Start with the work, rather than the tool
A leadership team can spend weeks comparing AI platforms without agreeing what needs to improve. A more useful opening question is: where does the business repeatedly lose time, information or momentum? Look for documents that need retyping, questions that require searching several systems, or reports that take too long to prepare. Speak to the people doing that work. Their workarounds often reveal more than a list of software features.
Make the opportunity specific
Replace broad ambitions such as ‘use AI in customer service’ with a bounded task: help support staff find approved product information and draft a response for review. Define who uses it, what information it needs and what the output should be. A specific task makes it easier to compare AI with alternatives, including a process change or a feature you already own. It also exposes constraints before they become expensive surprises.
Check five things before investing
Assess frequency: does the task happen often enough to justify change? Assess friction: what time, delay or rework does it create? Check the data: is the required information accessible, current and suitable? Consider consequences: what happens if the answer is wrong? Finally, identify an owner who can make decisions and support adoption. An attractive idea without usable information or an engaged owner may need preparation before development.
Build a shortlist you can defend
Rank opportunities by expected business value, delivery effort and risk. Record assumptions rather than turning guesses into precise-looking scores. A high-volume, low-consequence task with clear review steps may be a better first project than an impressive but poorly defined transformation. Keep the shortlist small enough to act on. Decide which opportunities to test, which foundations to improve and which ideas to leave for later.
Include safeguards in the plan
Responsibility is part of readiness. Agree what data may be used, who can access it and which decisions require human review. NIST’s voluntary AI Risk Management Framework organises risk work around Govern, Map, Measure and Manage. For a business starting out, the practical lesson is to consider responsibilities, context and evaluation alongside the opportunity, rather than after launch.
How AiQ helps
AiQ’s advisory and strategy work connects business goals with practical technology choices. A readiness engagement can review workflows, assess data and systems, prioritise opportunities and develop an actionable roadmap. The aim is an informed next decision: a focused pilot, a training programme, better foundations or a simpler solution. You do not need a complete AI strategy before the first conversation. Bring one process you would like to improve.