Tag Archives: management

Building Cross-Functional Teams

Frequently people ask me how our growth team is structured at Pinterest. In our case, it is a cross-functional team. Engineers, product managers, analysts, and designers all work together on shared goals. Pinterest believes the best results arise when people from different backgrounds work together on a problem. I’ve thought a lot about how to develop effective cross-functional teams in an organization, and I’d like to show how to do that successfully. These, in some ways represent how Pinterest is structured, and in other ways, don’t.

Step 1: Define Metrics
In order for a cross-functional team to be successful, it needs a North Star metric. If you’re creating one broad team, it’s a broad metric, like MAUs or revenue. I prefer creating multiple, smaller cross-functional teams that carve a piece out of the main goal, like signups or new user revenue. Then, you can create another small team for something like retained users, or repeat user revenue.

Step 2: Build the Team
A core team for cross-functional team trying to impact a metric is usually:
Product Manager
Potential other members:
QA person

Depending on the goal, you may not need many of these, or the product manager can be the catch all for analysis, research, etc. One person should be the owner of the team. Usually, this would be the person with the most context. At Pinterest, it’s either an engineer or a product manager. Ownership should seem arbitrary as the team should organically align on initiatives over time, deciding between short and long term projects, based on a shared understanding of what is likely to move the key metrics. So, ownership is more so management has one person to go to with questions than anything related to authority.

Step 3: Re-train Managers
There shouldn’t be any managers on cross-functional teams. Managers are the glue between different cross-functional teams, making sure all the teams align to the global strategy, and don’t improve their metrics at the cost of another team’s metrics, which is easy to do. Here is what some managers roles will look like in this scenario:
Design Director: align visual style across the entire application
Director of Analytics: align use of tools across teams for easy translation of data back and forth and sharing of pertinent data across teams
Marketing Director: allocate budget effectively, communicate how teams’s activities affect each other and balance
Director of Product: ensure product opportunities on one team can be leveraged by other teams, prevent disjointed product experience

Benefits of Cross-Functional Teams:
1) Improvement of cross-departmental communication: You would be amazed at how quickly individual contributors of different teams start to understand other department’s needs once they sit with them for a while and work on a shared goal. The marketer starts to understand why having dozens of tags firing is bad for the engineer. Once designers internalize the metrics from analysts, they start to work differently, and get satisfied by moving metrics instead of how beautiful their design is. The engineer sees how hard it is for the analyst to measure impact and starts to design better tracking systems and design better database storage.
2) Better ideas: The best ideas typically come from the intersection of people with different backgrounds working together on a problem. Designers, engineers, marketers, analysts et al. think differently. They solve problems differently. They have different strengths and weaknesses. Having them work together on problems almost always ensures a more optimal result.
3) Better prioritization: Instead of a product manager getting a list of requests from various teams, all those stakeholder are actively brainstorming together and measuring projects on potential impact to the metrics.
4) Increased speed: When teams aren’t spending time coordinating with other teams, disagreeing on goals and priorities, and struggling with inefficiencies, they produce results much faster.

Currently listening to Through Force of Will by Torn Hawk.

More On Building Effective Relationships At Work

I don’t think there’s anything I’ve heard people complain more about than co-workers or managers or employees they don’t get along with. People tend to categorize their co-workers as pure good or evil, leading to toxic relationships. Sometimes, these perceptions start from a simple misunderstanding that goes unresolved and grows over time. Sometimes, work styles are just incompatible. Most workplace relationship advice I had heard previously focused on getting to know the “real” person. Friendship, it seemed, was the only key to working better with these people, and friendship could only be attained by learning about people’s families, their passions, etc. This always felt forced to me, and when people attempted it on me, it felt manipulative. Also, some of the most effective teams I’ve worked on did not have this friendship. If that process works for you, stick with it. But if it doesn’t, let me tell you a story about how I learned to build more effective relationships at work.

