Modern major ATS parsers handle most two-column resumes, but not all parsers are modern, and columns built with text boxes or tables still scramble reading order in older systems. A single column has a zero-percent structural failure rate. Use two columns only when you know the audience — and never put critical keywords only in the sidebar.
A parser reads the text layer of your PDF in extraction order, then tries to segment it into experience, education, and skills. A cleanly built two-column layout — real column flow, left-to-right blocks — usually extracts in a sensible order on current engines behind Workday, Greenhouse, Lever, and their peers. The engines were forced to get better because designed resumes became common.
The failures cluster in construction technique. Columns faked with tables can interleave: parser reads row one across both columns, then row two, producing sentences spliced from your sidebar and your job history. Text boxes are worse — several extractors skip floating objects entirely, so whole sections silently vanish. The visual result is identical to real columns; the machine result is not, and you cannot tell which you have by looking.
You do not know which ATS, which version, and which parsing engine sits behind any given application. Against that uncertainty, the two layouts have asymmetric risk: single-column fails structurally never; two-column fails somewhere between rarely and badly, depending on construction and the recipient's software vintage. The upside of columns — denser sidebar, designed look — is real but modest, and recruiters do not award interviews for layout.
A workable compromise if you love the look: keep the left or right rail for low-stakes content — contact block, languages, interests — and keep everything a screen depends on in the main flow: titles, employers, dates, achievements, and the skills the job description names. Then a sidebar mishap costs decoration, not candidacy. Test either way: copy-paste your PDF into a plain text editor and read the order that comes out, because that is roughly what the machine reads.
Single column wins for high-volume online applying, conservative industries, and any pipeline where an unknown ATS sits between you and a human — which is most of them. It also wins under content pressure: full-width lines hold quantified bullets better than a narrow main column squeezed by a sidebar.
Two columns win where a human opens the file directly: design and creative roles where the resume is itself a work sample, direct-email applications, networking handoffs, and printed copies at interviews. Many candidates sensibly keep both — a designed version for humans, a single-column version for portals. If maintaining two files sounds tedious, that is what the single-column version is for: it works acceptably in both worlds, while the reverse is not guaranteed.
The current engines behind major platforms usually do, when columns are built as real column flow. Table-based and text-box layouts still fail unpredictably, and older ATS versions remain in wide use — so correct parsing is likely, not guaranteed.
Select all text in your PDF and paste it into a plain text editor. If sections come out whole and in a sensible order, parsers will cope. If sidebar and body text interleave mid-sentence, a machine reads the same garble.
Put skills in the sidebar only if they also appear in your bullets. Keyword matching runs on extracted text, and if the sidebar extracts badly, skills listed only there can vanish from your parsed profile — sidebars are safest for low-stakes content.
Upload your resume to the free scorer — it parses your actual file and shows whether your layout survives machine reading intact.
