2023 was a turbulent year in tech. Tech companies ranging from FAANG to venture-backed startups faced layoffs brought on by economic uncertainty, decreased funding velocity, and valuation resets. The rapid adoption of generative AI has also brought on new challenges. In 2024, 45% of firms are exploring how to integrate AI into their products and processes. The upshot of these trends is a frenetic development cycle and an increased pressure to ship, which, in many cases, is driving high rates of software engineer burnout.
In uncertain times, delivering value is critical. Finding a high-performing team structure based on your company’s resources, deliverables, and business needs is essential. Engineering teams must strike the right balance of seasoned senior talent and fresh junior perspective to handle the pace of development and continually evolve. The question is, “Which is the right model for my organization, and why?”
Below, we’ll explain the different approaches to engineering team structures and how they can empower your engineers to do good work. We’ll also cover the impact of seniority on the org chart, when and how to tap into early career talent, and the considerations for teams looking to do so.
According to Matt Eccleston, former VP of Engineering at Dropbox, “There’s no such thing as a perfect org chart.” Still, it’s possible to weigh the tradeoffs of each structure or even mix-and-match your way to success. These are some of the most common ways teams approach the question.
This is a traditional team structure that many of the older companies still follow. It involves grouping people by roles, such as front-end, back-end, mobile, and infrastructure, and having multiple levels of management to oversee them. It’s a familiar structure that companies might find easy to understand, and it allows developers to collaborate on shared goals and skill areas.
The downside is the risk of slower product delivery owing to the coordination and effort needed to share information across teams. There’s also the potential for disconnect between the lower- and higher-level employees, as well as slower decision-making due to longer chains of command.
The matrix organization seeks to remedy the siloed nature of the function-based structure. While role-based groups exist in this model, engineers from different groups come together to work on software projects, each with its own project lead. Meta (Facebook) uses a version of this model with product-based and geographic divisions and a flat hierarchy.
While this approach helps foster cross-team collaboration, it tends to treat software delivery as a one-and-done project—overlooking the need for continuous iteration. Plus, engineers with multiple reporting lines can cause confusion and friction regarding priorities and ownership of deliverables.
This model has attracted a lot of attention over the last decade, even though Spotify itself has moved on from it. The Spotify model consists of autonomous teams (or “squads”) in charge of different functions. Teams have the flexibility to make decisions without having to chase down approvals, essentially functioning as small startups.
The problem with this model, again, is that team interactions can become complicated. Adjusting to this model can involve significant effort from employees who are more used to traditional structures, and they may struggle to decide who makes the final decisions in the absence of a clear hierarchy.
There is also the risk that different squads might choose different workflows, leading to inconsistencies in the company offerings. Loyalties also tend to become squad-oriented, rather than product or company-oriented, which can lead to clashes when trying to make decisions that affect the product or the company as a whole.
As the name suggests, this model organizes cross-functional teams around building a unified product. The organization assembles teams by joining all the roles needed to focus on a specific product area, reporting up to one line manager. Several large companies, such as Airbnb, use a version of this model by building product teams around core personas, such as "guest" or "host. "
This is similar in some ways to the matrix model, but there are two core differences. One — the product team model permanently reassigns employees to a cross-functional team, rather than just bringing them together for a specific project. And two — as a consequence of point one, the product team model removes the problem of dual reporting lines, allowing engineers to be clear about their priorities
While this model optimizes collaboration and delivery speed, it may not be as flexible as it sounds. Interpersonal friction is a major risk as engineers spend more time adjusting to new management, operating norms, and product knowledge when functions change.
Moreover, there is the risk that engineers within the same function will be siloed on the basis of the product they’re working on. This could inhibit their ability to collaborate or agree on things that affect the function as a whole.
This is the model most startups and small companies tend to favor. It’s ideal for smaller teams, usually 10-15 people, as it minimizes hierarchy and keeps things moving quickly. It consists of large generalists, with senior engineers or product managers serving as functional team leads. The focus is on quick decision-making, high participation from each employee, and clear alignment on goals and expectations.
On the flip side, the lack of formal hierarchy could lead to clashes within the team as those with more dominant personalities exert their influence over others. Newer/junior employees might feel overwhelmed by their autonomy and struggle without a supervisor to guide them.
There is also a risk that the company might hire too many junior engineers during the early building phase, and then struggle to provide them with appropriate learning and growth opportunities later on.
An organization’s structure will almost certainly change as the company grows. It’s important to reevaluate your team structure every three to six months to assess what your current priorities are and whether the structure you have is best suited to those priorities. Could your team benefit from some extra developer support? Do they perhaps feel overwhelmed by having too many stakeholders to report to? Keep an eye out for signs that things might need to change, and be prepared to implement those changes quickly.
Building an optimal software engineering team is, in many ways, like designing a piece of software. You’ll want to optimize for a few key priorities at the start and evolve as you go. Here are a few factors to keep in mind:
Engineering teams need to be able to make decisions on their own. Customers expect product and feature upgrades faster than ever, and too much red tape can be a barrier. At the same time, there needs to be accountability, so that each team is prioritizing the most impactful enhancements.
Takeaway: Have a structure that supports autonomous team operations while keeping cross-functional channels with product and UX teams open. Periodic mentor-led review sessions can help iron out any difficulties and ensure everyone is aligned on the right priorities.
The product roadmap changes quickly, especially for high-growth companies and startups looking for product-market fit. The organization structure should enable engineers to move across functions with minimal impact on the flow of work and delivery timelines as focus areas shift.
