Earlier in this blog, I released 16 chapters from A Recruitment AiDE , a discipline on effective key hire recruitment. AiDE is something I've developed throughout my career, from curiousity, from a desire to find improvement, and because through experimentation I became a better recruiter. You might say it's a copywriting framework, but recruitment is different to attracting customers. For me adverts are the consequence of the work required to make sure employers are hiring the right people for the right role, and interpreting that message with meaning for readers who might just start the path to becoming an employee. But when you've advertised the way you've always done, and sometimes it works, sometimes it doesn't, it's easy to think 'what a bunch of hot air'. Adverts work for some vacancies, not for all! You might think you'll get to too much AI slop in terms of applications, so what's the point? You might think headhunting is best. Yet what you'd miss is 'the consequence'. Because better messaging helps all recruitment, irrespective of the way you do it. And if you get it right, you'll hire more suited people who'll enjoy their roles. Fortunately little of this is speculative. In itself it's the consequence of the work that has gone before, by many experts in many arenas. Some of which I share here, in some of the evidence which validates AiDE. It's very long, and the research cited longer still, but you can read the originals and make up your own mind. Then maybe you can try AiDE and see the results. Experimenting like I have over 25 and change years. (btw this chapter may have some errors from how I copied it from the manuscript, given citations sat as a footnote) Chapter 2: The evidence Why each dimension of AiDE has a research foundation This chapter shares research that supports what I learnt, developed and deployed through AiDE, because "Well you would say that" is such a natural objection. Good recruitment advertising is not complicated in principle. Write for the people you want to attract. Be specific. Tell the truth. State the salary. Be decent. And yet the evidence tells a different story. Across decades of research into how language shapes decisions, how candidates read and respond to job advertising, and what makes a hire succeed or fail, the same gulf recurs: between what employers write and what candidates actually need. Something that may lead to failed recruitment. This chapter maps that evidence onto AiDE's four dimensions: Attention, ikigai, Definition, and Experience. The research did not produce AiDE. AiDE reflects a set of candidate-first principles developed over twenty-five years of practice. But the research independently supports those principles and understanding why each dimension matters makes it easier to apply them well. A note on sources: the evidence assembled here draws on peer-reviewed academic research, large-scale industry data, and practitioner findings from conversion copywriting - a field that works from A/B testing and measurable outcomes rather than controlled laboratory studies. Each source type has different strengths. Where practitioner literature is cited, this is noted. Where academic studies are cited, the original papers are referenced at the end. The reader is encouraged to verify independently. Attention The AIDA framework - Attention, Interest, Desire, Action - was introduced by advertising pioneer Elias St. Elmo Lewis.¹ It is one of the oldest continuously applied frameworks in commercial communication. Its longevity reflects a current truth: attention must be earned before anything else can follow. Interest, desire, and action are unreachable if the opening fails. Research into how candidates actually process job advertisements confirms how early this failure can occur. A study by Jones, Shultz, and Chapman (2006) found that when candidates were browsing rather than actively searching - the typical state of a passive reader - the presentation of an advert had a disproportionate influence on whether they would pursue it further.² When attention and interest were not secured in the opening, the quality of the remaining content became largely irrelevant: the candidate had already moved on. Further research into job advert design found that job seekers "identify a small number of early favourites on the basis of very limited information," which they use as a template for subsequent search behaviour.³ An advert that does not establish itself as an early favourite - typically in the opening sentences - does not recover its position later, regardless of what follows. The implication is direct. The standard job advert opening - company name, generic description, duty list - is not just dull. It structurally fails the passive candidate at the moment that determines whether they keep reading at all. What conversion copywriting tells us about opening in the reader's world Conversion copywriting is a practitioner discipline grounded in A/B testing and measurable outcomes rather than controlled academic studies - but no less empirically founded for that. Its central finding, consistently replicated across thousands of tests, is that copy using the reader's own language outperforms copy written from the sender's perspective. A commonly cited demonstration is a split-test run by Joanna Wiebe on a rehabilitation clinic's website. The control headline read "Your addiction ends here." The challenger was drawn verbatim from a book review on addiction: "If you think you need rehab, you do." The challenger generated over 400% more clicks on the main call to action and over 26% more form submissions on the following page.⁴ When your opening uses the exact words a reader uses to describe their own situation, it creates an immediate recognition response: this person understands me.⁵ This is the outside-in principle in measurable form: begin in the candidate's world, with something they already know to be true, and the reader's default scepticism is suspended long enough for the rest of the advert to make its case. The conversion copywriting tradition also establishes that the opening carries disproportionate weight - estimates suggest that roughly 80% of a page's conversion performance is determined before the reader has absorbed the first paragraph. In job advertising terms this is the opening sentence: the one visible in a search results summary, the one that determines the click. AiDE's Attention dimension asks a single question of every advert opening: could this appear in any other advert for any other employer? If the answer is yes, the opening has not yet done its work. ikigai The most operationally direct evidence for AiDE's ikigai dimension comes from a study by Joseph A. Schmidt and colleagues, published in the Journal of Business Psychology, reported by the Association for Psychological Science. Schmidt's team worked with a large multinational engineering consulting firm to manipulate the wording of live online job advertisements across engineering, technical design, and project administration roles. The manipulation was straightforward. Some adverts were written to emphasise what the job would offer the candidate: variety, autonomy, the opportunity for development, the chance to work across multiple areas. Others emphasised what the employer required: the qualifications expected, the tasks demanded, the standards to be met. The role was identical. Only the orientation of the language changed. The offer-focused adverts drew more applications relative to views. More significantly, they attracted applicants independently rated as higher quality across three dimensions: educational qualifications, stability of employment history, and relevant experience.