Engineering is one of the most critical functions at a tech company; in fact, as every company becomes a tech company in some form, it’s critical to every company. But in tech companies in particular, engineering is responsible for developing the core product at the heart of their business: what they’re selling, what people are using and buying. And while there are potentially many leaders in the technical product function — from architects to CTOs — the head of engineering or VP of engineering is the executive that grows and manages the engineering team, is responsible for assessing the time it will take to deliver features (or products), and delivering quality releases on that schedule.
But understanding the role only in terms of logistics oversimplifies its scope, missing the many other critical and nuanced functions a star VP of engineering provides. While scaling an organization, it’s challenging (if not impossible) to determine delivery dates, hit all the right dates, and prioritize development by group consensus. This doesn’t even include the importance of setting the right culture, which is critical for retaining engineering talent that would otherwise go elsewhere … in this market, it’s easier for them to move than stay. A good VP of engineering would own all those things, while also shipping quality product and giving the rest of the organization transparency into the process.
Not understanding these other aspects of the job, startups often hold off on hiring an engineering manager too late in the product development cycle. Then they have to deal with all the internal issues that arise from not having a competent VP of engineering managing things: broken process, a poorly assembled team, a bad balance between engineering and product management, a frustrating engineering culture.
I’ve had the good fortune to work with multiple exceptional engineering executives; every time we brought one on board, they addressed major internal issues — and in every case I wished I had have brought them in sooner! So below are some of critical aspects of the job that you should keep in mind as you search for the right candidate…
Product planning
Most people rightly focus on an engineering manager’s ability to execute, but their role prior to development is also critical to the product-planning phase.
For an engineering organization inside a tech company, product development efforts often require expanding on R&D in order to deliver a particular product or feature set at a more practical, production-hardy scale. And while product management/marketing will provide insight into the features, market size, and pricing, the engineering leadership provides feedback into the resources (headcount, equipment, etc.) needed to deliver by a given date.
This is generally an iterative process; it’s not as neatly sequential as one might assume. The CEO is working within a set of budgetary and other constraints, and needs to strike a balance between features that will expand the market or value of the product, with the headcount and capex spend required to get there. Accurately pulling together the execution estimates, and then managing and updating that forecast, is critical to running a strong product business. Without a strong VP of engineering, this can be a perilous task.
Why can’t one just directly source the information from the engineering team? The pressure on engineering is extra strong, especially in a startup; the discourse often positions engineering as if they’re the only thing standing between the company and its success. Put in that position, engineers will often do what they think is right for the company — take on technical debt, work nights and weekends, or skimp on quality assurance — to get a release out on time. Individual engineers will sometimes be approached independently by product managers or field sales reps to get an “accurate” read of what it really takes to get a particular feature out, and since no one wants to be the naysayer, these estimates result in an overly optimistic timeframe that doesn’t take into account the broader agenda and competing priorities of the organization. Which means corners will be cut, technical debt will accrue, and features will disproportionately skew to only those needed by the immediate sales cycle (vs. other strategic development).
A good VP of engineering, however, brings sanity to this process with a unified view of the broader objectives. They know when to pull in a feature at the expense of something else, and how to account for that later.
Building the engineering team and culture
Clearly, the power of a good head of engineering can’t be reduced to managing product features and releases alone. The most important function of a VP of engineering is to build out the engineering team and set a startup’s engineering culture, especially when the organization gets to the size where that middle layer of management is required. And while there are clear markers of skill level for strong engineers — open source credibility, performance on coding tests, etc. — management is a more qualitative domain.
A strong VP of engineering should therefore be able to determine who will be good managers for the team, hire them, and be able to groom that skill set over time. Because building a strong team isn’t a static thing: Startups grow fast, and the engineering organization needs to scale with it. It’s also very easy to lose the culture of the early team as size doubles or quadruples. A strong VP of engineering knows how to balance maintaining existing culture with evolving a new one. And as an engineering organization starts to balkanize into sub-projects — which have to interoperate, yet are held to separate deadlines — good engineering leadership can manage the dynamic between the groups.
