In commercial aviation, most people know there's a "Pilot" and a "Co-Pilot" up front (also known as the "Captain" and "First Officer"). The ranks of Captain and First Officer denote seniority but both pilots, in practice, take turns filling one of two roles: "pilot flying" and "pilot monitoring". The pilot flying is at the controls actually operating the plane. The pilot monitoring is usually the one talking to Air Traffic Control, running checklists, and watching over the various aircraft systems to make sure they're healthy throughout the flight.

Obviously, both pilots share the goal of getting the passengers to the destination safely and on time while managing the costs of fuel etc. secondarily. They have the same engines, fuel tank and aircraft systems and no way to alter that once airborne. They succeed or fail together.

It would be a hilariously bad idea to give one pilot the goal of getting to the destination as quickly as possible while giving the other pilot the goal of limiting fuel use as much as possible and making sure the wings stay on. Or, to take an even stupider example, give one pilot the goal of landing in Los Angeles and the other the goal of landing in San Francisco. Obviously that wouldn't work at because those goals are in opposition to each other. You can get there fast if you don't care about fuel or the long-term health of the aircraft, or you can optimize for fuel use and aircraft stress if you don't need to worry about the travel time as much. It sounds ridiculous! And yet, this is how a lot of organizations structure EM and PM goals.

How the EM/PM relationship is similar to Pilots

An EM and PM have to use the same pool of available capacity to achieve their objectives. They are assigned a team of developers and what that team of developers is able to do in a given period is a zero-sum-game. In a modern structure, following the kinds of practices that lead to good DORA metrics, their ownership is probably going to be a mix of building new features and care and feeding of what they've shipped before.

Like a plane's finite fuel tank, the capacity of the team is fixed and exhaustible. Boiled down to its essence, the job of leaders is managing that capacity to produce the best outcomes for the business. Problems arise, however, in determining what those outcomes are because they're going to be filtered through the lens of each individual's incentive structure. Once fuel is spent, it’s spent. Add more destinations without adjusting the plan, and you’re guaranteeing failure.

Accountability and Incentives

It would not be controversial to say that a typical Product Manager is accountable and incentivized to define and deliver new features that increase user adoption and drive revenue growth. Nor would it be controversial to say that a typical Engineering Manager is accountable and incentivized to make sure that the code that is shipped is relatively error-free, doesn't crash, treats customer data appropriately, scales to meet demand, etc etc. You know the drill. I think this is fine and reflects the reality that EMs and PMs aren't like pilots in that their roles are not fully interchangeable and there is some specialization in those areas they're accountable for. But if you just stop there, you have a problem. You've created the scenario where one pilot is trying to get to the destination as quickly as possible and one is trying to save fuel and keep the airplane from falling apart. You need to go a step further and coalesce all of those things into shared goals for the team.

An interview question I like to ask both prospective EM and PM candidates is "How do you balance the need to ship new stuff with the need to take care of the stuff you've already shipped?" The most common answer is something like "We reserve X% of our story points for 'Engineering Work' and the rest is for 'Product Work'. This way of doing things is a coping strategy disguised as a solution. It's the wrong framing entirely because everything is product work. Performance is a feature, reliability is a feature, scalability is a feature. A product that doesn't work isn't a product, and so a product manager needs to care about that sort of thing too. Pretending there's a clean line between "Product Work" and "Engineering Work" is how teams can quietly drift off-course.

On the flip-side, I'm not letting engineering managers off the hook either. New features and product improvements are usually key to maintaining business growth and keeping existing customers happy. All the scalability and performance in the world don't matter if you don't have users. Over-rotating on trying to get to "five nines" of uptime when you're not making any revenue is a waste of your precious capacity that could be spent on things that grow the business and that growth will bring more opportunities for solving interesting technical challenges and more personal growth opportunities for everyone who is there. EMs shouldn't ignore the health of the business any more than PMs should ignore the health of the systems their teams own. Different hats are fine, different destinations are not.

