Objectives and Key Results (OKR) is a goal-setting framework that pairs a qualitative Objective (what you're trying to achieve) with measurable Key Results (the outcomes that prove the objective was achieved). Popularized by Intel and Google, OKRs are widely used in technology and ecommerce companies as a way to connect strategy to execution and align teams around shared outcomes.

The structure of a good OKR

An OKR has two layers:

Objective: a clear, qualitative statement of the goal. Inspirational, time-bound, and ambitious. Example: "Become the default skincare brand for sensitive-skin customers in the US."
Key Results: 3–5 measurable outcomes that, taken together, prove the objective was achieved. Each KR has a number and a deadline. Example KRs for the objective above: "Reach 250,000 sensitive-skin segment customers by Q4," "Achieve 4.7+ average rating across hero products," "Generate 40% of new-customer revenue from sensitive-skin marketing campaigns."

Why OKRs are worth the quarter

A small team can genuinely move two or three numbers in a quarter. The cost of refusing to choose is the same payroll spread across ten initiatives that each move nothing, and that cost stays invisible because everything shipped on time. For an ecommerce brand the discipline that pays is forcing every Key Result to name a number that eventually shows up in the accounts — repeat purchase rate, contribution margin, first-order acquisition cost — rather than a launch date. A feature delivered on schedule that changes no number has still consumed a quarter of a team's salaries. The trade-off is uncomfortable in the planning meeting: naming three Key Results means saying out loud which projects are not being done this quarter, and most organizations would rather keep the list long and the accountability vague.

What separates good OKRs from bad ones

Outcomes, not activities. "Launch new product page" is an activity. "Lift product page conversion rate from 2.1% to 3.0%" is an outcome. Activities are what you do; outcomes are what changes as a result. OKRs measure outcomes.
Ambitious but not impossible. Healthy OKRs are designed so a 70–80% achievement rate is considered success. Hitting 100% on every OKR usually means the targets were too easy.
Connected to strategy. OKRs cascade from company strategy. Department or team OKRs that don't ladder up to a top-level objective produce activity without alignment.
Limited in number. Three to five OKRs per team per quarter is a working maximum. Beyond that, focus disappears and execution suffers.
Public and reviewed. OKRs that are documented and discussed regularly drive accountability; OKRs filed away in a quarterly planning doc don't.

OKR cycle

Planning: typically quarterly. Company OKRs set first, then department OKRs cascade, then team OKRs.
Mid-quarter check-in: roughly 4–6 weeks in. Where is the OKR pacing? What's blocking progress? Adjust if needed.
Quarterly review: what was achieved, what wasn't, why. The retrospective drives the next planning cycle.
Annual planning: ties quarterly OKRs to broader yearly strategy.

OKR vs. KPI vs. balanced scorecard

KPI: ongoing metric tied to operational performance. Measured continuously.
OKR: goal-setting framework with a defined time horizon (usually quarterly). Combines qualitative objective with measurable Key Results.
Balanced scorecard: performance-management framework that tracks four perspectives (financial, customer, internal process, learning and growth). Older framework with similar intent.

The frameworks aren't mutually exclusive. Many companies use OKRs for goal-setting and KPIs for ongoing performance measurement, with Key Results often being specific KPI targets.

Common OKR mistakes

Treating activities as Key Results. The most common error. "Launch X" or "build Y" are tasks, not outcomes. Real KRs measure what changes when the task is done.
Setting too many. Six or seven OKRs per team produces dilution. Three is often enough; five is the practical maximum.
Sandbagging. Setting easy targets to hit 100% defeats the framework. Healthy OKRs are designed to fail 20–30% of the time.
No mid-quarter visibility. OKRs that aren't reviewed mid-cycle produce surprises at the quarter-end review. Regular check-ins surface problems early.
Cascading too rigidly. Strict top-down OKR cascades produce alignment but kill bottom-up energy. Healthy OKR systems balance top-down direction with team-level ownership of how to achieve it.

The step most teams skip is the baseline. A Key Result like "lift repeat purchase rate to 32%" means nothing if nobody can state today's number and name what moves it, and the quarter gets spent arguing about measurement instead of doing the work. Establishing those baselines is ordinary ecommerce audit and strategy work, and it belongs before the planning session, not after it. Deciding which of them deserve a quarter of attention is ecommerce marketing strategy.

A Key Result is only as good as the metric behind it, and choosing among conversion rate, AOV, LTV, CAC, and the rest requires knowing what each one actually measures and what a healthy number looks like — the reference point laid out in this rundown of the ecommerce KPIs worth tracking.