The Pragmatic Engineer

The Pragmatic Engineer

Preparing for Promotions Ahead of Time

Advice to get ready for a promotion, and a reality check on how they work.

Gergely Orosz's avatar
Gergely Orosz
Sep 28, 2021
∙ Paid

Question: I have missed out on promotions in the last cycle and am disappointed. What is your advice for maximising my chance of a promotion, next time? And what about just leaving this job, and joining somewhere else at the next level?

Whether you’re a manager or an individual contributor, the best time to start preparing for a promotion is months and sometimes even years before it’s due. At most tech companies, the next cycle of promotions will not start for another several months. 

In this issue we cover topics on preparing for promotions:

  1. Promotions at Big Tech. How does the system work, and what are good mental models to understand the process?

  2. Promotion philosophy vs reality at various types of companies.

  3. Advice for those wanting to get promoted on preparing ahead of time. What can – and should – you do, months ahead?

  4. Advice for managers on steps to take now to avoid surprises, so you don’t see people leaving unexpectedly, come promotions time. Bonus: a promotions map template for inspiration.

  5. Wait for promotion vs switching jobs? My take on this common question.

For a better understanding on how promotions and performance reviews work, see the article How performance calibrations work at tech companies. For advice on performance reviews, see the article Preparing for performance reviews ahead of time.

1. Promotions at Big Tech

Compensation, promotions and performance and how they are connected is different at Big Tech, than at most other companies. At most “traditional” companies, getting promoted is the obvious way – and often only one – to get more compensation. End-of-year bonuses are often tied heavily to company performance, sometimes fully independent of your own, personal performance. 

Big Tech takes a different view on promotions and compensation with regards to base salary, bonuses and equity compensation:

  • Pay for performance is a common approach at these companies. Your cash bonus and equity refreshers are tied to your performance, relative to your peers. The highest performing people often get outsized bonuses, 5-10x times that of the “average” bonuses.

  • Rules around promotion eligibility. Most companies require at least a 12-month tenure in the current role to be eligible for promotion, and some also mandate that only top performers – those calibrated to be in the upper half of ratings – be nominated for promotion. While most companies publish promotion criteria, some organizations can apply more strict criteria, for example, on tenure expectations.

  • A promotion moves compensation to the beginning of the next band. These companies have internal pay bands for base salary, and total compensation target. When someone is promoted, they receive a minimum salary increase (usually 8-10%) and are moved to somewhere near the base of the next level’s compensation band. Equity refreshers are allocated with this approach as well; promoted employees typically get some additional equity refresh, though this can be delayed by a cycle, and is often less than they hoped.

  • A top performing employee at a lower level can make more than an average performer at the next level. For example at Uber, top performing senior engineers or engineering managers would see as much as $320,000 in cash and equity bonuses in the US, and €170,000 in Europe. These people would make more compensation in the year than average performing engineers or managers at the next level. However, being recognized as a top performer – 3-5% of the senior engineering population – is a very hard feat.

  • Promotions are not the only way to recognize impact. Because of pay for performance, people with high impact can be recognized with strong bonuses, retention grants and by other financial means. Starting at the senior engineering level, promotions don’t become as pressing for many engineers, because the financial incentive is lower, compared to doing well at the current level.

Expectations for impact and competencies is a common way for these companies to frame performance and promotions conversations. All of them have expectations documented for all engineering and engineering management levels. To be promoted, people have to demonstrate performing at the next level on these two dimensions:

  • “What” – the impact of their work. They are expected to demonstrate business impact at the next level. For example, for a Staff engineer, this expectation might be to deliver a long-term effort to solve a meaningful problem at the organization level. The impact of the work is typically easier to quantify for customer-facing Product work, than some Platform work. Product work is often tied to company Key Performance Indicators (KPIs) like incremental revenue or cost savings. Platform work is often tied to KPIs one level below this, like systems latency decrease, reliability increasing or developer productivity increasing.

  • “How” – measuring against the competencies. Each company defines dimensions of expectations at each level, called competencies. For example, at Uber, these competencies were Software Engineering, Design & Architecture, Execution & Results, Collaboration, Creating Efficiency and Citizenship. Performing at the next level with these competencies almost always requires people to work well with others and stakeholders. In practice, some of the competencies will carry more weight than others. These differences are unique to the company, the promotion process, and sometimes even to the group within the company a person works in.

One of the biggest challenges for senior, and above-senior promotions at Big Tech is the continuous increase of impact radius and the complexity of problems. For example, to get promoted to senior engineering level, one is typically expected to showcase team-level impact. Staff engineers need to show organization or company-wide impact. Senior engineering managers and directors face similarly raised expectations.

While a company is growing rapidly, there are plenty of opportunities that fit these bills. A hypergrowth company has no shortage of organization-wide woes that engineers and managers can attempt to solve. When they succeed, promotions typically follow.

However, the more mature a company becomes, the fewer these opportunities; and the more complex and time-consuming they are to solve. The opportunities might become so rare that failure is not an option, as it would blow the promotion opportunity of an engineer or manager who knows how rare these organization-wide problems are. 

Having fewer “promotion-worthy” opportunities also incentivises coming up with success criteria that can turn a failing project into something that can be perceived as a success. And why would people not do so? They put in the effort and did the work; admitting the project did not succeed would result in poorer performance reviews and promotion outcomes at most places. I’m playing devil’s advocate here.

Promotion-driven development

Promotion-driven development is an unfortunate, but very real consequence of the impact-driven nature of promotions and performance reviews at Big Tech. With each level, your work is expected to impact larger groups, and drive more impactful business outcomes. 

How can you impact a large group of engineers, and drive a major business outcome? An obvious answer is to solve an engineering problem the organization has. However, to get promoted, the “what” (impact) is not enough. You need to perform at the next level of the “how” (competencies) as well. In practice, this means building something non-trivial that involves coordination of large groups, and is a difficult engineering challenge.

This is exactly why, instead of using an existing third-party framework for an organization-wide issue, engineers working towards Staff promotions will write their own, in-house solution. As justification, they will find edge cases where existing solutions do not work well enough, and they’ll go ahead and execute this complex, impactful – and promotion-worthy – project. 

Though at times building custom tooling is justified by unique needs, there is little to no incentive for any engineer to go with an off-the-shelf vendor solution that solves the problem. This approach would be labelled as trivial and would not meet the complexity expectations the software engineering and architecture/design competencies demand at senior and-above levels. 

Promotion Driven Development is one of the reasons why all of Big Tech has built custom solutions for everything, from developer tools, through observability tools, to rebuilding products that did not take off. I’m not exaggerating: Google built 20 different chat products over 16 years, often in parallel. It is certain that each of these chat projects got dozens of engineers and managers promoted to the next level, before another group made the case to start a complex project from scratch with an even larger impact, instead of fixing up the existing one.

2. Promotion Philosophy vs Reality

When should someone get promoted? The reasons are similar at most companies, and tend to be one or more of these:

This post is for paid subscribers

Already a paid subscriber? Sign in
© 2025 Gergely Orosz · Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for great culture