 Hi everyone. Thank you for watching my talk. My name is Roshan Gupta and today I'm going to talk about OKRs and specifically three tips to get the most out of OKRs. But first, let me introduce myself. So I've been building products for over 22 years. The first date as a software engineer and the last 14 as a product manager. I've worked in companies of all sizes, from tiny startups to companies that IPO'd to big companies like Google, Facebook and Yahoo. I've worked across a lot of different verticals like e-commerce ads, consumer software, enterprise software, your name a few. And currently I help lead the Android messages team at Google. So in all these different environments, I've seen and experienced what works and what doesn't. And today I hope to share some of that knowledge with you. So let's talk about OKRs. You've probably heard this term before. O stands for objective, KR stands for key result. And together, OKRs are a method for teams to set quarterly goals. It's a way for businesses to plan the next quarter. OKRs are popular with companies like Google, Facebook, Amazon and more. And as a product manager, you'll probably work for a company that uses OKRs or perhaps you already did. So here's the thing. I've been developing products for 22 years. And I can tell you, there's the theory of how OKRs should work. And then there's the practice. And there can be quite different. So let's start with the theory of OKRs. So in the traditional OKR process, you first have a kickoff, usually weeks in advance of the quarter beginning. Leadership and the product teams get together, they collaborate, they have discussions, they have debates, they incorporate feedback, they have plenty of time to align on what the quarterly goal should be. Alignment is done, there's high fives, hugs, maybe some tears of joy. And now they have this beautiful plan and it's time to execute for the quarter. And so they use this plan and they check in on it frequently throughout the quarter. Maybe they'll course correct based on the plan. You know, it comes down to planning the work and working the plan. It all goes extremely well. And as the next quarter approaches, the team is excited to do it again. And so that's how OKRs in theory should work. However, let's talk about the practice of OKRs, because in reality, it goes more like this at companies. First, the quarter is almost there, maybe a week before, if you're lucky. Everyone realizes you have to define OKRs. And so you have that wake kickoff. Oftentimes the teams will kind of grumble, groan, they'll hesitate. They'll wonder why they're doing all this work. It's never useful. But hey, you have to plan your business for the quarter. And so the teams scramble against really tight deadlines to come up with OKRs. And you'll find that teams come up with wildly inconsistent OKRs. Some teams have a handful of OKRs and some teams have a whole bunch of OKRs. Oftentimes the OKRs that they come up with just roll over quarter to quarter. You'll then have a review of the OKRs. And more often than not, it's extremely rushed. There's time to talk about one or two of the OKRs. You try to talk about all of them very quickly. Teams often feel pressure to take on more and more goals for the quarter. But let's say they get past all this, the tight deadlines, the review process, and now they actually have to use the OKRs. Typically what happens is, a few weeks in, everyone forgets about the OKRs. If you're lucky, sometimes an organization will do check-ins mid-quarter on those OKRs. But again, they're rushed OKR reviews. And then towards the end of the quarter, you have the grading. And the grading is almost always contentious. There's heated debates. Teams feel like the grades reflect on their performance, their efforts. There's an inflation of grades as teams try to be more optimistic with where they were at. But after all this heated debate, they have the grades. But more often than not, leadership doesn't look at the grades. And why? It's because typically it's the end of the quarter, they realize they have to plan next quarter's OKRs and the whole process repeats itself. Every time you go through this process, though, the grumbles, the groans, and the hesitation grows stronger. So do parts of this process sound familiar or maybe all of the process? If so, I'm sorry to say you are suffering from a case of the OKRs. What started with good intentions, create a plan, ends up being a best overhead for the team and a worse source of tension and delay. So should we give up on OKRs? Teams have to plan their business. And at the end of the day, OKRs are just a tool and tools are only useful if they're used in the right way. So today I'm going to share three tips with you on how to get the most out of OKRs. We're going to talk about optimizing for predictability. I'll introduce you to the concept of tactics. And finally, we'll talk about using top 10 lists to run efficient reviews. So let's dig in. So imagine there are two teams. The first team shows you how many OKRs that they get done in this table. And you can see team two gets a larger percentage of their OKRs completed every quarter. And remember, you as a product manager have to plan the budget, assign resources, update customers and partners on timing. You have to time launches. You have to line up PR and press and marketing and legal reviews. So which of these two teams do you think would help you run your business effectively? Most of you will probably say team two. They plan the work and then they work the plan and you can count on those results. So now what if I told you that team one took on double the amount of goals? Does that change your mind? My guess is it doesn't change your mind. So that is the strategy. We want to use OKRs to emphasize predictability and optimize for predictability, not for accountability. A lot of times OKRs are used to make sure teams are working hard or as a way to ensure that they're performing well. But you got to trust that your team is working hard and the goal is about predictability. During OKR reviews, you want to use these OKRs and ask questions like why did we not get it done when we thought we could get it done? How can we be more predictable next quarter? So now let's talk about the benefits of this strategy. So remember when we talked about the practice of OKRs, our teams felt pressure to take on more OKRs than they could handle. So if you build a culture that values consistency of delivery over volume, you'll help eliminate that pressure. Another big benefit is grading becomes much less contentious. Teams truly believe that they're being graded on the