JobTailored
All guides

What a skills-first layout looks like after a parse

By · Published

A functional resume, also sold as a skills-based resume, replaces the dated list of jobs with a set of skill headings. Program Management, Client Relations, Process Improvement, each with achievement bullets underneath, and then a thin block near the bottom naming the employers with the dates shrunk or left off. Almost every page explaining the format recommends it to the same three readers: career changers, people with a gap in their history, and people writing a first resume.

What those pages do not show is what the layout does to a document that software reads before a person does. So I wrote one person's history twice, identical sentences in both versions, and ran each through our own uploader this morning. This page is what came back, and what to build instead.

The format moves one thing and loses another

Two things change. The skills section grows into the body of the document and takes the achievement bullets with it, and the employment history shrinks to a list of names with the dates reduced or absent. Nothing is added. What leaves is the pairing: which job each accomplishment belongs to, and when.

Why that pairing matters to the human reader is the subject of a career change resume that keeps its dates, which argues the case against switching formats at all. This page is the mechanical half, what becomes of the file before any of that reading happens.

What came back from the parse

Our uploader turns a file into text and then into a structured record: contact details, a summary, a skills list, roles with their bullets attached, education. It is a heuristic rather than a model, so it works by looking for the headings people actually use, and it is the parser I can describe exactly rather than guess at.

The reverse-chronological version came back as two roles. Each carried its own date range and its own bullets, under the right employer, in the order they were written.

The functional version, written the way the templates show it with bullets under headings like Program Management and Financial Analysis and the employers in a block headed Employment History, came back with no work history at all. Not one role. The achievement lines and the employer name had been read as entries in the skills list, because the last heading the parser recognised was the skills one and nothing after it looked like the start of a job. A reader of that record would see a long list of competencies and a person who had apparently never worked anywhere.

Then I gave the format its best case: same skill clusters, but the whole block headed Experience, a word the parser knows. That version did produce roles, four of them, titled Program Management, Financial Analysis, the employer's name, and the university's name. The two skill clusters had been read as job titles. The employer had been read as a job with nothing in it.

The second result is the one worth sitting with, because it did not fail loudly. It produced a confident, well-formed record of a career that does not exist. If a recruiter later searches that record for an employer or filters on a date, there is nothing there to match.

One parser, ours, and other systems will differ in the details. The shape of the risk does not differ: a layout whose headings are skill names asks every parser to guess which block is a job, and you never get to see the guess. If you already have a functional resume, that part is checkable rather than theoretical. Our free ATS resume checker shows you the text that comes back out of your own file, so you can see whether your roles survived the trip.

Where it is disqualifying rather than merely risky

Some employers ask for the dates in writing. NIST's application tips for its own vacancies tell applicants to include, for each position, "your job title, employer, dates of employment (month and year), hours worked per week", and warn that the people reading it "cannot infer experience from job titles or general statements". A format whose selling point is that the dates are played down cannot satisfy a requirement that the dates be stated, position by position. Federal hiring is the clearest case rather than the only one: any application that asks you to confirm employment dates is asking for exactly what the layout removed.

The one case where it holds up

Handed straight to a person, a skills-first page can work. A referral, a recruiter you are already in conversation with, someone who asked for it after meeting you, anywhere a human opens the document and nothing parses it first. Most guides to the format make this point and it is the one to keep from them. It is also narrower than it sounds, because it describes the document you hand over rather than the one you submit through a form, and most applications are the second kind.

Build the hybrid instead

The emphasis the format promises is available without paying for it:

  • Keep the dated roles. Reverse-chronological, with employer, title and month-and-year dates on every one. This is the spine, and it is what makes everything above it checkable.
  • Put a skills block above the history. This is where the vocabulary of the posting belongs, and it reads as a summary of the document rather than a replacement for it. Our resume skills generator will draft the list from the posting and your background if the one you have is a jumble.
  • Reorder the bullets inside each role so the transferable work leads. Promoting a bullet costs nothing and invents nothing. Where the bullet you have just promoted ends on a duty rather than an outcome, how to quantify resume bullet points covers how to recover the figure it should end on.
  • Write the top line for the role you want. This is the one place a career change is argued in the document itself, and the resume summary generator drafts it from the role you are aiming at rather than the one you last held.
  • Keep the headings boring. Experience, Skills, Education. What a plain heading buys you is set out in what an ATS resume is, and our free ATS resume templates page grades the usual layouts on how they come through an import.

Once the hybrid is built, run it through the ATS resume checker and read the record it gives back rather than the score: every role present, each with its dates and its own bullets still attached to it. That is the whole test the functional version fails.

Sources