Often businesses and startups use AI in order to
produce insights and predictions specific to their industry
automate internal tasks that waste their and/or their team member's time
In this post we'll dig a bit deeper into each one of these general use-cases and discuss how you can get started thinking about each one in the context of your business.
Using AI to make industry-specific predictions
Suppose you could use AI to make industry-specific predictions. What kinds of tasks could you solve, and what this would accomplish for your firm? These tasks / accomplishments could look like the following:
If I could reliably predict “thing X” in my industry then I could improve customers’ experience with my “product or service Y” or "improve lead generation / conversion with my marketing funnel Y”
What sorts of specific questions like these could you posit about your industry? These can relate to key industry metrics, customer profiles, the sky's the limit.
Just remember, AI needs data to mimic - or “ape” - the sort of industry-specific prediction tasks you might want to solve. It isn't magic.
If you have industry-specific problems in mind ask yourself: do you have data relevant to automate these tasks? If not, what kind of data do you think you would need?
Erasing internal bottlenecks using AI
AI is often used to automate internal bottlenecks many businesses have that involve low-level and repetitive cognitive tasks, ones that consistently eat up sizable quantities of high-skilled workers’ time.
Such AI automate-able tasks are usually driven by a need to organize data internally - either for the benefit of human users or in order to satisfy certain baked-in system requirements - and are often done “by hand.” Examples of these sorts of processes include
regular cataloging / organizing of digital assets based roughly on their content
merging of multiple files based on the similarity of numerical or text-based entries
summarizing of large quantities of text for organizational purposes
Does your business have any of these kinds of internal bottlenecks? If yes, jot down the sort of data and simple cognitive task is involved with each. Engineered appropriately, an AI solution might be just what you need to unclog these sorts of bottlenecks.
Now suppose you could use AI to tackle internal tasks listed above, qualitatively. What do you think this would accomplish for your firm? These sorts of wins could look like the following:
If I had an AI-driven automation tool for doing “thing X” I could make “process Y” in my company Z% more efficient, or remove an expensive “bottleneck Y” from my work process
Once again here, AI needs data to mimic - or “ape” - the sort of internal tasks you might want to automate. So ask yourself: do you have data relevant to automate these tasks?
Having thought about the questions mentioned above and jotted down your initial thoughts, you're on your way to drafting the first version of your "AI Strategy" - a living document that helps you organize and prioritize your business's growing AI interests - which we outline in the next post.