When we hired our VP of Marketing at GrubHub, it created two problems for me. The first was I had gotten used to not having an active manager and doing things my own way. The second was I had developed a very direct style from working closely with the founders and other members of the team for a long time. As I continued my normal working style, that created problems for my new manager. She didn’t appreciate the direct tone of my emails, interpreting them as harsh criticism of her and others. She didn’t like the way I evaluated ideas. She liked short bullet points for emails. I tended to write paragraphs that covered a lot of details. Things went on like this for a few months, until she basically told me I had to change. This is a moment every employee should understand. The manager has communicated some feedback, and you can either ignore it and likely get fired, or apply it and stick around. So, the rule for building effective relationships with managers is to adapt to their style. They don’t have to adapt as they can just hire for people that fit their style.

After I adapted, I began to build a better relationship with my manager. She really valued personal growth of her team. So, as part of her process, we examined all of the issues I was having at work on a quarterly basis (which I highly recommend). When I was having an issue with a certain junior person at work, I described how this person operated, where their shortcomings were, how they needed more direction, how they responded to my requests, etc. She gave me some advice that really resonated: “Assume that person won’t change. How can you change to work with this person most effectively?”

As members of the workplace, it’s easy for us to see the flaws in how other people work. We spend a lot of energy hoping that those people will improve their performance in these areas. This is wasteful energy. While you should give direct feedback whenever possible, you can’t assume it will be heeded. So, you have to think about what you could change in how you work with someone to be more effective as a combined team. Sometimes, very simple changes can make all the difference. The only way to do this is to try different approaches and see what works and what doesn’t. Many people do this with their managers, but it’s even more critical to do it with your other co-workers. Identify the issues, brainstorm other approaches, and test them.

The Three Levels of Trust with Successful Co-Worker Partnerships

I think a lot about how to create and maintain highly effective teams. To be a part of a highly effective team, you need to have mutual trust with your team members. Many companies approach this problem organizationally, but the more I’ve thought about it, the more I’m convinced individual tackling it one co-worker at a time is the right solution. So, I thought about what trust really means with your co-workers, and found that there are really different levels of trust in an organization. I’ll break them down here.

Layer 1: Same Goals
It’s amazing how many workplace relationships never even get to this step. I’ve been lucky to work with really exceptional teams with shared objectives and aligned incentives, but the first question you should ask when trying to establish trust with a co-worker is, do you really have the same goals. In more political organizations, co-workers tend to see things as zero sum. If she gets what she wants, I won’t get what I want. This can be related to project allocation, budgets, headcount, etc. Even in non-political organizations, I’ve found other people tend to assume we don’t share the same goals.

So, what do you do if you don’t share the same goals? I have tried to break down what we’re trying to accomplish to find some middle ground. Usually, people are at companies for at least some similar reasons. If you have frank conversations with your co-workers, you can typically find those out and build from there.

Why is it important to have the same goals? If you don’t, you can never be confident a co-worker won’t undermine you/your plans, try to make you look bad, get you fired, etc. This sounds kind of rash and unrealistic, but you’ll be surprised how often these things happen.

Layer 2: Doing What You Say
Once you’ve agreed to the same goals, you need to divide work to reach those goals. The second layer of trust is having confidence that if someone has said they will do something, they will actually do it. This probably sounds minor, but it’s probably a more common problem than sharing the same goals. Co-workers are constantly bombarded with tasks, and can easily get side-tracked. Some people also over-commit regularly and let co-workers down. Some people really mean to get stuff done, then when they excitement wears off, they get lazy. If you can’t trust someone to accomplish what they say will accomplish, you will not have a successful partnership with them. Now, it’s important that you commit to this as well. You can’t have a successful partnership if you don’t care of your tasks as well.

