Now that you have a methodology for ensuring your Objectives and Key Results are screened against and informed by your purpose, let’s look at a leaner methodology to implement OKRs.
Back to basics: why are we using OKRs again? We use any measure of output to ensure our work creates the value we seek. Objectives and key results attempt to define that measure of success. By defining what success looks like, for individuals and for the team, you are sharing a treasure map. With a treasure map, the odds of arriving at the right place at the right time are generally greater than wandering at random.
Same here: if we define success, we may well achieve it. If we don’t define success, we won’t know either way.
What’s an objective?
In the world of setting forth the definition of success, it helps to be precise about definitions themselves. If we said our objective was to “crush it” (odd phrasing on a few levels) in our field, it’s likely we may not know if we in fact “crushed it.” However, if we said “be the leader” in our field, that may be a phrase we could define with some detail.
Our shortlist of objectives, usually numbering fewer than five, should sum to define our success. They should do this with fact-based language that can be clearly interpreted by all members of the team—from admins to sales to accounting and UX. Words like “world-class” and “coolest” are suspect, as they may be highly subjective.
Objectives often include a collection of success factors such as a measure of market success (“Become number one in market share,” or “become industry standard”) or internal metrics (“hit cost reduction plan” or “complete all engineering hires on time”). If met over the period of measurement, this collection of completed objectives should prove we’ve succeeded. A simple test is to ask: “Great. So, if we, in fact, met these goals and did nothing else, would we call this quarter/year a win?” If yes, move on. If not, build more specificity.
A word on inspiration in objective drafting. Some people believe objectives need to be inspirational. Others find keeping it clean and direct to better paint a picture for the team. As one of our CEOs said to me recently, “We found the tapestry that non-grandiose objectives wove was way more inspirational than any single objective.” I’ve seen both pathways work. In either method, be sure there’s no room for ambiguity.
What’s the difference between an objective and key result?
Key results are the details. Key results are where we can connect the aspirational statements of the objective with reality needed to truly measure the outcome. Key results are where the aspirational rubber meets the reality road.
The simple lexicon can be our “Objective is X, as measured by key results X1, X2, X3”. This “as measured by” phrase will come in handy when testing OKRs as they get assembled.
A good key result is specific, measurable, achievable, relevant and time-bound. While this age-old acronym of SMART goal setting is valuable, I propose an upgrade. Since all goals which are measurable are by definition “time-based,” let’s ditch the “T” and instead add “P” for purposeful. Now, with a little vowel shuffling, we have RAMPS. If you take one thing from this article, take this acronym. If your key results fit the acronym, and if they roll into objectives that truly define what success looks like, you’re off to the races.
- Relevant. This may sound obvious, but I have seen many excellent teams put up OKRs which were important, but less-than-relevant. For example, one OKR may be “successfully transition to agile development methodology as measured by 100% use of new dev platform.” This may be an important activity, but will it truly rise to be a measure of success? If we make the transition, but it ends up slowing us down instead of speeding us up, were we really measuring the right goal? Here, too, the negative screen is instructive: if we don’t hit this objective, can we still be “successful” during the measurement period?
- Achievable. Key results need to be at least achievable. Without that ability, teams will be demotivated, not motivated, by the process of setting goals. Some goals may be determined to be stretch goals, and perhaps only 70% of the target is expected as “plan,” but these details need to be specifically laid out in the OKRs. (Some people refer to this “A” as “Aspirational” in that setting.)
- Measurable. As discussed below in Specific, the key to results such as “48% market share” is that we must be sure to have a specific report to point to. Be sure this report will be available during the appropriate review period, or this KR will not, in fact, be measurable. Other customer-centric measures, such as “increase net promoter score from 32 to greater than 75” sounds very specific. However, don’t forget the “as measured by the external quarterly survey.” To be measurable, you’ll need to be time-bound. Annual, quarterly, monthly—the increments are up to you. Just be sure to specify one.
- Purposeful. Imagine you are a media company. Do your objectives and key results align with your purpose? Are you certain that “Open China [objective], as measured by $10M in first-quarter top-line sales [key result]” will allow you to stay true to your value of “Always support freedom of expression?” Google struggled with exactly this issue as its China market goals seemed to collide (according to some execs) with their value statement “Don’t be evil.”
- Specific. Objectives that are measured by non-specific key results will not work. Saying we will be a “leader in the space (objective), as measured by “beating the competition” is a fail. Without defining “beating” and “competition,” we will not know if we’ve ever met the target. However, saying we must “be a leader in the space, as measured by our ability to win greater than 48% market share, according to XYZ industry published the report” is a clear definition of success.
So, what makes these OKRs “lean?” The speed of development can drive making OKRs in start-ups a faster process. But I find these two tests to be instructive in keeping it lean—in development, but also in implementation.
- The One-Page Test. Your entire team-wide objectives and key results for up to four quarters (i.e., four different review periods) should fit on one page. This is not easy, however, by focusing on fewer than five objectives with fewer than five key results measuring each, you can create a grid for the year, and that should fit on one spreadsheet, or be cut and pasted into one presentation slide.
- The One-Day Test. With most organizations, assuming you have built out clear purpose and values statements, OKRs do not need to be more than a one-day process with an executive team. If the execs have done prep-work with their team, understand the basics, and have access to company measurements around financials and schedules, this work can be done in a workshop model. With your entire exec team working for a full day to align goals using the techniques laid out above and below, it’s possible to at least get to the one-page draft within a day. From there, a quick iteration of feedback from teams should bring together a final within a few days.
These concepts are of course easier said than done. They require the prep work to have clear and agreed to purpose and values. They require clearly understood financial plans, team growth options, financing information, etc. However, if you’re this far along in your exploration, I’m assuming these things are in place, or at least within reach.
Of course, even with this work done, there will be exceptions and tricks of the trade in implementation.
- Subjective measures. Some measurements will feel impossible to get right in “Measurability.” If you have two managers not playing well together, how is this resolution measured? If you have someone who clearly needs to upskill their “leadership,” how is this measured? At the end of the day, there are times when subjective (non-numerical) measures are the best option. Things like “as measured by the evaluation of the exec team” or “as measured by feedback gathered from at least five people across the team by your manager” are acceptable. These measurements are not perfect and can be misused. However, if the managers are doing their jobs correctly, these are acceptable.
- Pushing down a level. How far down do we take this? My rule of thumb is to roll this out over time starting with team-wide goals, pushing down with success. That is, as the team accepts the use of OKRs, sees the exec team continually referring to them and measuring success with them, different parts of the organization should be excited to push (or pull) these methods down into their own orgs. We have CEOs using OKRs only at the highest level, and others doing them with individuals. With our small team at Obvious, we use OKRs at the team-level as well as the individual level. For a hard-fast rule: always start at the top.
- Changes along the way. OKRs should be considered living documents. Depending on the ability of the management team, specifics of the usage level, and discipline, I recommend reviewing OKRs quarterly and considering making modifications (if absolutely necessary) at mid-year. Mis-set goals happen. These should be acknowledged, cataloged, and modified.
I find that keeping this effort lean helps with some basic constraining tools. I’ve created a simple template in Google Sheets which a number of our portfolio companies have stolen to make their own. In the next post, I’ll share the details of that document, as well as some rules of the road to keep you on track as you move forward.