Build, Buy or Wait: Sourcing AI Tools
Decide whether to buy an AI product, build on a model, extend what you own, or hold off
A taste of a lesson
Our developer says building our own document summariser on an API would cost almost nothing. Is that right?
The model usage may well be cheap. The build is not just usage, though. Someone has to write and secure the integration, test quality on your documents, handle model changes, fix it when it breaks and support users. If that developer spends a few days a month on it, that time is the real cost. So compare like with like: vendor fees plus review time versus usage plus build and upkeep time. Quick exercise: estimate how many hours a month your developer would spend maintaining it in year two.
Written by the teacher as an example. In your lesson the tutor answers your own questions, and like any AI it can be wrong.
What you will be able to do
- Compare buy, extend, build and wait on cost, control, speed and risk
- Estimate total cost of ownership including people time and maintenance
- Identify lock in risks and plan a realistic exit route
- Write a fair one page decision memo with clear revisit triggers
Lesson plan
- 1 Define the need before the options State the job, users, volume and constraints precisely enough to compare routes fairly. Start
- 2 The four routes compared Understand what buying, extending, building and waiting each give and cost you. Start
- 3 Total cost of ownership Build a cost comparison that counts people, maintenance and switching, not only fees. Start
- 4 Data, control and lock in Check where data goes under each route and how hard it would be to leave. Start
- 5 Testing before deciding Design a short, fair head to head test on your own examples. Start
- 6 The decision memo Write a one page recommendation that a sceptical reader would find fair. Start
Try asking
About this tutor
For managers, heads of department and technical leads facing a sourcing decision: an off the shelf AI product, an AI feature inside software you already pay for, a custom build on a model provider, or waiting. You will learn the real differences in cost, control, speed, data handling and lock in between these routes, and why the honest answer is often a mix. We work through your own decision step by step: define the need, list the options, estimate total cost of ownership, check where your data goes, and plan an exit. You finish with a short decision memo that lays out the options fairly, states your recommendation and names the signals that would make you revisit it.
Reviews
4.3
3 ratingsSample
- Kwame A.Sample
Good structure and very even handed. I would have liked a worked example closer to manufacturing, but the memo template is now in our shared drive.
- Ines V.Sample
The cost of ownership table changed our conversation. We had compared a licence to API usage only and forgot the engineer's time completely.
- Lena F.Sample
I did not expect 'extend what you own' to be the answer. We turned on features in software we already had and dropped a new purchase.
About the teacher
Helps leaders choose AI work that pays for itself and drop the projects that do not
9 tutors 353 lessons taught Sample
I teach managers and executives how to make sound decisions about AI without needing to become engineers. My background is in strategy and operations work inside companies, where I spent a lot of time turning vague ambitions into projects with a budget, an owner and a way to tell if they worked. I teach with real decisions: a use case...
See Sofia's profile and tutorsMore like this
Other tutors on the same or nearby topics.