Layer 3: Covering
A truly highly effective partnership is not just about having the same goals and doing what you say you will do, but also covering all the gray area in between what you agreed to do and what actually needs to get done. Consider a typical project. As a product manager, I may write the strategy doc or requirements I said I was going to write or do the appropriate research, and it may cover everything I talked about with the engineer I’m partnering with on the project. But, sometimes, it won’t really cover everything it should cover for the project to be a success. Not only could I have missed something, but there could be something that couldn’t be foreseen that’s really important to the success of the project. In that case, both me as the product manager and the engineer could have done everything we said we would do, and the project would still not be successful. Layer 3 is about covering for these gray areas. You want to feel confident in a team that if you forget something, your co-worker will catch it and address it. You need to be able to do the same.

It’s really stressful being in a role where you feel you need to be “always on”, and if that you’re not 100% perfect on everything you do, everything will fail. Covering is about have a team member that can pick up the slack when you miss something or when something comes up.

So, if you’re trying to build better relationships at work, think about which layer you’re at with your co-workers and how you can ascend to layer 3. If you build layer 3 relationships with multiple members on your team, you’ll execute at an extremely high level, and will be way more likely to succeed.

The Incredible Unbundling of Marketing

Having worked in marketing for almost a decade now, I have seen a lot of change. One of the most fascinating is the change of what people around me think marketing is and what it is not. To establish the baseline of how I think of it, and how marketers typically think of it, it helps to look at the official definition from the American Marketing Association:

Marketing is the activity, set of institutions, and processes for creating, communicating, delivering, and exchanging offerings that have value for customers, clients, partners, and society at large. Source.

Since that’s a mouthful, marketers tend to shorthand with a series of P’s (four to seven depending on who you ask). For products, those are product (the creating part of the definition), price (the exchanging part of the definition) , promotion (the communicating part of the definition), place (the delivering part of the definition), positioning (the value part of the definition), people (the people who do the activity), and packaging (another part of the communicating piece of the definition). For services, those are product, price, promotion, place, people, process, and physical evidence. These are consistent with how I was taught in my marketing undergraduate classes as well as those in my MBA.

If you visit that link above, you’ll notice the AMA also goes through the trouble of defining marketing research on the same page:

Marketing research is the function that links the consumer, customer, and public to the marketer through information–information used to identify and define marketing opportunities and problems; generate, refine, and evaluate marketing actions; monitor marketing performance; and improve understanding of marketing as a process. Marketing research specifies the information required to address these issues, designs the method for collecting information, manages and implements the data collection process, analyzes the results, and communicates the findings and their implications.

I’ll come back to research, but first, if schools are teaching what marketing is consistently, is this how marketing is being defined in the marketplace? At least in the tech industry where I’ve spent my entire career, increasingly no. Let’s break down some of these functions.

This is the process of creating something of value for customers. This is almost always its own organization lately, and with varying degrees of interaction with marketing. In technology companies, product managers are more likely to have engineering backgrounds than marketing backgrounds. I myself am a part of the Product org at Pinterest (though I was in the marketing org at every other company).

This is the process of determining the willingness to pay of different consumer segments, and setting a price that is attractive to the segments that are attractive to the company. Who owns this very much depends on the org. I have seen pricing owned by finance and sales more than marketing in my career.

Place is where the product/service is sold. In technology, the internet is the prominent place, and determining whether an app strategy makes sense is the key question people need to answer regarding place. Sometimes a marketing decision, sometimes a separate product org’s decision. For sales-driven companies, this is frequently owned by the sales org.

This is the act of making potential customers aware of and driving purchase of the product/service. Even in promotion, marketing functions are being splintered through multiple departments. PR is sometimes its own separate department. Many of the more direct marketing channels for technology companies (SEO, email, notifications, viral loops, conversion optimization) are unbundled into a separate product and engineering team typically called “growth”. Marketing still mostly has a stronghold on events, campaigns, and community management.

Positioning is the strategy of how a product is presented to potential customers. Many people refer to this as brand, but positioning includes functions of determining a market segment and deciding on a value proposition for that segment. Much of that can and should occur before a product is built. Positioning is about why a company exists and what is stands for. Much of this is still owned by marketing, but I have seen many companies independently position their product and create core values that do not reflect the positioning. This makes it hard to align market expectations with internal processes, and brands suffer as a result.

