How to Add a Job Board to Your Website

Most careers pages are a paragraph about culture and an email address. That is not a job board, and the gap between the two is measured in applications you never received.

A proper board does three things a paragraph cannot. It shows every open role in one scannable place, so a candidate can tell in ten seconds whether there is anything for them. It gives each role its own page, which is what makes it findable in search and shareable in a message. And it closes roles when they are filled, so you stop fielding applications for a job that no longer exists.

This guide covers building that: what belongs in a listing, which filters earn their place, how expiry should work, and the structural detail that decides whether your roles show up in job search results at all. You can build the board from a job board widget template and spend your effort on the content, which is where the applications actually come from.

Worth saying up front what this is not about. The board is the shop window: it lists roles and lets people browse them. The application form, the thing a candidate fills in with their details and their CV, is a separate feature with separate problems. This article is about the window.

Why the careers page is worth more than it looks

Recruiters are expensive and job boards charge per posting, so it is easy to treat your own careers page as an afterthought. That gets the economics backwards.

Candidates who arrive at your own site are the best ones you will see. They found you rather than a listing, which means they were already interested in the company rather than shopping a category. Their applications are better, their interviews are better, and they cost nothing per hire beyond the page existing.

Search is the second reason. People search for job titles with a place attached, constantly, and roles with their own well-structured pages can appear for those searches. A single careers page listing five roles in body text competes for nothing. Five individual role pages, each properly marked up, each have a chance.

The third reason is quieter. Your careers page gets read by people who are not applying: customers checking whether you are growing, partners sizing you up, candidates who will apply in a year. A board with real roles on it signals a functioning company in a way a culture paragraph does not.

What every listing needs

There is a minimum below which a listing generates questions instead of applications.

The job title, written the way a candidate would search for it. This is the field people get wrong most often. Internal titles like "Growth Ninja" or "Member of Technical Staff II" are invisible: nobody searches for them. Put the recognisable title first and save the internal one for the offer letter.

The location, and be specific about the arrangement. Remote, hybrid or on-site, and if remote, which countries or time zones you can actually hire in. "Remote" with no qualifier wastes everyone's time, because most companies cannot in fact employ someone anywhere on earth.

The employment type. Full time, part time, contract, internship. It changes who applies and it takes one dropdown.

The team or department, which is how a candidate skims a board of twenty roles down to the three that concern them.

A description with actual content. What the person will do, what they need, and what the first few months look like. The temptation is a list of requirements; the thing that gets replies is a clear picture of the work.

The salary or a range. This is the one that provokes an argument internally and settles it in the data every time: listings with pay information get more applications, and increasingly they are required by law in a growing list of jurisdictions. If you cannot publish a number, publish a band. Publishing nothing filters out exactly the confident candidates you wanted.

The posting date, which tells a candidate whether they are looking at something live or something that has been sitting there since spring.

Optional but useful: a closing date if there is a real one, a named hiring manager, and a short note on the interview process. The last of those costs a sentence and reduces drop-off noticeably, because uncertainty is the main reason good candidates stall.

Filters, and how many is too many

A board with three roles needs no filters. A board with forty needs them badly, and the instinct is to add more than help.

The two that almost always earn their place are department or team, and location or remote status. Between them they cover the questions nearly every candidate asks first: is there anything in my discipline, and can I work from where I live.

Employment type is worth adding if you genuinely mix full time with contract and internships. If ninety percent of your roles are one type, the filter is clutter.

Seniority is tempting and rarely works, because the labels are not comparable between companies. A candidate's own sense of whether a role is right for them comes from the description, not from a badge saying mid-level.

A free-text search box is worth having above roughly twenty roles and pointless below it. On a short board it is slower than reading.

Two rules for whichever you keep. Show the counts, so a candidate can see that Engineering has six roles and Finance has none without clicking. And never let a filter combination produce an empty page with no way out: if there is nothing matching, say so and offer to clear the filter.

