Build a promotion case when the management rung narrows
Map the rungs with demand, match your evidence to the strongest one, and draft a promotion case that fits the org chart.

The management rung is narrowing
If you are a software engineer deciding whether to keep aiming for management, start with the shape of the org chart, not the title you want. The share of U.S. software engineers moving into engineering or software management roles decreased from 8% in 2019–2022 to 6% in 2023–2026. The same narrowing shows up among U.S. SEO specialists, who optimize search visibility: the proportion making advancement moves into management fell from 17% in 2019–2022 to 15% in 2023–2026. The default move is still staying inside the craft: half of U.S. software engineers moved into another software engineering role as their next profile job.
Management remains real, but the opening is less frequent, so your case has to be sharper. A smaller rung changes what a promotion case must prove. You cannot rely on the title alone. You need evidence that your work already reaches the next level: broader scope, clearer ownership, and outcomes that matter outside your immediate team. If the management role is scarce, the case has to show why you can fill that role without expanding the org chart.
Adjacent rungs need real demand
Adjacent rungs are only useful when the org chart has a place for them. Among U.S. software engineers who changed occupations, AI engineering and machine learning engineering appeared as new transition destinations, with neither role prominent before 2023. The founder option is smaller, but it is moving: the proportion of U.S. software engineers who moved into founder roles increased from 2% in 2019–2022 to 3% in 2023–2026. The same founder shift is visible among U.S. SEO specialists: the proportion who moved into founder roles rose from just over 2% in 2019–2022 to just under 4% in 2023–2026.
Use that demand as a check. A rung with no openings, no budget, and no clear reporting line is not a promotion target. Look for an open role, an existing budget line, and a reporting line that makes sense without inventing a new position. If you can point to all three, the rung is real. If you can point to only one, treat it as a stretch option, not a promotion target.
The case must fit the rung your evidence supports
A promotion case needs proof, not intent. It should connect the rung you want to the work you have already done. Keep it short enough that a busy manager can read it in a single sitting, and concrete enough that the evidence can be checked against real outcomes.
- Count the open rungs above you. Pull the org chart and list the manager, senior manager, and adjacent technical lead roles that exist in your company. Done means a short list of actual roles, not aspirations, with the number of open or likely open rungs in the next levels above you.
- List adjacent rungs with real demand. Add the rungs where your company has hiring, internal movement, or a named owner. You are done when you have a short list of technical, product, and founder-adjacent rungs that someone else has already occupied or is actively filling.
- Match your evidence to the strongest rung. Compare your shipped work, team impact, and scope to the outcomes that rung requires. The mistake is to choose the rung with the best title and then assume the missing evidence will appear. The output is a single rung where your evidence covers most of the required proof, with the remaining shortfall named clearly.
- Draft the promotion case around that rung. Write the rung name, the evidence, the shortfall, and the ask on a single page. The result is a case a manager could defend without inventing a new role, a new title, or a new reporting line.
Keep the ask concrete. Say the rung, the level, the effective date, and the evidence you are asking the reviewer to weigh. If the case depends on a new role, a new title, or a new reporting line, it is harder to defend. That gives the reviewer a clear standard to apply, and it keeps the ask tied to work that is already visible.