The Pay-Shape Reference by Role Class, and What a Total Hides

Before you commit a search to a role class or an industry, you need to know how pay is built there, not just how big the number gets. This reference is for the job seeker choosing where to aim: it translates a quoted total in an unfamiliar field into its guaranteed floor, its at-risk portion, and its realizable equity, so a base-heavy public-sector total and an OTE-loaded sales total can sit in the same table on the same basis. Jump to the row you need; each section stands alone.

The published compensation guides benchmark one role's number or decode one posting's range. None maps how pay is composed across segments so you can choose before you have an offer in hand. That is the gap this fills.

Why a total number hides more than it shows

A quoted total tells you the ceiling, not the floor. The same headline can mean "money I will bank" in one field and "money I might earn if a lot goes right" in another, and the only way to know which is to split the number into its parts.

Pay has three components: guaranteed base, at-risk variable, and equity. Base pays regardless of performance. Variable pays on attainment, which is rarely 100 percent. Equity pays on a liquidity event that may never come, or comes years late and diluted. A single total blends all three into one figure that flatters whichever component is least real.

This is why the "bigger number" test fails across segments. In government, wages and salaries account for 61.6 percent of employer compensation cost, with benefits at 38.4 percent - a base-dominant shape that guarantees most of its headline. In quota-carrying sales, half the headline may be variable that only pays at exactly 100 percent quota, and median attainment across many plans can sit in the 30s and 40s after a quota hike. The public total guarantees more per headline dollar even when the sales total is larger.

The headline tells you the ceiling. The mix tells you the floor. Only the floor is yours.

Pay shape by role class: the lookup table

Here is the default pay shape for the major role classes a searcher chooses between. Read the row for your target, note what carries the variance, and treat these as starting assumptions to confirm, not guarantees.

Two patterns matter here. First, base share is inversely related to how much of the story is upside: the more a field talks about total, the less of it is guaranteed. A full-cycle SaaS AE runs a 50/50 base-to-variable mix; an SDR or BDR runs 60/40 or 70/30; in industrial and non-tech markets, splits lean heavily toward base, often 70/30, 60/40, or even 80/20.

Second, the variable itself takes different forms that fail differently. An IB analyst's bonus is 100 percent cash and typically ranges from 70 to 115 percent of base, so a first-year analyst on a $115K base might receive an $80K to $130K bonus, totalling roughly $190K to $245K. A tech IC's third component is equity, which introduces realizability risk that a cash bonus never has. Consulting sits between: base-heavy with a capped performance bonus of 10 to 35 percent of base depending on level and rating, where the cap is a ceiling and not a promise.

What a quoted total actually guarantees, component by component

The guaranteed floor is base alone, the only amount paid regardless of performance. Everything above it is a claim about the future that needs a discount before it belongs in a comparison.

Start with the variable. A 60/40 mix on $200,000 OTE means $120,000 guaranteed base and $80,000 variable earned when the rep hits quota exactly. OTE - on-target earnings - is the pay at precisely 100 percent attainment, a single point on a distribution that describes almost nobody. The mix stays 50/50 on paper, but realized comp drifts toward base because nobody hits 100 percent consistently. A 40/60 mix on a quota that 80 percent of the team misses is a pay cut with a story.

The accelerator is where OTE misleads most. A base rate applies up to quota and an accelerated rate applies above it; accelerators commonly run 1.5x to 2x the base rate, and per WorldatWork most enterprise sales plans include this structure. That makes realized pay bimodal: overperformers clear OTE, and the median lands under it. The single OTE number sits in a valley almost no one occupies.

Now equity. A cash bonus varies in size but is spendable when paid. Equity varies in realizability: whether it converts to money at all, when, and at what discount. These are two different risks and they need two different discounts. Do not fold them into one adjustment.

Discounting equity: private versus public, options versus RSUs

Equity risk scales with company stage, not with your role. The same "TC" is worth materially less at a private firm because the shares have no market price, sit behind a liquidation preference, and often carry double-trigger vesting that delays value.

Two adjustments do most of the work. Common employee stock is priced below preferred, often 20 to 40 percent lower, because of the rights gap. On top of that, a discount for lack of marketability applies: the 2024 AICPA practice aid recommends 15 to 35 percent for typical venture-backed private companies, reaching 40 to 50 percent for very early or distressed firms. As a percent of value, the illiquidity discount should shrink as the firm grows larger. Companies switch to RSUs on average 5.5 years after incorporation, at an average post-money valuation of $1.05B, so the presence of RSUs is itself a maturity signal.

Options and RSUs are not the same instrument. Stock options require payment to exercise; if the stock price is below the exercise price they are underwater and typically will not be exercised. Public companies grant options-to-RSU at commonly 2-to-1 for higher-volatility stocks, up to 3-to-1 for mature companies, which tells you roughly how many options it takes to match one RSU of value.

The procedure: decode a quoted total into a comparable figure

The job is to turn a headline into three numbers you can defend: guaranteed floor, risk-adjusted variable, and risk-adjusted equity. Work in order, and stop when both targets are stated on the same basis.

Practitioners disagree on one point of order. Some discount equity before variable, because private equity can be worth zero and there is no point risk-adjusting a commission line if the equity swamps everything. If your target is an early private company, run step five first.

