As a Business User

If you're considering integrating LLMs, the first question your investors or team will ask you is "why?".

At LogicLoop, there were three main reasons we were attracted to the promise of LLMs, increasing our TAM, improving our user experience and increasing our ROI.

To Increase TAM

Our core business alerting platform needed users to write SQL, a specialized skill set that prevented most business operations teams from becoming independent with the product.

Our hypothesis was that a natural language to SQL module would help expand usage at existing customers, and attract new customers that didn’t have technical business teams.

To Improve User Experience

Furthermore, writing SQL takes time. You first have to come up with an idea, write SQL, debug it and then optimize it. Even for professionals, new queries can take a week or more to formalize and productionize.

Our hope was that data analysts would automate the rote aspects of query writing, like boilerplate join templates. An improved user experience should translate to greater usage, and hence more revenue.

To Increase ROI

Part of LogicLoop's value prop is reducing the amount of time and effort it takes for operations workflows to be built. By reducing that time further, we can show a greater difference in the pre and post-LogicLoop scenarios, driving a better ROI argument.

Similarly, as you’re evaluating whether to add LLM-based features, consider the business goals you want to achieve first, aka the Jobs to be Done.

Simply adding LLM-features because you believe it will increase your company valuation or hype might work short-term, but won't work long-term. You’ll need to quickly leverage that Level 1 thinking into something that will build sustained revenue and differentiation for your business. LLMs should not have the power to make or break your entire business model. In other words,

LLMs are part of your solution statement, not your problem statement.