Packaging is the “coat of paint” that defines how a product is presented physically. It is every visual element of your product. This piece has mostly remained a marketing function, though I have seen separate design teams own this before.

This term has largely been replaced by the phrase “culture fit” in companies I have worked for, and is measured either by individual hiring managers or by HR or recruiting teams. As a result, this test has represented less about whether this person is a good reflection of our positioning to our customers and more about how well they will work with others internally, creating diversity problems.

These are the systems developed to deliver on positioning as someone experiences a service, like the line at a Chipotle or someone picking you up when you rent from Enterprise. These are increasingly managed by an Operations team.

Physical Evidence
With a service, there is a lack of tangibility to it, making it hard to value. Physical evidence re-inserts something physical into a less tangible experience to either create a memory or create an easier way for a customer to evaluate a service. This can be something out of the ordinary like a pink mustache with Lyft or a chocolate under your pillow at a hotel.

Now, let’s look at marketing research. In this case, I’ll discuss two newer disciplines encroaching on marketing’s stranglehold of these responsibilities.

User Experience
User experience teams frequently include their own research functions that do qualitative and quantitative analysis to identify problems and opportunities. Qualitatively, this occurs through one on one interviews or by monitoring individual product usage. Qualitatively, this occurs through surveys.

Marketing Performance Analysis
Data science or analytics teams have started to handle more of the monitoring of performance of product usage or marketing campaigns’ impact on growth. The rise of big data has made these processes require specialized statistical as well as technical skills.

Why is this unbundling happening?
I wish I had a stronger theory as to why this unbundling is occurring. Perhaps it is a reaction to years of abuse by advertising agencies and CPG companies trying to get us to buy cigarettes and saturated fats casting an evil stigma around the term marketing. Perhaps it is the technical founder re-imagining these skills with engineers at the core instead of MBA’s. Whatever the cause, marketing is being attacked on all sides, which has the result of redefining marketing with only the least impactful and measurable components, casting further doubt on the value of marketing.

What do we lose with unbundling?
I think the main thing one should worry about is whether an unbundled marketing structure or a bundled marketing structure is more effective, or at least knowing the trade-offs. The main issue that seems to occur with unbundling is that these separate functions lack a shared raison d’etre. Ways I have seen this manifest on the direct marketing side are growth experiments that violate brand guidelines, or a bias toward quantitative research when qualitative research may provide more insight. With operations, I have started to see as this moves further away from separate brand teams, efficiency trumps experience, and trade off discussions between those two things happen less often than they should.

The main issue with a bundled marketing organization is one of management. While the team is more likely to be aligned under one goal and set of rules, there are few if any people capable of managing a department with this large a scope that can have enough of an understanding of these functions to be valuable managers. This is both a failure of educational institutions to teach the “doing” element of marketing instead of just the strategy, a lack of on-the-job training to broaden employees’ view of the organization, and increasing complexity in performing all of the above actions. In this type of org, I foresee the strategy being great, but the execution being terrible because the strategy lacks an understanding of how execution really works.

Where do we go from here?
As I look through this analysis, I can’t help but feel conflicted. While I believe in the definition of marketing and experience some of the pains of these functions growing less and less aligned, I don’t see a rebundling fixing more problems than it creates in these organizations. So I can only hope that this post showcases the value of all of these elements working together, and that people working in these specialized roles start to take a broader view of what’s going on in the rest of the organization, start partnering more with these other teams, and create a better experience for the customer.

What do you think about the unbundling of marketing? How do you think we should react to it?

Thanks to Katie Garlinghouse for reading an early draft of this blog post.

Skill Optionality for Career Growth

Most firms operate via division of labor and specialization. So, as a budding employee at most companies, the task is to figure out what to specialize in so you have a place in the organization. This task is valuable in that in ensues a role for you in the organization and allows you to become a subject matter expert over time in a specific field. The world needs subject matter experts, so picking the right specialty early on is an important decision.

