with a new project.
Besides, a new project sometimes means an initial ramp-up concerning a new technology which needs to be strategized well in terms of having an in-house expert who can help onboard with new technology. Not just the technology adoption crucial, but the overall solution needs to be adopted well within the organization.
This takes us to our next point which highlights the need for evangelists who advocate for the need for your solution and also become your champions in this journey. It gives a significant confidence boost seeing the adoption of your solution within the enterprise.
An AI strategist is an orchestrator who works in collaboration with multiple teams including business, product, data science and analytics, and engineering.
Source: Image by tirachardz on Freepik
An effective AI strategy requires the strategist to be well aware of the workflows and key fundamental drivers of an AI project such as:
- Cognizant of the impact of AI solutions on the end user
- Understanding ethical concerns such as biased or unfair outcomes, or access to sensitive and private information
- Assessing if the process that needs to be modeled is repeatable and if data has patterns for the algorithms to learn from
- Is the proposed solution scalable in terms of technology?
- Does it cut across the horizontals and invite widespread adoption in terms of use cases?
- What components of the entire solution need development from scratch or can be outsourced?
- Accounting for what technical skills are required to implement the strategy – are they available in-house within the teams or need external training or hiring altogether?
The post explained the significance of a strategy and highlighted the key factors when designing an AI strategy specifically. It is intended to bring out the complexities and list down the fundamental pillars of building a successful AI strategy.
Vidhi Chugh is an AI strategist and a digital transformation leader working at the intersection of product, sciences, and engineering to build scalable machine learning systems. She is an award-winning innovation leader, an author, and an international speaker. She is on a mission to democratize machine learning and break the jargon for everyone to be a part of this transformation.
