What are outcomes?
A Clear Definition That Helps Everyone Understand
As product managers, we often talk about outcomes and outputs. Think of outputs as the things we build - like new features in our software. Outcomes, on the other hand, are the real changes we want to make happen.
In my years of leading product teams, I've seen something again and again: many people find it hard to think about outcomes. When teams write user stories, plan their roadmaps, or set their OKRs (Objectives and Key Results), they often struggle to define good outcomes. Instead, they usually end up writing lists of things they want to build.
I recently found one of the best ways to think about outcomes in the book called "Who does what by how much?" by Jeff Gothelf and Josh Seiden (Sense & Respond Press, 2024). They explain it very simply:
"An outcome is a change in customer behavior."
This helps us understand what an outcome is NOT:
It's not something you deliver or build
It's not just launching a new feature
It's not just doing tasks on your list
An outcome is about how your users or customers actually change the way they do things when using your product.
The authors make an important point about OKRs too: every Key Result should be an outcome. If your Key Results are just about building things (outputs), you're missing the real goal of product development - creating meaningful change for your users.
Which LLM? → All of them!
When I use a chatbot AI, I use whatever fits best to my use case. With Abacus ChatLLM, I choose between ChatGPT-5, Claude 4, Gemini, Perplexity, Grok, Llama… whatever I need, for less than a ChatGPT subscription.
[I am not affiliated with Abacus, but I am a convinced subscriber. If you use the link and subscribe, I will receive a referral fee.]
What I read
As usual, I will list some of the best articles I read on the Internet. I will keep a list of the best articles (currently >800) at https://www.digital-product-management.com. These are today’s picks:
Software engineering laws: Parkinson's law, Conway’s law, Zawinski's law, and 10 others, all in one place!
Deep Dive Into MCP and the Future of AI Tooling: What MCP is, how it changes the way AI interacts with tools, what developers are already building with it, and the challenges that still need solving.
B2B Customer Segmentation: How B2B product teams can segment their customer data.
I must say that I disagree a wee bit on your take on outcome. I believe Jeff Patton says it best: outcome is what comes out after you deliver an output; e.g. what users/customers say, do and feel after using your product. Disagree?
That’s a good point. I can only confirm that. It’s not uncommon that, in the beginning, the goal was to fundamentally improve something. At some point, all that’s left is a new product that is different from the old one but brings no substantial improvement.