Specializing Too Early

The problem with that decision is when most employees are forced to make that decision i.e. in one of their first few jobs or during college when picking a major, they really have no idea what a great specialty will be, either from a matching their personality perspective or being needed in the marketplace perspective. Furthermore, universities are literally the least knowledgeable about what specialties will be valuable in the market because they largely exist outside of the labor market. This is why we’re pumping out hundreds of thousands of journalism majors even though there are very few jobs in journalism above the poverty line.

Making the Bet

Andy Johns has a great post on identifying individual strengths and making bets on skills that will become more valuable over time in the labor market. I would encourage anyone thinking about how to position their career to read it. The challenge is in making the right bets, and how bets can scale long-term for a career. For example, let’s say you make a bet to focus on social media marketing. Is that skill likely to be more or less valuable in four years? Now, the answer to this question has two components: 1) how many more firms will want these skills in the future and 2) how many more people will have these skills in the future. If the answer to 1) is a lot more and the answer to 2) is not much more, that’s a gold mine, but if the answer to both is a lot more, it may not be that great of a bet.

Scaling the Bet

The second component is how that bet scales. To answer this question, you need to know far can that skill can scale upward in an organization or in a career with that specialization. Sticking with the social media example, will there be Directors of Social Media Marketing in the future? VP’s? Chief Social Officers? If that seems likely, that’s another reason to focus on that niche. Unfortunately, no matter what specialty someone looks at, the answer is almost always that there is a cap to how far that specialty goes before one needs to acquire more skills to advance further. Sticking with social media, that track may end at Social Media Manager, and then you need to have PR skills to advance to the next phase at most organizations, or more online marketing skills like SEM, SEO, and email marketing. After that, whichever track you choose, you would need to fill in those other skills to become a VP of Marketing or a CMO.

This scenario becomes a problem for many people, as there are no convenient ways to develop new specialties over time. If you are working on social media all day, chances are low that your manager will proactively make sure you develop PR or SEM skills to prepare you to advance. So, people get stuck with the specialties and levels they are at after a while, and if those specialties become less important, they become less important over time. In a field like marketing, the skills needed at companies change very dramatically, as new marketing channels emerge over time, and as marketing channels mature, they tend to be less effective. Going back to school doesn’t help, as they tend to teach less skill-based functions, and certificates are mostly seen as a joke by employers.

Going T-Shaped

Fortunately, there are some ways to dodge these main issues, which are 1) not being able to predict the future on which skills are best to specialize in because the market moves too fast, and 2) not being able to add new skills over time to successfully manage more disciplines over time. Most advocate a position called being t-shaped ( good example here and example for search marketers here). The t-shaped model is great, but it misses some key features for career development I want to highlight.

1) One big bet vs. a few small bets

Becoming a t-shaped employee still requires placing a big bet on one area of depth and an 80/20 approach to close disciplines around that area of depth. There is still a large risk if that area of depth devolves in importance over time. The T-shape implies a static field and demand of skills that does not represent the labor market.

2) Individual contributor vs. manager

Management (and especially upper management) positions are not t-shaped. They are largely about strategy and goals. Having a ton of depth in a field like social media for a VP of Marketing is not especially useful as that person is expected to have someone that reports to him/her that has more depth. A senior manager needs to be able to spot opportunities to meet goals, and manage a diverse group of people to work together to reach those goals. Currently, many managers have a jarring experience where they begin to manage multiple disciplines and people not related to the depth that got them promoted.

3) Different T shapes work better at different companies

Sticking with the marketing examples, going deep in social media might be great for one company, but not very effective for others. Only knowing a little about other areas will likely make you not a great fit if social doesn’t move the needle. Ideally, you want your skills to work on a broad array of companies (though not too broad, like tech and industrial).

