When I was a newbie, I'd I often forget the difference between a KPI and an OKR, so let's start with that.

OKR = Objectives and Key Results

A popular leadership process for setting, communicating and monitoring quarterly goals and results in organizations.

Intended to unite a company, team, and individual with common goals.

OKRs consist of a list of 3-5 high-level objectives.

Under each objective, there should be 3-4 measurable results, not more. OKR results can be based on growth, performance, revenue or engagement. Often they are numerical, but they can also show if something is done or undone, so a binary 0 or 1. These can also be milestones.

KPI = Key Performance Indicator

Key performance indicators determine specific factors needed to achieve success. These are measures that must be watched. They are not goals, but they can be associated with goals. They get specific, and they must be measureable. (Although, once a measure becomes a goal/target–you must question it's validity relevance.)

KPIs are important, and few.

There may be measures we track, but only few KPIs.

A few KPI examples:


  • Accounting costs
  • Accounts payable turnover
  • Actual expenses
  • Budgeted expenses
  • Gross profit
  • Gross profit margin
  • Net income
  • Total sales
  • Variable costs

Customer Service

  • Agent's full-time employees (FTEs) as percentage of total call center FTEs
  • Average after-call work time
  • Average number of calls/service request per handler
  • Average queue time of incoming phone calls
  • Email backlog
  • Number of complaints


  • Ad click-through ratio (CTR) clicks ÷ impressions
  • Brand credibility
  • Brand strength
  • Cost per lead
  • Response rate


  • Actual calls
  • Average deal size
  • Call quota
  • Closed sales
  • Customer churn ratio

Information Technology

  • Account create success
  • Alert-to-ticket ratio
  • Network availability
  • Server growth rate
  • Web server availability
  • Software development quality


  • On-time orders
  • Open orders
  • Percentage increase in productivity
  • Process capability


Starting fresh? Basic metrics are just fine. Call them vanity metrics. It's fine.

In general, you'll want to understand

  • Product usage/adoption (Daily/Monthly Average Users)
  • Churn rate
  • % of users who take a specific action

These tend to be lagging indiciators. But how do we understand the customer behavior ahead of time, find our leading indicators, and improve the product experience?

How do we actually measure a success feature launch?

  • How do we measure success? How do we know that our users are appreciating this feature?
  • How do we know when our user gets frustrated with this feature?
  • What is the goal for this feature?
  • Who are our user segments?

But you won't be using these to actually improve your product experience...

So, how do you actually improve the product experience? Measure it. Measure the experience.

Get specific, down to the type of user and their pain points.


  1. Understanding the measurement’s purpose
  2. Mapping Measurable Results
  3. Designing Meaningful Measures
  4. Building Buy-in to Measures
  5. Implementing Measures
  6. Reporting Performance Measures
  7. Interpreting Signals from Measures

Primary role for a product manager is

  • Understand the problem
  • Empathy for the customer, understanding the the technology, etc. etc. etc...
  • Being able to communicate the problem

Understand the customer's pain points and build measures around them