Organizational alignment is a challenge that all companies go through. Sometimes we're running so fast that we forget how important it is to create alignment and that it's one of our core responsibilities as engineering leaders.

Elon Musk has a great analogy for this. Imagine that people's impact is a vector. Vectors pointing in different directions mean wasted energy and momentum. As leaders, we need to guide people in the same direction.

On the right side of the image above, we can see the difference in the impact of moving the company in a specific, aligned direction. Comparing the left and right sides of the image, we can see the wasted energy due to the effects of misalignment. Now imagine the waste if, instead of four people, we had 200. Let's state the obvious: it will be impossible to get perfect alignment, and we shouldn't even try to achieve it. Still, we must be aware that we are ultimately accountable for optimizing this process as leaders.

Alignment is built through clear direction, goals, and a proper understanding of that direction. Communication is the key driver of this. When we talk about goals, we need to put engineering and product goals at the same level, as these two orgs define the product's future and the process of getting the product out.

The alignment process should always come from a high-level intention definition, as with OKRs, so it cascades down. Cascading down can come in a push motion (where we push goals to the teams) or in a pull motion (teams identify how they will contribute to high-level goals.).

No company should work only in push or pull mode. We should be using both motions together. On top of that, we should also leave room for team-level optimizations. We want to promote team autonomy and accountability from the pull motion and local optimizations. However, as leaders, we need to create some momentum for critical changes - this comes from a push motion.

As shown in the image below, we want to start with goals at the top and cascade them down through the organization, team by team. A big part of this process implies ensuring all managers are aligned on the goals for the upcoming cycle. This process helps uncover if everyone is truly aligned on what matters and if communication is flowing up and down, which is critical when we are growing fast.

As leaders, we must guarantee that all teams contribute towards a specific goal. To enable this, we define a top-level objective and clarify our expectations to the team. Teams will then identify how they will work towards the goal.

We mustn't be dogmatic regarding goals. In the example above, when we state that we want to have a cycle time of less than 7 hours, we can observe that the Finance team doesn't need to do anything. The team can either ignore the objective or add it to their team goals. They may add it to their goals to have constant visibility through the goals review period. This will allow teams to course-correct if something is not going well. We recommend that teams always add organizational goals to their team goals in this push motion.

In a pull motion, we set high-level goals for the planning cycle, but we only share the intent and not specific details with the teams. Teams will then tell us if and how they will contribute to those objectives. If we see that we don't have enough quorum to move a goal forward, we should stop and think about whether:

If we are in the 1st situation, we should kill the goal, look for an alternative, and restart the process. Or we can drop the goal and come back to it later down the road.

If we are in the 2nd situation, we should communicate with the whole organization on the objective's intent and get together with the managers. We must talk with the managers to explain why that goal is important and where the communication failed. Managers need to support their teams on goal-setting decisions, which means they should be able to explain the goals without our intervention (this is critical for fast-growing companies).

Remember that alignment needs to happen in recurrent cycles and that we need to define our level of investment between delivering products and improving engineering. A good Acid test is to have between 3 to 5 top-level engineering improvement metrics.

If you have any questions or you want to get more strategies on how to improve, drop us a line, and our Engineering Success team will always be available to share their experience with your team.