How we failed at setting goals

Published on Nov 28th, 2018 by Ryan McKay-Fleming

As we’re putting together our plan for Chalk for the year, I’ve been thinking a lot about goal setting. We’ve been building Chalk for over six years now. In that time we’ve tried to formalize company-wide goals twice. The problem is, we never continued after a single quarter of using them.

OKRs

The system we tried was Objectives and Key Results (OKRs). It’s a pretty simple system. Your goal is the objective and key results are how you measure your progress towards the goal. Here’s an example that John Doerr used to introduce them to Google.

Objective: Teach you OKRs
Key Results:

  • Show Example Use Case
  • Highlight Benefits
  • Provide links to additional resources

He then went on to show how they apply to an organization. The leadership sets organization-wide OKRs, and they cascade down the hierarchy.

Football GM Objective: Make money for Owners
Key Results:

  • Win Super Bowl
  • Fill Stands to 88%

Head Coach Objective: Win Super Bowl
Key Results:

  • 200-yards passing
  • No. 3 in defence stats
  • Average 25-yard punt return

Public Relations Objective: Fill Stands to 88%
Key Results:

  • Hire two colourful players
  • Highlight key players

Our First Few Attempts

That seemed simple enough, so here are some OKRs we set for ourselves:

Objective: Reach $1M in Annual Recurring Revenue
Key Results:

  • Close 20 deals
  • Increase ACV to $10K

Objective: Improve Planboard Set Up
Key Results:

  • Convert Planboard to Single Page app by April 28, 2017
  • Launch new scheduling system by June 30, 2017
  • Convert Setup to mobile style by July 31, 2017
  • Increase setup conversion to 90%

Objective: Ensure 12 clients up for renewal don’t churn
Key Results:

  • 100% clients have roadmap discussed with point of contact by end of quarter
  • 100% clients had baseline implementation completed by end of quarter
  • Add 1 advocate at each of the clients
  • Write-up the ‘The implementation process scenarios’ section for wiki - Mar 28th
  • Get understanding of categories for usage metrics indicating renewal per segment

How’d we do?

No idea. Goals are useless if you don’t keep track of your progress. Which we didn’t.

So, why didn’t we keep up with them?

Too many objectives – Each department had way too many objectives. The goal “Ensure 12 clients up for renewal don’t churn objective” came from a document eight pages long! For one department! There are six more objectives, and each one has a massive list of key results (that example isn’t even the longest). Who can remember that many objectives or key results?

We felt discouraged by the results – We thought our key results were reasonable, but we didn’t come anywhere close to achieving them. We knew we weren’t going to come anywhere close to what we set out, so we ignored them. Then forgot about them.

How to set good OKRs

How do we make OKRs work this time around? Based on the issues we encountered the last two times, here’s how I’m going to teach OKRs now:

Choosing the right objective

To find good objectives, ask yourself “What are my top three concerns?”

They need to be grounded in reality. For instance, as a startup, if one of my concerns is a lack of revenue growth, it’s easy to come up with “Reach $1M in Annual Recurring Revenue”. However, that’s likely a recipe for failure and disappointment at an early stage company. A more appropriate one would be “Determine the viability of the sales model.”

Choosing good key results

It helps to think of objectives as experiments. Key results are how you measure the success of the experiment.

Key results lie on a spectrum between in your control and out of your control. It’s best to have a mix of both.

For “Determine the viability of the sales model,” here are some example key results:

  1. Call 50 customers per sales rep per day – Did we actually do any work on this objective? In our control.
  2. Conduct 100 customer pricing surveys – Did we get enough information to declare this experiment a success? Partially in our control.
  3. Achieve an average survey result of $10K – Is the distribution model viable? Out of our control.

This is a good mix of key results because it tells us a lot about the success or failure of the objective. Specifically, if it was a failure, we will know if it was our fault (not actually working on it) or if it was outside of our control (people don’t want to pay at least $10K).

Focus

Finally, you only have so much time and mental capacity. Choose only three objectives, and track only three to five key results per objective.

Next Steps

Setting goals doesn’t have to be hard or discouraging. Adopt an experimental mindset. Ground yourself in reality. Measure what you can control and what you can’t.