Target: __________            Target: __________
Guaranteed base:  $______     Guaranteed base:  $______
Variable at 100%: $______     Variable at 100%: $______
  x median attainment: __%      x median attainment: __%
  = risk-adj variable: $____    = risk-adj variable: $____
Equity headline:  $______     Equity headline:  $______
  x (1 - DLOM): __%             x (1 - DLOM): __%
  x liquidity prob: __%         x liquidity prob: __%
  = risk-adj equity: $____      = risk-adj equity: $____
NORMALIZED TOTAL: $______     NORMALIZED TOTAL: $______

Tailoring an application to each of these targets - and reflecting the right mix language back in a cover letter - is exactly the friction Refolk removes once you have decided which segment to chase.

How this goes wrong: failure modes and false positives

Most bad comparisons come from a small set of repeatable errors. Each has a false positive that looks like a good offer and a check that exposes it.

• Reading OTE as guaranteed. A $200K OTE treated as salary. Isolate base only, then ask for median attainment. Realized comp drifts toward base because nobody hits 100 percent.
• Equity headline read as cash. A private "$400K TC" treated as spendable. Apply a DLOM of 15 to 35 percent and confirm a liquidity event exists; much of the total is illiquid equity value.
• Ignoring quota inflation. The mix looks generous but the quota was raised. Check the rolling 12-month median attainment; below 65 percent signals decorative variable, and medians can sit in the 30s and 40s after hikes.
• Treating options like RSUs. Valuing underwater options at grant value. Options require payment to exercise; below the exercise price they typically will not be exercised at all.
• Comparing gross totals across sectors. A base-heavy public total and an OTE-loaded sales total look equal at the headline and are not on a risk-adjusted basis. Normalize to the guaranteed floor first.
• Assuming bonus caps are targets. Consulting and IB ceilings are not promises. The maximum in the offer letter is a ceiling; only the top 5 to 10 percent of performers reach it.
• Trusting self-report medians as population truth. Self-report samples skew high through selection bias. Cross-reference with government employer-cost data.

The deepest of these is quota inflation, because it is invisible in the mix. A 50/50 plan reads the same before and after a quota hike, but the realized variable collapses. The attainment distribution, not the mix ratio, is where the upside actually lives. Ask for it in writing before you sign.

Why the sources disagree, and which to trust for what

No single source gives you clean base, variable, and equity together with population-level rigor. You pick sources by what each one proves and where each one lies.

Government employer-cost data is a rigorous survey that breaks cost into wages and salaries, supplemental pay including nonproduction bonuses, and benefits. Its blind spot is equity: it measures employer cost, not stock, so it undercounts pay shape in equity-heavy tech. Use it as the trustworthy anchor for base share and for fields without equity. In private industry, wages and salaries averaged $32.60 per hour and 69.9 percent of employer cost; supplemental pay averaged $1.84 per hour in private industry versus $0.66 in state and local government.

Self-report platforms separate base, bonus, and equity from what workers volunteer, which is the only clean view of equity you will get. Their weakness is selection bias: the sample reflects workers who choose to report, which skews high. One such source reported 2025 median total comp of $226K for a software engineer, $312K for senior, and $457K for staff - useful for shape, not for a population median. Notably, 70 percent of reporting tech workers receive equity, while only around one-third of companies are observed offering stock, which is itself a selection tell.

The practical rule: use government data for the floor and for fields without equity, use self-report platforms for the shape of equity-heavy fields, and cross-reference the two whenever a total is both self-reported and equity-loaded.

Thin markets make totals harder to sanity-check

Talent-pool depth signals mobility and benchmarkability, not pay level. A deep pool means many comparable offers to triangulate against; a thin one means a quoted total is hard to sanity-check because there are fewer peers to compare it to.

In Refolk's index of professional profiles, the US Account Executive pool is 240,568 against 17,286 in the UK, a 13.9x gap, and the US Software Engineer pool is 349,217 against 22,300 in Germany, a 15.7x gap. Within the US, the software engineer pool is 1.45x the account executive pool.

The mechanism matters for decoding a total. In a thin market, fewer comparable offers exist, so a quoted number is harder to place. If you are weighing a role abroad or in a small segment, budget more time for benchmarking and lean harder on component-level sources, because the headline has less local context to correct it.

Before you commit your search: the check

Run this before you pick a target segment or accept that a headline is worth chasing. If any item fails, you are comparing marketing, not pay.

Keeping this current

Pay shape drifts, so re-check the two inputs most likely to move rather than trusting a stored value. The first is attainment: median quota attainment across large plan samples swung from 11.1 percent to 74.3 percent in a single year, so the rolling 12-month figure for your specific plan is the number to ask for, every time. The second is equity discounting: the AICPA practice aid updates its DLOM guidance, so confirm the current recommended range before you model a private grant.

Everything else in this reference is structural and moves slowly. Sales mixes cluster where they have long clustered, IB analyst bonuses stay cash-heavy with deferral rising by seniority, and government stays base-dominant with rich benefits. When you decide where to aim next, open this table first, decode the total into its floor and its two discounts, and compare on that basis. Once you have chosen the segment, Refolk writes the resume from your history and tailors it to each posting in that field, so the deciding work you did here carries straight into the applications.