So you did your research on OKRs:
- You reviewed the Google video on OKRs.
- You read a bunch of blogs on OKRs.
- You know Intel, LinkedIn, Twitter, and dozens of hot start-ups in the Silicon Valley use the OKRs methodology to define, align, and make progress on their most important goals.
- You saw The Executioner’s Tale
- Maybe you even joined the Objectives and Key Results Group on LinkedIn.
It all sounds great, so you decide to launch OKRs at your company, well, at least for your team. After the initial excitement, you realize you’re still not clear how to get started. So, you review the 30 or so questions and answers posted on OKRs in Quora.
This blog summarizes lessons learned based on several real-world interviews with business leaders currently rolling out OKRs at their organizations.
Note to busy readers: I’ve included a few key points in BOLD RED in case you want to skim this over!
One manager at a mid-sized company, still in their first quarter of using OKRs, summarizes his experience with this approach to performance management:
“Our CEO said ‘customer retention is our top priority.’ Traditionally, customer retention was our customer success team’s problem; they were the ones that managed ongoing client interactions and renewals. At first, everyone assumed customer success would just work harder to drive customer retention, and other departments continued to focus on their same priorities. Then every team leader started asking: what can my team do to increase customer retention? The discipline inherent in OKRs got us focused so that every single area is working together to drive customer retention.”
He went on to provide specific examples about how OKRs quickly created a culture of alignment across the company:
Product Team “Our feature requests used to focus on what we thought new customers would want or what would be cool to differentiate from the competition, but now our product team asks the same question before approving a new feature request: how does this product improvement drive customer retention?”
Marketing Team “We shifted our traditional event-marketing approach of sponsoring channel partner events to launching our first annual user conference. We took the time at the user conference to interview customers and gather valuable survey data.”
Sales Team “Even our sales team is now taking time to call on our install base and ask open questions around how we can add more value. They do this simply to build the relationship and emphasize the importance of working together over the long haul. Sales reps now have an explicit key result “Meet onsite with senior management to conduct 2-hour product feedback session with all Tier 1 customers 60-120 days prior to their contract renewal date”
3 Benefits of OKRs:
- “Focus is a constant theme in every team meeting. We begin with OKRs and limit the number of objectives and key results to reduce our conversations to make measurable progress on our most important goals.”
- “Teamwork is way better! Everyone on the team and even on other teams now work toward the same goals (e.g. customer retention), which is encouraging for everyone. We celebrate other’s results not just accomplishments within our own department.”
- “Learning each quarter does not work unless you are focused and make the time to explicitly reflect on what you’re learning. In order to be better at what you’re doing, you need to look back. The work of defining and reviewing OKRs creates a continuous learning environment.”
Hold on here… this blog sounds like yet another sales pitch for OKRs! Let’s get to the good stuff: practical tips and pitfalls to avoid when rolling out OKRs based on those who have gone before you.
4 OKRs Process Tips from the Field:
- Cadence: While it’s good to adjust due dates of OKRs throughout the quarter and review progress weekly, we find it often works best to set a bi-weekly OKR in the form: “By the end of next sprint 2 weeks, the 3 lead engineers will accomplish: (insert a single engineering key result here).”
- Ownership: In general key results have an individual owner, but in this case, putting in the time to split key result into child key results and assigning each to one of the 3 engineers would have taken more time to make it fair for everyone. It’s often better to let team members work as a team and succeed or fail together when OKRs have a short timeframes.
- Adaptable: When we set quarterly OKRs, we were very specific about the things that matter to us most (e.g. customer retention). We started the quarter thinking we need 4 projects in place to drive customer retention. We completed the first 2, but we never did the last 2. In fact, we ended up finding more important projects to drive customer retention. It’s important to constantly be ready to shift focus throughout the quarter and proactively question why you’re doing what you’re doing. I can sense a cultural shift where everyone feels encouraged to question their priorities and make sure we’re working on the right stuff at the right time.
- Team Meetings: We begin all team meetings by showing OKRs. This is a critical requirement for focus. Starting a team meeting with anything other than OKRs means we’re losing focus. After reviewing the OKRs, we talk about where we’re ahead, where we short, and areas we need to improve. We reflect on objectives and ask ourselves was the original objective realistic? Did we target the moon and sink in a hole or make it to the top of a mountain? Each team meeting ends differently, but always starts with OKRs and drives toward next steps.
Rolling out OKRs to create a metrics-driven culture focused on results sounds great, but it can be quite a cultural change. Let’s summarize some common pitfalls.
Best Practices for Avoiding 3 Common Pitfalls with OKRs:
Pitfall 1: “Many on our team do not update progress on OKRs each week” This is a problem for many people (Note: 30-50% do not regularly update progress regularly based on my interviews so far). They may update progress on OKRs that are 100% achieved or ahead of schedule, but they are not rigorous about providing status updates on items where they falling behind. In certain areas, like product development, this is not acceptable since 50% complete with a feature in production does us no good at all. Some key results are scored 0 or 1 since there’s no value for the middle ground.
- Current Best Practice: Score and grade OKRs each week rather than waiting until the end of the quarter. Managers cannot possibly score every goal on their team. Instead, individuals should self-score their OKRs each week. A zero in one area is fine as long as they can indicate a good score in another area!
- Future Best Practice: Automated or semi-automated, continuous scoring of OKRs.
Pitfall 2: “Weekly OKRs start to look like tasks, not goals.” On the one hand, we have OKRs where you are trying to achieve a specific, measurable impact by a certain date. On other end we have simple tasks like: “email the client” or “meet with the VP”, where no clear result is expected. Quarterly OKRs are clearly not tasks and often take the classic form: “Achieve x% increase of key business metric Y.” But we often need to breakdown high-level KRs into more detailed KRs as well as reduce the timeline from quarterly to monthly or even weekly. These detailed KRs take the form “complete x by date y” and tend to look more like tasks than goals. But the whole point of OKRs is to get us focused on goals, not tasks! And yet, we still need to take action and monitor progress on OKRs frequently. For example, a 60% improvement by end of quarter translates to a 20% improvement each month which means we should see about 5% improvement each week.
- Best Practice: Consider the distinction between task and goal as a continuum rather than an either-or. As long as quarterly key results are clearly defined as goals, not tasks, you are OK. Breaking down a quarterly OKR into a “weekly OKR” is an excellent idea. In fact, this will ensure that you focus on tasks that impact your most important goals.
Pitfall 3:”Failing to jointly define OKRs across departmental functions” An SVP of Engineering told me his team’s objective was to “Enable Sales to Double.” He listed a key result “Contribute to 10 pre-sales technical discussions in Q3 with senior IT leader resulting in a confirmation that our solution meets all client security requirements.” He had one more KR related to developing training content. Then, I asked the big question: “What else needs to happen in order for your team to achieve the objective ‘enable sales to double?” The SVP said that’s all. So, I gave the homework assignment to run these 2 KRs by the VP of Sales. However, the SVP of Engineering sent the 2 KRs to the SVP of Sales via an email rather than taking the time to sit down and agree on the set of KRs with the right wording to ensure horizontal alignment. For more on this topic, please see Why Integrated Driver-Based Planning is Key to Your Business.