quality of their prediction versus the quality of their efforts. And finally, let me give you some advice as you try to implement the strategy. The first is be a broken record. You need to emphasize that OKRs are about predictability and not accountability. Do it during kickoff, do it during OKR reviews, do it during grading. But the team truly has to believe they're being evaluated on the quality of their prediction and not their efforts. The second thing is find ways to reward predictability. It's less about how much they're getting done and more about achieving consistent results quarter after quarter. And then the last tip I have is I often get asked the question, hey, my teams can be predictable if we do very little work. And so the goal is not to do as little work as possible in the name of consistency. But my advice is focus on consistency first. Get predictability and then you can figure out how to increase throughput and achieve that right balance. So let's talk about my next tip, tactics. So let's look at these OKRs. How would you grade them? If you look at them, only one out of three key results were completed. But the objective was completed. So was the team successful, were they half successful? It's a bit unclear. The other thing that's odd is if you look at the objective, is that improving a metric for the sake of improving a metric? Is that the objective? Or is that a key result in service of a bigger objective? So we can solve all this by introducing the concept of tactics. Tactics separate the why and the what from the how. It makes each part of the OKR more clear and of the right granularity. And so if we rewrote the OKRs that we have on the left, perhaps the bigger objective, the multi-quarter objective, is to become the market leader in Europe. And now the key result, the milestone for this quarter in the service of that objective is to improve a specific KPI by 20%. How long proof are the tactics? And the key here is the tactics can change. The team is committing to or predicting that they can achieve that key result this quarter. But how they achieve it is up to them. So if I have to summarize this strategy, introduce the concept of tactics, which is used to capture the how. But the most important thing is don't grade the tactics. Tactics can change and focus instead on the objectives and the key results. And what are the benefits that you'll see if you implement this strategy? First, you'll have fewer OKRs. You won't have those long lists of OKRs anymore because you're separating out the how from the what. Your key results become more streamlined. Second, you'll have more apples-to-apples comparisons between teams because objectives and key results and tactics are all of the right granularity. And finally, introducing tactics creates focus. During OKR reviews, leadership in the teams can focus on the value, the why and the what. And the teams themselves are empowered to figure out the how, how they will achieve that value. And finally, my advice for implementing this strategy, don't put a limit on the number of tactics that can be added. Often in OKRs, team members want to feel represented. They want to feel like their projects are, you know, being discussed. And so, allow team members to add as many tactics as they want to represent their work. The key, though, is during the reviews themselves, focus on the why and the what. One way to do this, if you track OKRs in the spreadsheet, for example, you can create rows of tactics and group them and roll them up under a key result so you can expand and collapse them as you need. So that's tactics. Let's talk about OKR reviews. So, OKR reviews are often rushed or sometimes, frankly, skipped because they've proven to be ineffective. And there are a couple of reasons why this happens. The first reason is teams fear reporting bad results. They feel like a bad score or a bad grade is a reflection on their performance and their efforts. And we talked about how to solve this. It's about the quality of their prediction, not the quality of their efforts. Another reason OKR reviews aren't efficient is that oftentimes, those OKRs are inconsistent. There are too many OKRs or too few OKRs. And we talked about introducing tactics as a way to solve that problem. And then there's a third reason why OKR reviews aren't efficient. And that's the time crunch. And to solve the time crunch, I want to introduce you to the concept of top 10 OKR lists. So here's how it works. Each team should nominate 10 OKRs that they will report on on a consistent basis. And every time you have an OKR review, you focus on these top 10 first. Time permitting, you can go over the other OKRs, but you always cover those top 10. To make this even more powerful, every layer of the org should pick its own top 10. So you may have top 10 among a couple of teams. But when you go to the next level and need to report up in your organization, you create a top 10 from those top 10. So that's the strategy. What are the benefits of doing this? So the most obvious benefit is you'll have a consistent set of OKRs that are tracked every single review. It also makes preparing for the reviews easier for teams because they can prioritize these top 10. You'll find that reviews will be easier to prep for, they'll finish on time, and they'll yield useful discussions. And finally, my advice on implementing this strategy, if top 10 is too much, go with top five. Go with top seven. The goal is to review a consistent set of OKRs every single time. Another thing to help make sure those reviews are efficient, when you look at those top 10, start with the most risky OKRs and go to the least risky OKRs. Again, if you do this strategy right, you'll have time to cover all of them, but it's just another tweak you can make to the process. And finally, even though you're only reviewing 10 of the OKRs, you have to trust that the team is working on the OKRs that aren't being talked about. Remember, this is not about accountability. You trust that the teams are working hard. This is about having streamlined OKR reviews that are useful. So to sum up, OKRs can be a useful tool. Product teams need to plan their business, but there's the theory of OKRs, and then there's the practice of OKRs. Today I covered three tips on how to get the most out of OKRs, optimized for predictability, not accountability, introduce the concept of tactics, and use top 10 lists to run efficient reviews. On a final note, just remember, if you wield the OKR tool correctly, you just might save your team from a case of the dreaded OKRs and turn them into believers. Thank you for attending my talk.