My time at SafetyCulture - Goals and Objectives

Now that I've stepped down from my role as CTO of SafetyCulture I’ve fielded many questions about what it was like, and how we did things. So I'm taking some time to reflect on what I think went well and what I would do better if I were to do it all again.

One topic that stands out in my mind is the setting of goals and objectives. And how we go about doing that. Historically, I have thought of goals as being important for two reasons. They give us a direction to work towards and they allow us to determine our progress along the way. But I found at SafetyCulture there was a third reason to put effort into defining them.?

Alignment?

Goals can be very good at helping align what people do. Or at least align what they care about. If the goals of everyone in the team are consistent and aligned, you will usually have everyone in the team pulling together in the same direction. However, if there is even a slight inconsistency, or different priorities amongst people's goals, you create tension in the team. I saw this several times, and it was never intentional. It happens by accident. Whenever there was something off about a team dynamic, or where people don’t agree on priorities, you would usually find their goals didn’t line up. Or at least their interpretation of the goals didn’t.?

We achieved a sudden boost in alignment when we introduced goals and objectives into the engineering organisation. We saw that as an immediate benefit. With my time again however, I would make sure that within a team people’s individual goals lined up closely, and actively look for where there might be misalignments.

Oppositional goals?

One frequent challenge we had was encountering goals that appeared to be in opposition to each other.The most classic case was speed versus quality. You can have it fast or you can have it good. Setting aside the truth or otherwise of this cliche, it's true enough in many people’s minds that it feels like oppositional tension. Sometimes when two goals are working against each other (in this case, speed versus quality) it's easy to spot that they're in opposition. However, there are plenty of other goals that are in opposition but it's just not as obvious at first. For example, asking someone to become the “go-to person” for something. Become a visible and respected expert in a particular topic. But then we also ask them to adopt new skills and in particular new technologies, and so enter into the mindset of a beginner. For some people, maybe most, asking them to rapidly cycle between expert and beginner frames of mind can be taxing.You might consider these goals as opposing each other, albeit indirectly. Giving someone ambitious delivery goals while asking them to mentor others can feel oppositional unless they have (or can develop) great time management skills.?

When someone feels like they have oppositional goals, it creates personal stress. If I had my time again, I would put more effort into identifying and resolving instances of oppositional goals to avoid creating this stress. I suspect you cannot remove them entirely, but can at least minimise them, and provide extra support to people who for whatever reason have to carry goals that are in opposition.

Linkages

The next challenge was to make sure that for any goal, you could see some direct link to another (usually higher) goal. If people cannot see the linkage between their individual goals and those of the organisation, they will find the goals irrelevant at best or incredibly demotivating at worst. Neither outcome is good. And it’s not enough to establish a linkage, it has to be visible and easy to identify. We achieved this to some degree by using the idea of cascading goals. But for this to work well, you have to be quite explicit that this is what you are doing. It can help if you as the CTO to establish your goals early and share them. We were quite good at cascading goals in that first layer of management, but we could have done more to make the cascade process “real” for frontline managers.?

Where we should have done more is making the linkages visible. It would have been better if a front line engineer could trace their individual goals back to the goals of their manager, and then back to their manager etc… all the way to the company goals. We could mostly talk this through with people, but we could have built a mechanism that allowed people to “self serve” this exploration of the linkages between goals.?

Outcomes or activities?

The next challenge was whether to use outcome metrics or activity metrics. Both seem valid to me, but it also seems to be popular for some managers to flex and assert only outcomes matter. I don't agree with this philosophy. I think outcome metrics are a bit like lagging indicators. You can see what has been done/achieved. And it is motivating to have something “concrete” to aspire to. However, I think activity metrics are like leading indicators. They tell you what's likely to happen in the future (although of course there is no guarantee). From my time in sales we would have a quota as an outcome metric, and we would also have activity metrics like the number of customers called and the number of customer meetings we had. This provided a more rounded picture. It also allowed us to differentiate between people who failed to reach a quota (outcome) because of a lack of activity versus those who didn’t reach the outcome for some other (usually more complex) reason.?

Upon reflection, I would like to have spent more time differentiating between outcome and activity metrics, while still leaning towards outcomes. We might develop a more nuanced view of people’s performance if we thoughtfully included the right activity metrics. For example, where we have shared goals, we could use an activity metric to describe someone’s contribution to the shared goal. This would allow us to get the alignment benefits of a shared goal, while retaining some sense of individual responsibility and ownership towards the shared goal.

Shared or individual goals?

Should we have shared goals or should our goals only apply to what we can control individually? This is a conversation that always brings out strong opinions. We used a combination of the two but leaned towards individual goals. This had the benefit that people felt like they were in control of their own destiny (i.e. performance rating). Where we had shared goals, ownership and responsibility got a little cloudy. But shared goals are great for establishing alignment.

If I were to do it all again, I would still use a combination of shared and individual goals - I think the combination is powerful - but I would spend much more time making sure the shared goals have strong visible linkages in all directions. So people can see how their efforts contribute to a shared goal.

Drift

Ask everyone in a team meeting to (privately) write down what they think the three goals of the team are on a piece of paper. When you collect and compare them, you'll may be surprised at how divergent the answers are. As a team leader, you often fall into the trap of believing everyone is clear on the goals, just because you're clear on the goals. This is alignment. There is always some divergence, but what I’ve noticed is that this divergence grows with time. And time is a matter of weeks, not months. People’s understanding of the team goals will drift apart over time. The more time you leave between revising and reviewing the goals, the more drift. You drift out of alignment in a matter of weeks if you don’t reconnect or recalibrate to your original goals.

We made quarterly reviews of goals the preferred practice - but in hindsight, that’s rarely enough. Some of the better frontline managers would check in with the goals every month. They seemed to encounter the least amount of drift (and, their teams routinely scored best in sentiment surveys for sense of direction and purpose). In retrospect, we would have benefited from standardising on monthly reviews as common practice to better manage drift.

Packaging

So how did we package this up? We chose as simple a mechanism as we could imagine. Which was really just a set of spreadsheets. We wanted the focus to be on the content, not on the tool (Imagine asking a bunch of engineers what tool you should use - good luck managing 200 different opinions from only 150 engineers). For the goal of simplicity, I think we were successful. We didn't spend time or energy debating whether to use tool A or tool B. So, spreadsheets fulfilled their intended purpose to capture goals in written form, providing a low tech, simple way of doing these things.

Once the team grew past the 150 mark however, we saw growing pains with this approach. At 150 it was messy but manageable. I don't imagine this would scale far beyond 200. And it's hard to display linkages between goals. So what tool should we use? After scouring the market I could not come up with anything that was impressive. They are either too sophisticated and as a result create extra workloads and distractions. Or they are too inflexible and cannot accommodate how you want to do things in your shop. My search for the right tool continues.

Conclusion

Overall, I think we managed goals and objectives reasonably well at SafetyCulture, but I can certainly see ways we could have done things better. Alignment, shared v individual goals, outcome vs activity metrics, linkages and packaging. All things to improve upon. But that's hindsight for you. The issues outlined above are much more clear to me now than they were at the time. I was fortunate that my engineering leadership team, and our engineers more broadly, were full of incredibly talented people who could navigate us through these issues as they arose. And there’s a lesson there as well. If you hire great people you don’t have to be perfect at setting goals and objectives - they will help you through it. You live and you learn.