I have a love-hate relationship with OKR. If you love OKR, skip the second section.

The good

Used well, OKR can provide a framework for productive communication. Priorities can be discussed, and reasonable goals can be set; productive communication at least has a chance of leading to productive work. The length of time is also useful (not too short and not too long) but beware of Parkinson’s Law. The critical question is, “Are we doing the right thing?” If the answer is “yes”, or “no”, forget OKR, you have a more fundamental problem (pride and/or stupidity). If the answer is “probably”, then OKR can be used to manage workload and keep a team on the same page. OKR can be easily misused to manage ashes (how long is your backlog?) whereas its chief benefit is in starting new fires, even small ones.

The bad

It’s a great tool for talking about work and, in case there’s any time left for actual work, measuring that which was talked about, then talking about the measurements. And repeat. The potential for self-perpetuating, useless meetings is high, and is

  1. Directly proportional to the length and impenetrability of the set’s “wording”
  2. Indirectly proportional to the clarity of the key results

There’s the rub: clear definition can take ages (especially when more than one person is involved) and there is no guarantee that the work done is of any better quality, because—guess what?—I would have done the work anyway. Facta non verba. And, because having measurable goals inevitably focuses work on achieving them, subtle and abstract nuances that might matter a lot could be lost (or worse, willingly ignored).

If all my customers are pissed off and I “increase customer satisfaction by 10%”, they are probably still pissed off.

Much more beneficial would be defining what is valuable and meaningful to the organisation and building on that before creating new structures, but the latter is easier; defining value is hard. It must by nature be unique to every organisation and it is (much) more than profit or EBIT or a 10% increase in a something-KPI.

Improved communication across the organisation

Everyone potentially knows what everyone else is doing and all dependencies have been identified. Firstly, gatekeepers still exist and the interesting, secret (valuable?) stuff stays secret. Secondly, people are either too busy rearranging their own backlogs to take notice, or they take the time to “know everything” and start interfering where they shouldn’t. A little knowledge is a dangerous thing.

A fundamental restructuring of the organisation

OKR breaks down stuffy, old-hat organisational silos and creates sexier modern ones—functional silos are augmented with OKR-dialects. The whole organisation breaks down into those who understand OKR, those who don’t, and those who don’t care, although the boundaries are fuzzy. A whole new subset of people is created to take part in a whole new subset of meetings. These people have bureaucratic responsibility and wholly no authority whatsoever, intentionally!! Thus the OKR process is dissected, perfected, talked to death and resurrected, while the real work is being done by others.

Flexibility

I can strong-arm a complex problem into a simple one-dimensional goal, or I can straitjacket the simplest of problems into flowery adjective-laden prose. In both cases, any real progress is stymied.

And the gardening

It might sound like I’m anti-OKR until now, but that is not the case. I am anti-time-wasting, anti-bells-and-whistles, things like that. OKR becomes powerful when there is no organisational ambiguity, no misuse as a signalling device, and no unnecessary meetings; in short, when I use it on myself. It combines two scourges of modern life and makes the whole greater than the sum of its parts, namely:

  • The wish list good stuff I want to do and never get round to.
  • The todo list annoying stuff I have to do (all the time).

After a pilot phase in March (really), I gave private OKRs a proper shot in Q2, which is where the landscape gardening comes in.
The need (wish) for a sandpit revamp in our garden was floating around for ages before a confluence of circumstances pushed the project headlong into realisation:

  • It stopped raining, briefly. Remember that?!
  • A pile of logs desperately seeking a practical use
  • The wasteland of the previous sandpit becoming unbearable
  • Clamouring children
  • An itch to dig holes and move heavy loads to compensate for all that sedentary office work

The wording of my first set was crap, but just having something written down was enough because the “customer requirement” (kids playing happily) was clear. The near term deadline added the requisite motivation to keep chipping away. Also, my deep knowledge of creating meandering, disjointed todo lists helped me fashion a focused, cascading series of tasks ultimately leading to—tada!—a very nearly completed sandpit (result) and kids delightedly throwing sand at each other (objective).

I record todos and track progress in Organice

orga

Still, there will always be the question, do we succeed because of OKR, or in spite of it? At any rate, my Q3 set is much shorter and more focused; for OKR, unarguably, less is more. While writing this, I thought of great creative endeavours, undeniable genre-breaking, life-changing stuff…The Beatles didn’t use OKR to create Sgt. Pepper’s, but by the time they got to Let It Be, it might have helped.

Dig a pony

digging

sandpit

The inspiration behind my experimentation was this article in HBR