Employer Branding Guide

Why do strong candidates decline before you ever meet them?

Because they formed a judgement from evidence you did not choose. Somebody weighing a move looks at the careers page, the tone of the job ad, the public profiles of people already doing that job, review sites, and whatever a former colleague says in a message. Almost none of that is controlled by a recruiter's outreach. When the only available signal is a generic advert and a page of stock photography, candidates fall back on whatever they can verify elsewhere, and a person with options will not spend a Saturday researching an employer who has told them nothing. This is why weak outreach response and surprise declines are so often a reputation problem misread as a sourcing problem. Volume does not fix it. What fixes it is giving a candidate enough specific, checkable detail to form a view: what the team is building, how the work is organised, what the genuine trade-offs are, and who they would be sitting with every day. This is also why a strong recruiter cannot rescue a weak page. Outreach opens a conversation, and the page is what somebody checks before replying to it.

What is an employee value proposition and how do you find yours honestly?

An employee value proposition is the specific trade you offer: what somebody gets here that they would not get elsewhere, and what they give up in exchange. Both halves matter, and the second is the one most companies delete. You do not write an EVP in a workshop. You find it by asking people and listening for the phrase that repeats without prompting. That phrase, rather than the one the leadership team prefers, is the EVP. Then test it against evidence. If the claim is autonomy, name a decision an individual contributor made last month without asking permission. If the claim is growth, name three internal promotions in the past year. If nothing can be named, the claim is an aspiration, and aspirations belong in a strategy document rather than on a careers page. An EVP that cannot survive a new joiner's first fortnight stops being a brand problem and becomes a retention problem, showing up in first-year attrition rather than in a survey. Write it in the words people actually used, because a sentence borrowed from a competitor is recognised as borrowed.

• › Recent joiners: what nearly made you take the other offer instead?
• › Long-tenured people: why did you stay through the hardest quarter?
• › Leavers: what do you actually miss, and what would have kept you?
• › Everyone: what should a candidate be warned about before they join?

Why is the careers page the highest-leverage employer brand asset?

Because it is the one surface you fully control that candidates visit at the exact moment they are deciding, and every other channel points at it. What earns its place on the page: what the company builds and who for, in plain words; how teams are organised and what a week actually looks like; the hiring process laid out step by step with realistic timelines; your position on pay, location and remote work stated rather than implied; and real people with names attached. What does not earn its place: a values grid, stock photography of a meeting that never happened, and a paragraph about passion. Loading speed and mobile rendering matter more than design flourish, because many candidates arrive from a phone during a break. Keep every open role one click away and make applying obvious from any page. A careers page builder that lets the talent team edit without an engineering ticket is worth more than a beautiful page nobody can change. Review it twice a year, since a page describing the company of two years ago reads oddly to anyone joining today.

How do you get employee stories that do not sound scripted?

Interview people, do not brief them. The difference shows up in the first sentence. A briefed employee produces the line they think communications wants; an interviewed one produces a specific week, a project that went sideways, or a decision they were nervous about. Ask for specifics: what were you doing this time last year, what surprised you in your first month, what would you warn a friend about before they applied. Publish some of the awkward parts, because a story with no friction in it reads as marketing and gets discounted entirely. Format matters less than voice. A written question-and-answer piece, a ninety-second phone video, an engineering post about a real problem, and a conference talk all work. On advocacy, make sharing easy and voluntary: give people the link, offer a suggested line they are free to ignore, and attach no obligation. Mandated posting is obvious to every reader and costs more credibility than it buys. One honest story a month beats a campaign nobody finishes. Ask permission before publishing, keep the right to withdraw a story, and check in before reusing one a year later.

What should you do about employer review sites?

Read them as feedback first and reputation second. Comments that repeat across different reviewers are usually true, and they are cheaper to fix than to argue with. Respond publicly to a portion of reviews, positive and negative, in a human voice and without legal boilerplate, naming what you are changing where you genuinely are changing something. Do not reply to every review, since a wall of near-identical corporate responses reads worse than silence. Never incentivise reviews or run a campaign to bury a bad one: platform policies vary, profiles do get flagged for manipulation, and internally it converts a feedback channel into theatre. If you ask for reviews at all, ask everybody at a natural moment such as a work anniversary rather than asking only the people you expect to be positive. The uncomfortable part is that a pattern of reviews about one manager or one stage of your process is a management problem wearing a marketing costume, and no response strategy will fix it. Share the recurring themes with the leaders who can act on them, in private, with the review text attached rather than summarised.

How do you write a job ad that reads like a person wrote it?

Lead with the problem the person will own, not with a paragraph about the company. The first two lines decide whether anybody reads the rest, and a candidate is scanning for what they would actually do all day. After that: one honest paragraph on the team and how it works, the specific tools involved, a requirements list capped at around five genuine essentials, the salary or a plain explanation of why it is not shown, and the hiring process with a realistic timeline. Cut the values block and the boilerplate about dynamic environments. Write requirements as things the person must be able to do rather than as years of experience, since a year count is a proxy that quietly filters out people who can do the job. Read it aloud before publishing, because the sentences you stumble over are the ones that were assembled rather than written. Keep the ad aligned with the job description, and use job posting software so one version reaches every board instead of three drifting variants. Then check the application rate for that ad a month later against the version it replaced.

Where should employer brand content actually be published?

Own it first, then distribute where the audience already gathers. Owned means the careers page, the job ads and the team pages, because those keep working long after a campaign stops. Distribution means the places your specific candidates already read: communities attached to their craft, employee networks on the platforms they use daily, alumni of your own company, universities if you hire early-career, and any newsletter or event with the right audience. A corporate careers account posting into a general feed rarely earns attention on its own, and expecting it to is how employer brand budgets get spent with nothing to show for them. Employee networks outperform company channels for the simple reason that people trust a person. Paid amplification works for hard-to-fill roles, though only when the destination is worth arriving at, so fix the page before buying traffic to it. Everyone who engaged without applying belongs in talent pool software, since this work pays back over years rather than inside one quarter. Measure each channel by applications and hires rather than by reach, and stop the ones producing neither after two quarters.

How do you measure whether employer branding is working?

Split it into leading indicators you can move this quarter and lagging ones that confirm it worked. Leading: application rate per job view, the share of applications arriving direct and organic rather than paid, referral volume, response rate to outreach, and careers page conversion by device. Lagging: offer acceptance rate, source quality measured as pass-through by source rather than raw volume, time to fill on roles that were previously hardest, and decline reasons captured as a picklist rather than as free text nobody reads. Be honest about attribution. Employer brand is one input among many, it moves slowly, and none of these numbers isolates it from the labour market or a competitor's layoff round. What you are looking for is direction across several quarters against your own baseline, not a clean causal figure. Pull the numbers from the same source every time using your recruitment analytics software, and never report follower counts as an outcome. Set the baseline before changing anything, because the commonest failure here is an improvement nobody can evidence afterwards. Two quarters of stable measurement beats a dashboard built later.