The board template handles the mechanics, so this is a matter of deciding what to switch on. Set it up on the job board widget and adjust as the board grows.

Each role needs its own page

This is the structural point that most home-built careers pages miss, and it has the largest effect.

If all your roles live in one long page, you have one URL. It cannot rank for a specific job title, a candidate cannot send a colleague a link to just that role, and you have nothing to put in a social post. Everything competes with everything.

Give each role its own address and all three problems disappear. The page can be found for its own title, shared on its own, and closed on its own when the job is filled.

While you are there, mark the page up so job search engines can read it. There is a standard structured data format for job postings, and search engines use it to place roles in dedicated job search results. It wants the title, the description, the posting date, the employment type, the location and the hiring organisation. Most of that is what you already typed into the listing, which is the argument for using a board that emits it rather than hand-writing pages and hoping.

One warning on this. Structured data must describe a real, open role. Marking up filled or fictional positions to appear in job search results is the sort of thing that gets a site removed from those results, and the removal is much harder to undo than the shortcut was to take.

Expiry is a feature, not an afterthought

Nothing damages a careers page faster than roles that never close.

A candidate who applies for a job that was filled in March and hears nothing does not conclude that your page is stale. They conclude that your company ignores applicants, and they tell people. Meanwhile every application to a closed role is time somebody on your side spends sorting rather than hiring.

So decide the lifecycle before you publish. Set a closing date on each role, or a default window after which a listing is automatically hidden. Remove the role from the board the day it is filled, not the week after the new person starts. And decide what a closed role's page does: showing a short note that the position has been filled, with a link back to the board, is far better than a dead link, both for the candidate and because the page has probably accumulated some search value you would rather not throw away.

Structured data has its own view of this. A posting that has expired should say so rather than continuing to claim it is open, which is another reason to let the board manage the state instead of managing it by hand.

Building and embedding it

The build itself is quick once the content decisions are settled.

Start from a job board template so the listing structure, the filters and the individual role pages already exist. Then add your roles, writing each description properly rather than pasting the internal requisition, because the description is the part that produces applications.

Style it to the site. A careers page that looks like a different company is a real signal to candidates, who are reading everything for clues about what working there is like. Match the typeface, the colours and the button style you use elsewhere.

Decide where applications go. This is the equivalent of the lead routing problem on any other form: a board that collects applications into an unwatched inbox produces the worst possible candidate experience. Route them to the person who will read them, and reply to everyone, even briefly.

Then embed it. The widget produces a snippet that drops into WordPress, Squarespace, Wix, Webflow, Shopify or plain HTML. Put the board on a page linked from your main navigation, not buried in the footer, and link to it from your About page, because that is where interested people already are.

Test it as a candidate on a phone. Most job browsing happens on one, often in a spare five minutes, and the filters and long descriptions are what suffer on a small screen.

Writing a description that gets replies

The board is structure; the description is the part that converts. Most are written as internal requisitions and read like them, which is why so many produce silence.

Lead with the work, not the company. A candidate skimming six listings wants to know what they would spend Tuesday doing. Two or three concrete sentences about the actual work beats four paragraphs of mission statement, and the mission statement is not what they are deciding between.

Separate what is genuinely required from what is merely nice. Long undifferentiated requirement lists suppress applications from exactly the people you want, because thorough candidates read a list of twelve bullets as a checklist they fail and confident ones read it as a company that does not know what it needs. Three or four real requirements, then a short optional list, is honest and gets better results.

Say what the first ninety days look like. It is the single most useful paragraph you can add and almost nobody includes it. It tells a candidate whether the role is what the title implies, and it forces whoever wrote the listing to think about whether the job is actually defined.

Name the process. Two interviews and a task, or four rounds over three weeks, said plainly. Candidates drop out of processes they cannot see the shape of, and the sentence costs nothing.

Write the salary line carefully. If it is a band, say what moves someone within it. A range with no explanation invites the assumption that everyone lands at the bottom.