Takeaway: One approach is investing in more “jack-of-all-trades ” engineers at the growth stage rather than deep experts who might find it difficult to keep shifting focus. Hiring early-career engineers can also be a good move, as they’re typically more receptive to exploring multiple roles.
Maintaining a flat, close-knit structure is easier when you’re a team of 10 or 15. Once that number crosses 50, you’ll likely have multiple teams that interact only periodically. At this stage, communication is critical to ensure teams don’t lose sight of the bigger picture.
Takeaway: Regular check-ins should always be a priority, especially if a change is in the cards. For example, if you’re hiring a new leader or changing the way teams collaborate, make sure everyone understands exactly why those changes are necessary for the business and how their day-to-day will be different as a result.
Software companies find themselves growing faster than expected to keep up with market demands and customer needs. At early-stage companies, this can mean frequent pivots in product strategy and deliverables. It’s important to have an engineering org structure that can scale up and down as business needs change, without oversaturating the team or hindering collaboration.
Takeaway: The top-down hierarchical approach is generally not ideal when you’re scaling fast. Have a mix of teams, each with a defined purpose, and let those teams have a say in when and whom to hire.
Adopting and adjusting your organizational structure is just the first step. You’ll need to iterate on and build an approach that works for you, your business objectives, and your people. Here are a few best practices to get you started.
The best engineering teams strike a balance between autonomy and alignment—allowing them to ship fast while keeping stakeholders in the loop. But finding the right balance is a function of your goals. Are you prioritizing a quick go-to-market approach? Or are you focused on operational efficiency? Are you investing in robust R&D for each product, or are you taking a “grow-as-you-go” approach based on a minimum viable product (MVP)? When choosing or reevaluating your team structure, think about what you want to accomplish and how your team structure levels up to those larger goals.
It can get easy to treat engineering teams as amorphous units, especially as organizations grow. For leaders, it’s important to invest the time to help each engineer, whether new or experienced, to understand how their actions tie into bigger business goals. Who owns what? Who are the people/teams relying on them, and who’s in charge? This alignment helps engineers take full ownership of what they do and progress without excess oversight.
Any change in your org makeup needs to loop back to the why and the what then. For example, most organizations bring in experienced operators with a deep understanding of your challenges or problem space for leadership roles. On the other hand, external additions can block the growth path for someone else on the team, while the costs of bringing them up to speed and the risks of mismatching with senior engineers on the current team may very well offset the benefits. When it comes to people decisions, map out the scenarios and thoroughly evaluate the pros and cons before proceeding.
An engineering org structure should fit the company’s goals, not the other way around. And given that goals can evolve, it’s generally wiser to implement smaller iterations on the go rather than making drastic changes. Remember, no two teams ever work the same way — and no two leaders ever lead the same way. The goal should be figuring out how to bring the best out of your teams and leaders so that they’re willing to speak up, ask for what they need, and adjust whenever they have to.
Any organizational structure is only as good as the people on the team. Getting the right balance of engineering skill sets is critical—and so is finding the right mix of early-career, mid-career, and senior team members.
Early-career hires are eager to learn, tech-savvy, and come significantly cheaper than more senior hires. However, it can take time — anywhere between three to nine months — to onboard early-career engineers and ramp them up to where they’re contributing optimally. Companies looking for immediate change may not have the luxury of time.
Mid-career hires have more industry experience and can adjust to their new roles much sooner. They can, however, be more resistant to new ideas than early-career hires owing to their added seniority. Plus, mid-career engineers have much shorter tenures at organizations compared to a few years ago. Over-investing in them can thus lead to retention problems and a lack of organizational continuity.
And finally, while many companies are emphasizing senior technical talent to help them through the current economic downturn, over-indexing senior talent comes with its own set of challenges. These teams run the risk of not being open enough to new ideas and opportunities and taking a “my way or the highway” approach to leadership and management. And of course, hiring senior talent comes at a hefty price tag.
This isn’t to say that any one category is superior to the other. How you structure your hiring, again, will depend on what your priorities are as a company. If you need to make a quick pivot in overall direction, hiring senior talent with strategic experience makes sense. And if you are looking to build a cohort of talent with long-term loyalty, choosing early-career hires and investing in their potential is the way to go.
Investing in early-career talent is widely regarded as one of the best strategies for bringing fresh perspectives and dynamic talent to the table. Gen Z, in particular, is eager to embrace new techniques and processes, such as coding with generative AI, and tends to pick up such skills quickly as a digital-native generation. When provided with strong mentorship and growth opportunities, they can also become loyal teammates, making them excellent long-term assets. And hiring early-career talent is significantly cheaper overall than hiring mid-career or senior-level talent.
But this doesn’t mean you should rush to hire junior engineers without consulting existing team members first. Early-career hires will require mentoring, so the timing and scale of your hiring are conversations to have with your current team.
“What I ask myself, at any given point,” says Nathan Esquenazi, CodePath CTO, “is whether I have the leadership in each of the workstreams and the infrastructure to support junior engineers properly and help them develop their skills by getting the mentorship they need.”
When hiring early-career talent, it’s important to have an experienced team lead to manage each major workflow. This is preferably a senior engineer with deep product expertise. Tapping mid-level engineers for day-to-day supervision and mentoring is also a good idea.
Here are a few key questions to assess whether your team is ready to bring on junior talent: