The Posting Pay Screen, Scored Before You Spend an Application

You are applying to many roles at once, and every application costs time you cannot get back. This guide gives you a pay screen you run before you spend that time: set one private floor once, then score any posting - range listed or hidden - into apply, flag-and-verify, or skip in under a minute. It runs earlier than the usual advice, which starts triangulating a target range only after a posting already has your attention.

Most compensation guidance handles the number too late. It teaches you to read one posting's band, or to answer the desired-salary field, once you have decided to apply. That is the wrong order at volume. When you are screening dozens of postings a week, and roughly half of them show no number at all, you need pay as a triage gate, not a negotiation input. The gate has one job: stop you from burning applications on roles that will never clear your floor.

Why a floor beats a target for triage

A single private walk-away floor is the only pay input that exists on every posting. A target range is useless the moment a posting shows no number, and employers negotiate toward the low end of any range they do show, so anchoring your triage on a target you cannot see is a fantasy.

The numbers force this. In the United States, the share of online postings with pay information has risen from an average of about 15 percent before January 2018 to roughly 53 percent since January 2024. That is real progress, but it also means nearly half of what you screen still arrives with no band. A target range cannot triage those. A floor can, because a floor is yours - you carry it into every posting whether or not the employer shows a number.

The floor is private. It is not the number you type into an application, not the number you tell a recruiter, and not your target. Its only purpose is to answer one question fast: is this posting worth an application on compensation grounds? Everything downstream - negotiation, the desired-salary field, the offer call - is a different job with its own guide. Here you are guarding your application budget.

A target range is a fantasy on half your postings. A floor is a number you carry into every single one.

Setting the floor

Set your market value first. Pull title, level, and location comps from multiple sources so you have a defensible market-low, a midpoint, and a stretch number. The documented method is explicit: before you can set a minimum, you need to understand what your skillset and experience are worth for this particular job and level.

Then adjust. Take the market-low and move it for the tradeoffs a specific search involves - commute, workload, risk, and how steep the learning curve is. Your floor is the market-low plus or minus that personal adjustment. For context on what the market looks like right now, the median advertised U.S. full-time salary sat at $62,400 at year-end, up 3.4 percent year over year. Use that only as a sanity check on your own comps, not as your floor.

Reading the pay field: three branches

Every posting lands in one of three branches, and you decide which in about ten seconds. A range is listed, the field is hidden, or the language is vague ("competitive," "market rate," "DOE"). Vague counts as hidden - it is a non-number dressed as a number.

The branch determines what you do next. A listed range goes to a direct floor comparison. A hidden or vague field goes to a band estimate from signals. Do not skip the classification step; treating "competitive" as if it were a number is how people talk themselves into applications they should have flagged.

Scoring a listed range against your floor

When a range is listed, check two things: the low end against your floor, and the width. The low end tells you whether the role can clear your bar. The width tells you whether the range is even usable.

Width is a data-quality signal, not just a number. Enforcement and case law are converging on an unwritten norm of roughly 30 to 50 percent range width. New Jersey caps the spread at 60 percent of the minimum. A 100 percent-plus range like $90k to $200k is exactly what enforcement actions in California, Colorado, and Washington have targeted as bad faith. So the width test comes first: if the range is wider than about 50 percent, treat it as effectively no range and route it to flag-and-verify. Only score a range against your floor when its width is plausible.

Once the width passes, the low-end test is simple. If the low end is at or above your floor, the number clears - apply. If the low end is below your floor, do not skip on reflex, because a listed range is often not a binding cap. Many transparency laws require the posted band but do not require the offer to land inside it; one employment view puts it plainly, that if there is a legitimate, non-discriminatory business reason to change the pay, the employer is free to do that under the law. A below-floor low end on a strong role becomes a flag, not a skip.

Estimating the band when no range is shown

When no range appears, estimate a band from observable signals and attach a confidence level. You will never be precise here, and that is fine - you are deciding whether to spend an application, not signing an offer.

Four signals do most of the work, and each one can lie:

• Level and title. A wide-scope, multi-level title is a weak signal. A posting can be written to be entry, mid, or senior with enough flexibility to shift it based on applicants received. Read multi-level scope language before assuming anything about pay.
• Employer type and size. Cross-reference against disclosure law. If the employer is large enough to be covered in its state and still shows no range, the omission carries more weight than for a tiny employer with no obligation.
• Benefits language. A base range says nothing about total comp. Ranges are usually base only, so a base-lean posting with strong equity or bonus language may pay well above what a base figure suggests.
• Overtime-threshold marker. A posting priced under $58,656 has been cited as a signal that the role may be non-exempt and overtime-eligible. Treat this as directional only: the 2024 federal rule behind that figure was later vacated in court, so it is a heuristic, not current law.

The disclosure-law cross-reference is worth building into your screen, because a missing range is not automatically an employer trying to hide low pay - roughly a quarter of listings covered by transparency laws still omit salary. Knowing the size threshold for the state tells you whether absence is suspicious or simply legal.