And keep it shorter than feels comfortable. A description that fits on a phone screen without three swipes gets read to the end. One that does not, gets skimmed for the pay and abandoned.

Keeping the board honest as it grows

A board with five roles maintains itself. A board with thirty does not, and the failure is gradual rather than obvious.

Give one person the job of reviewing it, on a schedule, however light. Roles get filled and nobody remembers to close them; a department gets renamed and the filter splits in two; a location changes from hybrid to remote and only the internal document is updated. None of these are dramatic on their own and together they turn a useful board into one candidates stop trusting.

Watch which roles get applications and which get views without them. A listing with traffic and no applications is usually a description problem or a missing pay range, and it is the cheapest thing on this list to fix. A listing with no traffic at all is usually a title problem, which is also cheap to fix once you notice.

Finally, keep the closed roles' pages doing something useful rather than deleting them outright. A filled position with a short note and a link back to the board is a small courtesy to whoever arrives late, and it holds onto whatever search value the page earned while it was open.

Common mistakes

Using internal job titles. Nobody searches for your invented title, so nobody finds the role.

One page for every role. It cannot rank, cannot be shared and cannot be closed independently.

Omitting pay. It reduces applications and is increasingly a legal requirement rather than a preference.

Listing "Remote" with no geography. You will get applications from countries you cannot employ in, and you will disappoint all of them.

Leaving filled roles up. It is the fastest way to convince good candidates you do not reply.

Too many filters. Four filters on a board of six roles is furniture, not help.

No posting date. Candidates assume the worst about undated listings, and they are usually right.

An unwatched inbox. The applications arrived and nobody read them, which is worse than not having a board.

Frequently asked questions

Do I need a job board if I only have two or three roles?

A board is still worth it, mostly for the individual role pages rather than the browsing. Two well-structured pages that can be found and shared beat one careers page mentioning both. What you can skip at that size is filters, which are pure clutter until you have enough roles to need narrowing down.

Will my jobs show up in Google's job search results?

They can, if each role has its own page with valid job posting structured data covering the title, description, posting date, employment type, location and hiring organisation. A board that emits that markup automatically is the practical route, because hand-maintaining it across a changing set of roles is where errors creep in. The role must genuinely be open: marking up filled or invented positions risks losing access to those results altogether.

Should I publish salary ranges?

Yes. Listings with pay information consistently attract more applications, and pay transparency is now legally required in a growing number of places. If a precise figure is impossible, publish a band and say what it depends on. Publishing nothing does not keep your options open, it filters out the candidates who have options.

What is the difference between a job board and an application form?

The board lists your open roles and lets a candidate browse and filter them. The application form is what a candidate fills in for a specific role, with their details and CV. They are separate features that work together, and it is worth getting the board right first: a good form attached to an unfindable role produces nothing.

How long should a job stay on the board?

Set a real closing date, or a default window of a few weeks after which it is hidden automatically. The important rule is not the exact length but that expiry is automatic rather than remembered, because remembering is the part that fails. Take a role down the day it is filled.

Can I add a job board to Squarespace or Wix?

Yes. The widget gives you an embed snippet, and both accept one through their code or embed block, as do WordPress, Webflow, Shopify and static HTML. Nothing about the board depends on the platform underneath it.

Should closed roles stay online?

Keep the page, change what it says. A filled role whose page shows a short note that the position is closed, with a link back to the current board, is more useful than a broken link and holds onto whatever search relevance it built up while it was open. What must not happen is a closed role continuing to advertise itself as open, either on the board or in its structured data, because that is how you collect applications nobody will read.

Where to start

A careers page becomes a hiring channel at the point where each role has its own findable page, the board closes roles by itself, and the listings say what the job pays. Those three things account for most of the difference between a page that produces applications and a paragraph with an email address under it.

Open a job board template, add your open roles with real descriptions and a pay range, switch on the two filters that help rather than the five that do not, and make sure applications land somewhere a person will read them today. The rest of the widget library covers the pieces around it.