Itamar has over 20 years of product management experience working with industry leaders such as Google and Microsoft. He’s now a consultant, speaker and writer.
Three Types of Launch
Part of the role of a product manager is to discover and help build products that have a high market impact. But how many of us can say that we launched a product or service that created a step-change in our industry?
Itamar highlights three types of product launches:
- Positive – where a product is launched and is incrementally improved upon. These are very rare.
- Negative – where launches are so bad they get rolled back.
- Middle – this is where most things launch and not much changes in the business. This is the space where we spend most of our time.
What is the point of working so hard on products that fail to make an impact? Itamar argues that it has to do with planning and execution.
The Planning Waterfall
In most companies, we try to plan the long-term future with an organisational strategy which feeds into a product strategy. This in turn feeds project plans and feature roadmaps, and the micro-planning of an agile product team. If we take a macro view of this process we see that nothing about it is agile.
The Planning Delusions
This planning execution creates two delusions:
- Stakeholders who live in the waterfall world think they can predict everything.
- The agile product team believes that if they keep on pushing small increments then everything will be okay.
At the centre of this is the product manager who is expected to convert everything from one world to another. They are tired, there’s not much time to talk to customers or to do research, and they can get demoralised because they aren’t moving the organisation forward.
Itamar’s framework has four levels of planning:
- Goals, which tell us what we are trying to achieve
- Ideas, which are hypothetical ways to achieve the goals
- Steps, that develop the ideas and test them
- Tasks, the things that fall out of the steps level to build
For this talk, Itamar focuses on the first three levels – Goals, Ideas and Steps
If you tell people where to go but not how to get there you’ll be amazed at the results. General George S. Patton
In tech, you can’t possibly predict everything. We need to set our teams up with a mission.
Good goals answer two questions:
- Where do we want to be and by when?
- How do we know when we are there?
This means we need to use metrics, and there are two types of metrics we need to focus on:
- Impact metrics. How do we capture value? Pick one good metric that measures how we capture value such as profit or revenue.
- Outcome metrics. How do we deliver value? We can use a North Star metric to measure how much value we deliver to the market: WhatsApp uses “sent messages” and Airbnb measures “nights booked”.
Everyone has ideas and most of these ideas are not very good. Even experts often misjudge what experiments will pay off. At Google and Bing, only 10-20% of experiments generate positive results.
If you want to have good ideas you must have many ideas. Most of them will be wrong, and what you have to learn is which ones to throw away. Linus Pauling
Scientists don’t lock themselves in a conference room to convince each other with opinions. They go out and test many ideas to look for evidence. So we shouldn’t decline all ideas, we should put them in a safe place – an Idea Bank. Then we can rank the ideas using the ICE (Impact, Confidence and Effort) scoring system.
At the steps stage, we should validate our ideas as quickly as we can to generate results. Then these results can then feed back into our Ideas Bank and update our ICE scores. Some ideas will go up, but most ideas will go down. Then we adjust our investment in our ideas accordingly.
At this stage people commonly make the mistake of building the “winner” once they’ve found it, confident that they have made the right decision based on evidence. To avoid this we need to break the project into smaller learning milestones.
The best thing about GIST is that it creates a clear thread from the goal of a project to the everyday task, giving the team a feeling of purpose. But it also democratises the planning process by giving developers and designers an opportunity to be involved in planning at an early stage. Read more here.