Goal Frameworks and Incentives

If you use OKRs or some other cascading system of goals where the C-Suite sets some broad goals for the company and those cascade into ever more specific versions as you walk down the org chart hierarchy, what happens when you get to the team level? Do the EM and PM have separate OKRs they're trying to achieve? Put them side by-side and ask yourself the following questions:

• If we had infinite capacity, are these all achievable? Or are some mutually exclusive?
  • Example: Launch a new feature with expensive storage requirements (PM) vs. Cut infra spend in half (EM)
• Does achieving any of these objectives make another one harder?
  • Example: Increase the number of shipped experiments and MVPs (PM) vs. Cut the inbound rate of customer support tickets escalated to the team. (EM)
• Do we actually have the capacity to do all of this?
  • Example: Ship a major feature in time for the annual customer conference (PM) vs. Upgrade our key framework which is 2 major versions behind and is going out of support. (EM)

If you answer 'no' to any of those, congratulations, you've just planned your team's failure to achieve their goals. To put it another way your planning is not finished yet! You need to keep going and narrow down the objectives to something that both leaders can get behind and do their best to commit to.

How to get aligned

1.	Put the EM and PM goals side by side.
2.	Identify conflicts and classify them (mutually exclusive, tension, compatible).
3.	Talk about business value without the full team in the room.
4.	Use “what’s best for the business” as a tie-breaker.
5.	Bring in a neutral facilitator if you’re stuck.
6.	Commit together. No outsourcing the decision.

You've already got your goals side-by-side. After you've identified the ones that are in conflict, have a conversation about business value. Don't bring in the whole team yet, they're engineers and the PM might feel outnumbered. Just have the conversation between the two of you and see if you can come to some kind of consensus about which of the items in conflict are more valuable. Prioritize that one and deprioritize the one that it's in oppostition to.

Use "what's best for the business" as your tie-breaker. That's not always going to be perfectly quantifiable so there's a lot of subjectivity and bias that's going to enter into it. This is where alignment up the org chart is very beneficial. These concepts don't just apply at the team level.

If you find you can't agree and aren't making progress, try to bring in a neutral third party to facilitate the discussion. If you have scrum-masters, agile coaches, program managers or the like they can often fill this role nicely. You could also consult a partner from Customer Support who will have a strong incentive to advocate for the customer above all to add more context to the discussion. Don't outsource the decision though, both the EM and PM need to ultimately agree (even reluctantly) on what's right for the team.

This has to go all the way to the top

If your org, like many, has parallel product and engineering hierarchies, alignment around goals at EACH level is critical. The VP Eng and VP Product should aim to have shared goals for the entire team. Same thing at the Director level, and every other level. That way if each team succeeds, each leader up the org chart succeeds, and ultimately the customer and the business are the beneficiaries. If you don't have that, doing it at the team level alone is going to approach impossible and you should send this post to your bosses. 😉 But in seriousness, you should spend some energy trying to get alignment up and down the org chart. If your VP Eng and VP Product don’t agree on direction, your team is flying two different flight plans. No amount of team-level cleverness can fully fix that. Your job is to surface it, not absorb it.

Things you can do:

• If your original set of goals conflicted, the goals at the next level up probably do to. Call that out and ask for it to be reconciled.
• Escalate when structural problems get in the way of the team achieving thier goals. Only the most dysfunctional organizations would create systems that guarantee failure on purpose.

Planning and alignment are continuous activities

You likely set goals on a schedule. Every n months you're expected to produce a new set and report on the outcomes from the last set. That's all well and good but I think most people who've done it a few times recognize that the act of doing it is more valuable than the artifacts the exercise produces are. (Often described as "Plans are useless, planning is critical.")

The world changes constantly, the knowledge you and your team have about what you're doing increases constantly, all of these things can impact your EM/PM alignment so it's also important to stay close, keep the lines of communication very open, and make adjustments whenever needed. The greatest predictor of a team's success, health, and happiness is the quality of the relationship between the EM and PM. Keeping that relationship healthy and fixing it if it starts to break down will keep the team on the right flight path with plenty of spare fuel for surprise diversions if they're needed.

Land that plane

Regardless of what framework you're using, or even if you're not using one at all, the team should have one set of goals they're working toward and the EM and PM should succeed or fail together on achieving those goals. Shared goals don’t guarantee a smooth landing. Misaligned ones guarantee a crash.

"A380 Cockpit" by Naddsy is licensed under CC BY 2.0 .