Solving the Problem

So how do you de-risk the choosing of a specialty and position yourself for more senior positions over time? That solution is to embrace skill optionality. What is skill optionality? It is maintaining exposure to different (especially emerging) areas of a field, and focusing, when necessary on specific niches, but not being stuck on one skill forever. Think of that t-shaped framework as a moving model where the component that is the stem of the T is gradually moving around as opportunities emerge in different fields and sharing depth with one or more other fields. So, instead of one long stem on SEO as in the Moz example above, it becomes a pi as SEO shares time with social media.

Skill optionality solves many problems for young or even experienced professionals. Young professionals gain exposure to many different fields, and are more likely to find one they excel at to either specialize in for the long-term or use as an initial depth from which to expand. For those with more experience, it breaks through the career ceiling of being just a subject matter expert, and protects one when certain expertises become less valuable. If you become experienced in four or five things, it is unlikely all five will become less important over time. Typically, as one skill becomes less important, another becomes more important. Skill optionality protects you from movements in the market related to skill needs.

I am not describing being a generalist, but just diversifying your portfolio of skills a little bit. I’ve seen a lot of blog posts that say every marketer in the future should know how to code, and use Photoshop, and know how to execute a laundry list of different marketing tactics. Chances are that’s too much for most people. How do you know if you’re going too general? Well, you should feel like the smartest person in the room on multiple subjects. If you’re a manager, it should that you or someone on your team is, and, if the latter, that you’re the next smartest after your team member. If you can’t claim a few broad areas as your domain, you’ve probably gone too far. In marketing, most companies win because they focus on a few channels, and become the best in the world at them. Skill optionality is not about executing in 30 areas, but knowing and adapting to which few maximally benefit your company/career, and being positioned to strategize as well as execute on those.

How to Execute

The next question you should be asking is how to execute this. I would like to make sure people understand that corporate America is not built for skill optionality; it is built for specialization. So, optimizing for this career path is fighting against the corporate ladder in every way early on in a career, but with maximum upside later on in a career. Some good strategies for young professionals is to shoot for positions that provide exposure to different areas in useful ways, like a web analyst who measures online marketing performance and therefore begins to understand different elements of online marketing. If that is not possible, another good strategy is to develop a specialty at a large company, and then move to a much smaller one where that skill is needed, but there is plenty of opportunity to develop new skills as no one else will be doing other things close to that specialty (or other things in the T one wants to develop). Some larger companies have rotational programs, which sound like a great way to get exposure to multiple areas, though, without having experienced these myself, I can’t say for sure.

The bottom line is that it’s important to pay attention to new opportunities to go broader in your skill set, and fight to get exposure to those opportunities. This will de-risk specialties that go out of favor in the market as well as better position you for management in the future. Even if you never want to manage, having multiple skill sets that you can adjust your depth on ensures you are always a viable candidate for many open positions, and will make you suitable for positions some companies desperately need.

Note: My examples are in marketing because that’s where I can get most concrete, but these examples can easily apply to engineers (via languages and elements of the stack), designers (with web, mobile, print, interactive, copy, coding, ui, ux), or just about any other discipline I can think of.

A Manager Should Act Like a Fullback

A common question I hear is what makes a good manager. There have been countless studies on this subject, and some good theories on specific pieces of management, such as situational leadership and being a shit umbrella. While both of those approaches are great, lately, I’ve been thinking that a manager is like a fullback.

For those that aren’t familiar with the player in football, a fullback is an offensive player who is primarily a blocker for running plays. On running plays, the offensive line opens a hole for the running back. The defense tries to close that hole by running into the hole. What a fullback does is go through the hole before the running back and pushes everyone out of the way so the running back can get through and and make a big play. If things go well, all you see if a running back running into the endzone without even being touched. The fullback is a crucial, but under-appreciated position in football because all attention is given to the person making the big play. But, in front of almost every great rusher there was been a fullback clearing a path for them.

So what a manager should do is learn about an opportunity from a star employee and clear a path organizationally for that employee to seize that opportunity for the company. The manager could come out a little bruised, but the employee hopefully scored a touchdown for the company.

Anyone else have any good analogs to management?