But when hiring, how do we know what good engineering leadership looks like? The easiest way to assess a VP of engineering beyond their on-paper qualifications is to take a look at their existing and prior teams: Did they meet scheduled release dates? Did they have a sensible (vs. haphazard) process? Was there too much discord (i.e., unproductive vs. creative friction) between teams? I’ve found that engineering managers are best reflected by the groups they manage. What you see in their teams is likely what you will get in them.
Ensuring execution
Ultimately, the output of any engineering organization is product. A well-functioning organization will not only build a high-quality product, but also hit pre-determined dates and accurately forecast updates to the timeline as it changes. A good VP, therefore, has full command of the software development process, and ensures the correct process is in place and followed.
Much has already been written about engineering processes and approaches, so I won’t dwell on that here; instead, I’ll focus on several characteristics of a strong engineering executive that aren’t immediately obvious on the surface but that help enormously in execution:
Maintaining morale (and therefore high work output). Often morale and output seem somewhat opposed; more side projects, time off, and ancillary training are assumed to lead to better morale. But startups often don’t have this luxury. Strong engineering management tends to give their teams enough ownership and latitude that they are happy and fulfilled in driving the product forward.
Building boxes, not arches. Programmers and architects generally appreciate elegance in their design. And often the designs are sufficiently general that a significant investment needs to be made in the platform before producing anything of value. I’ve heard this process likened to building using arches, where the structure can stand on its own until the last piece (the keystone) is slotted in at the top to hold everything together. This may be elegant, but in the sprint to finding repeatable product fit in a market, showing value and getting customer feedback is critical. Competent engineering management should therefore be able to push the team towards more practical, incremental designs that can garner useful external feedback quickly — without compromising the long-term generality of the system. The VP’s role here is not producing the architecture, but ensuring that incremental release is a real requirement in the design process.
Handling chaos. There’s a popular saying in tech businesses: “the only thing worse than not having any customers is having customers”. When a product launch goes right and catches the world on fire, the feedback from the market can be overwhelming. Even with marginal success, engineering will be inundated with customer escalations, feature requests, questions from the field, pressure from technical partners for integration, and meetings with customers. All of this must be addressed while continuing to drive the product forward. When they’re under pressure, organizations look up: A hallmark of a strong VP of engineering is to maintain composure while navigating this chaos.
Account management. While largely internally facing, the head of engineering can also play an important role in pre- and post- sales for companies that ship product. Customers will often want to talk with the VP of engineering to gain an understanding of the QA process for example. Early in the development lifecycle, there really isn’t much beyond vision and the strength of the team; a strong VP of engineering can use this alone to instill the necessary customer confidence. The VP of engineering is also often pulled in on the post-sales side as visible customer features start to slip or product issues come up. Having executive engineering leadership in the room with the customer is one of the most effective means of regaining customer confidence, whether it’s through describing the root cause of a bug they’ve faced and how it’s being addressed, or explaining the reasons behind a critical date slippage.
* * *
I remember when, in the early days of Nicira, we pushed out the first generally available version of our product. The walls immediately started to crumble. Sure, we had expected to uncover some bugs in the field, but the inbound pressure on engineering far exceeded what we had anticipated. In addition to customer escalations, the sales engineering team was requesting usability improvements to help streamline proof-of-concepts for what was an exceptionally complicated product; meanwhile, support was pushing for immediate changes needed for remote debugging. Customers immediately pushed the product past our published scaling limits, and were using it in ways we hadn’t imagined, causing problems we had no ability to reproduce in house.
As a company founder, every morning I felt like I was walking onto a ship where everything was leaking. The first person I’d often visit was our VP of engineering, Rob Enns, who had just recently joined the company. I’d ask him how far we were from sinking — “are we going to make it?” — and he’d reassure me and then tell me to leave him alone and to “go run the business”. Which I did. But looking back, things would have been a helluva lot easier if we had brought him on earlier!
I believe all startups should hire a VP of engineering, and early — bringing them on before the problems begin, even if this may seem a bit too soon relative to total headcount. Many young startups believe they don’t need a VP of engineering because their engineering team is either sufficiently senior or perhaps too small. But it’s hard to overstate the importance of the role. A strong VP of engineering provides the backbone for an engineering function that withstands the pressure of the business, while also growing the org and without letting it collapse under the weight of an ad hoc process. Setting the right culture and the right process is something you want endemic to the organization, independent of size — and that’s exactly what hiring the right VP of engineering at the right time will do.
