First, the bad news: there is no singular framework or set of checkboxes that cover the bases of what you’ll be exposed to as a PM. A big part of the job is problem-solving and navigating complexity.

Second, the good news: frameworks are a useful tool for just that! They help you problem solve and navigate complexity. But they’re (unfortunately) not an answer to the ultimate laws of the universe.

Product is an art, not a science. Your goal should not be to pick the perfect framework but instead to strengthen your product sense using frameworks as good training-wheels.

That said, in choosing any framework to use, apply some product thinking skills to take some steps in the right direction:

• First, understand the problem you’re trying solve by using a framework in the first place
• Then, explore which solutions (frameworks) may solve that problem, and decide on one
• After, tweak the framework for your unique context
• Lastly, focus on getting in reps to gain intuition

Let’s walk through each of these in less theoretical terms.

Understand the metrics problem (before picking a framework)

There are many frameworks around product metrics because product metrics are used for many different purposes. For example, I’ve used metrics for all of the following:

• Sizing and highlighting potential opportunities
• Setting goals
• Tracking strategic impact
• Understanding usage
• Proactively monitoring changes
• Developing confidence in a solution

Unfortunately, there’s no singular framework that’s going to solve all of those for you in a nice, tight way.

But not to worry — just as you’d pick a hammer for a nail, and a shoe for your foot, you can pick a ”works well-enough” framework for whichever metrics problem you’re trying to solve. You just first need to know what it is.

For example, when opportunity sizing, you may decide to use a framework like RICE, which may help you to prioritize and compare different opportunities.

Or, for developing more confidence in a solution, you may use a painted door test simply to validate and track interest in a new feature.

Or, for tracking strategic impact, you may pick a few key engagement metrics to monitor over time.

Or, for setting goals, you may use the OKR or NCT frameworks.Bottom line: whatever framework you choose should be related to a real goal or problem you’re trying to solve.

Explore frameworks that might fit that metrics problem

Once you’ve clarified what your real goal is, that hones in the solution space a bit more.

For example, let’s say I’m working on building a new feature that allows Amazon college student customers to share their cart with a parent so the parent can pay directly without needing to share credit card information with their kid.

After shipping this feature, my goal as a PM is trying to understand and measure feature usage, so I search around and find Reforge’s TARS framework, which helps me with the following:

• Target: What % of the targeted population saw this new feature
• Adopted: What % of them tried it successfully/unsuccessfully
• Retained: How does successful usage of this feature impact monthly retention for this college age population
• Satisfied: How does successful usage of this feature impact overall customer satisfaction

This is a great starting point. And most frameworks around measuring feature usage cover 80% of the same ground (e.g, are people using it? What segments? How do those that use it retain compared to other users, etc).

But there’s so much more that I may know about these Amazon customers that might also govern how I think about feature usage and success.

The Pareto Principle (80-20) is relevant here: it’s likely any framework you use misses key information that’s unique to my company or its customers but may matter most, which leads to the next step.

Tweak the framework 20% by recognizing what it doesn’t do

In our Amazon example above, the TARS framework helps me evaluate usage at a basic level…but it may miss some key things that I know based on context. For example:

• There are certain times of the year where this kind of feature is more important than others (e.g, at the beginning of a semester when students need to buy new books or travel if they live away from home). Feature usage is likely going to be directly tied to seasonality.
• Because students are on mobile devices more frequently, but parents are often more comfortable making purchase decisions on desktop, we might need to build and test across-platforms to ensure that usage isn’t low for that reason alone. Feature usage is likely going to be directly tied to platform access.

These two bits of context may matter significantly to the feature’s successful adoption. There’s always context happening around you that doesn’t get captured in a nice framework.

For example, the RICE framework may be able to help you prioritize opportunities in a vacuum, but it doesn’t account for complexity like:

• Which opportunities have x-team dependencies?
• How bought-in to this opportunity is the leadership team?
• Does your team have the right people to actual implement this kind of solution?

Similarly, OKRs are a useful framework for aligning teams around what matters, but the framework isn’t that great for:

• Weekly tracking of your team’s progress
• Sizing opportunities
• Understanding usage

The point is this: even if you have a great framework, it’s just a starting point. The framework doesn’t have the answer for you…it simply has some guiding principles that can help you think through the problem a bit more.

This is a tough part of PMing! The complexity of human decision-making doesn’t go away. You just get more comfortable with it over time.

Focus on getting successful reps to gain intuition

Lastly, I want to reiterate that your goal should not be to pick a perfect framework. Instead, it should be to develop great product sense and intuition over time, with frameworks as your training wheels.

It’s better to focus on using one framework successfully a few times and getting some reps in so that you can understand how it works, what it’s good for, and where it fails. This will help you to develop pattern-matching intuition that will make solving the next problem easier.

As long as you’re clear on what you’re trying to use the framework for, think of it as a sandbox you can play in and learn from. You can always step out, try something else, and start new again.