These thresholds move as new laws take effect, so re-check the current list for the states you apply into rather than trusting a fixed count. As a reference point, 18 states plus Washington, D.C. had statewide pay transparency laws on the books at the time this was written.

The gap rule: apply, flag, or skip

The gap rule turns your floor comparison and band estimate into one of three routes. It is a ten-second decision, and the routing depends as much on your position as on the number.

• Apply when the low end (or your estimated band) is at or above your floor. The number clears; spend the application.
• Flag-and-verify when the number is below your floor but within reach and the fit and employer are strong. Do not spend a full application on faith. Ask the recruiter for the band before deep investment.
• Skip when the number is far below your floor and the role is a common title drawing a large applicant pool.

Applicant volume is the hinge. Where volume is high and the role is standardized, there is almost no negotiating room - a marketing coordinator role with 150 applicants has little wiggle room beyond the stated range, so a below-floor number really is below-floor. Skip it. Where your skill is scarce, the band is softer and you can push above it, so a borderline number becomes a flag.

Scarcity, not title, sets your ceiling. In Refolk's index of professional profiles, US software engineers listing Rust number 608 against 55,849 listing Python - Rust talent is 91.9 times rarer. A candidate that scarce can negotiate above a posted band; a candidate in a deep pool cannot, and should trust the floor.

Geography compounds this. The same posted floor implies different apply-anyway odds depending on how deep the local pool runs.

With 8.1 times more software engineers in the US than the UK in Refolk's index, a deeper pool means the range is closer to firm and your room to push a below-floor number is smaller. Read your own leverage before you decide whether a borderline posting is a flag or a skip. If you want to check where you sit on the scarcity curve for a given skill and market, Refolk can size the pool for a specific role and location so you know whether you can push a band or should trust the floor.

The step-by-step screen

Here is the whole procedure in order. The first two steps run once, before the search. The rest run per posting.

There is a real disagreement worth naming. Some practitioners argue you should never let a listed range deter you and should apply broadly, because you can negotiate the number but you cannot negotiate from a table you never reached. That is a conversion-maximizing strategy, and it is defensible when applications are cheap. This guide takes the effort-conserving view, because at volume your applications are not cheap - each one is tailored time. The gate exists to protect that time. If your applications are near-free, lower the gate; if each one costs real effort, keep it high.

How this screen goes wrong

The screen fails in specific, repeatable ways, and every one produces a false positive - a role you scored wrong. This is the most valuable part of the standard, so give it weight. Each failure below has a check that catches it.

Two of these deserve extra attention because they cut in opposite directions. The 100 percent-plus range and the missing range both look like "no usable number," but they route the same way: flag-and-verify. A $90k to $200k band is not more information than a blank field; it is a blank field with extra words. And the "no range means low pay" assumption is the single most common way people skip roles they should have applied to, because a real share of range-free postings simply span levels.

The last failure mode is the quietest. A quarter of postings covered by transparency laws still omit salary, so the absence of a range does not reliably tell you the employer is small or exempt. The only way to know is to cross-reference employer size against the state threshold - which is why that cross-reference sits inside the band-estimate step rather than as an afterthought.

Asking for the band is leverage, not a red flag

Asking a recruiter for the band early is a normal, expected move that costs you nothing. Missing ranges cost employers applicants, and that fact is your leverage: 41 percent of professionals lose interest in a posting with no listed range, and in one poll of 2,000 U.S. job seekers, 33 percent said they would not interview without seeing salary first.

That means the recruiter already expects the question. Use it on every flagged role. The script is short and does not disclose your floor:

Hi [name], thanks for reaching out about the [role] position. Before we set up a screen, could you share the base salary band and a rough sense of total compensation for this level? I want to make sure we are aligned before we both invest time. Happy to walk through my background once I know we are in the same range.

Confirm two things on that call: the base band and total comp. Ranges are usually base only, and a base-lean role with strong equity or bonus can clear your floor even when the base does not. Reading base as total comp is how good roles get skipped.

Refolk can shorten the verify step. When a posting shows no range and you need someone who can confirm the band, Refolk can surface the recruiters and comp leaders at a specific company and size so you are asking the right person, not the general inbox. And once a role clears the screen, Refolk writes and tailors the application itself, so the effort you conserved at the gate goes into roles that will actually clear your floor.

Keeping the screen current

Run this final check before you trust any single decision. The screen is only as good as its inputs, and two of them drift: your floor and the disclosure-law thresholds.

Re-set your market value when the market moves under you - the median advertised salary rising 3.4 percent year over year is enough drift to matter over a long search. Re-check disclosure thresholds for the states you apply into, because new laws take effect on their own dates and the covered-employer count keeps rising. Both are quick to refresh, and both keep the gate from silently letting the wrong roles through, or blocking the right ones. The screen is not a one-time setup; it is a standard you keep sharp for as long as the search runs.