⁶ The hiring managers, unaware of the research, responded accordingly - they were more likely to call the highest-rated applicants in for interview. The finding was strongest for technical roles where candidate demand was high. The researchers concluded that high-quality applicants who perceive their skills as being in demand "have the luxury to be more discriminating about the number and types of positions to which they apply" and are "more sensitive to recruitment messages" than those who feel they have fewer options.⁷ The candidates most worth attracting are precisely those who will read a requirements-heavy advert and feel nothing. The advert that describes what the work genuinely offers gives that same person a reason to keep reading. Honesty as a mechanism: what realistic job preview research shows A parallel body of research - the Realistic Job Preview (RJP) literature, developed since the 1970s and meta-analysed most recently by Earnest, Allen and Landis (2011) across 52 studies⁸ - reaches the same conclusion from a different angle. An RJP provides candidates with honest information about both the positive and negative aspects of a role, with the explicit goal of reducing turnover by enabling better self-selection. The initial assumption was that RJPs worked by lowering inflated candidate expectations. The Earnest meta-analysis found something more interesting: the primary mechanism was perceived organisational honesty.⁹ Candidates who received an honest preview did not simply have more accurate expectations - they perceived the employer as more trustworthy, and that trust was the strongest predictor of subsequent retention. The implication runs directly through the ikigai dimension. An advert that describes both the genuine opportunity and the genuine challenge of a role is not just more honest than one that hides the warts. It is more effective - because candidates who apply having seen the full picture are applying for the right reasons, and the research shows they are more likely to stay. Candidates who received an honest preview perceived the employer as more trustworthy - and that trust was the strongest predictor of subsequent retention. The passive majority LinkedIn's global workforce data identifies approximately 75% of the working population as passive, with 45% "approachable passive": employed, not actively searching, but potentially open to the right opportunity.¹⁰ The remaining 25% are active candidates: currently looking, CVs live on job boards, checking their email. These two groups require fundamentally different things from an advert. Active candidates have already decided to move. They need information. Passive candidates have not made that decision - their default is no. For an advert to convert a passive reader from "not looking" to "genuinely interested", it has to give a specific, honest, compelling reason for a contented person to reconsider their situation. The ikigai dimension is the primary mechanism for reaching passive candidates through advertising. It is where the advert articulates not what the employer needs, but what the candidate might gain - and does so with enough specificity that a content professional has a genuine reason to weigh up a change they were not already planning. The next chapter introduces 'problem awareness' as a different way of considering active vs passive candidates, and is one which sets aside the notion that situation defines candidacy. Definition In 2011, Danielle Gaucher, Justin Friesen, and Aaron Kay published a study in the Journal of Personality and Social Psychology examining whether gendered language in job advertisements affected applicant behaviour. The methodology was substantial: an archival analysis of job advertisements, followed by three controlled experiments creating masculine and feminine versions of the same advert for identical roles. The headline finding is that masculine-coded language reduced appeal to women. But the mechanism is more instructive than the result. The reduced appeal was not explained by women perceiving themselves as less qualified. It was explained by a lower sense of belongingness: whether a candidate felt they would fit, be welcomed, and belong in the culture being described.¹⁰ ¹¹ Participants did not explicitly attribute identified gendered language as the reason for their response in any of the studies.¹² They rationalised decisions on other grounds without recognising that the words themselves had done the filtering. The study has since been replicated in start-up contexts.¹³ The implications reach well beyond gender. If carefully coded language can signal who belongs and who does not - and if candidates act on those signals without knowing they are doing so - then the question any employer should ask of their advert is not only whether it excludes anyone deliberately, but what it is signalling without knowing it. This is a Definition problem as much as a language problem. The same mechanism operates whenever a requirements list is unnecessarily exclusive, whenever insider jargon narrows the reader without narrowing the role, whenever cultural shorthand communicates a kind of person rather than a set of capabilities. The definition section of an advert is not neutral. In the Gaucher framework, it is a continuous projection about who the organisation is imagining when it writes - and who belongs there. The requirements section of an advert is not neutral. It is a continuous sign of who the organisation is imagining - and who belongs there. Specificity, salary, and the cost of vagueness Research into message specificity in recruitment advertising consistently shows that specific information increases candidate perceptions of ad truthfulness, organisational attractiveness, and intention to apply.¹⁴ More specific content shows a sincere and honest employer; lower specificity reduces credibility and increases negative attitudes toward the message. On salary specifically: a 2026 analysis of 18,283 full-time UK job listings found only 13% included numeric salary details.¹⁵ Reed's 2024 analysis found roles disclosing salary received approximately 60% more applications; one sector-specific dataset showed 67% more applications and response times over 50% faster when salary was listed.¹⁶ These are commercial reports reflecting indicative trends from proprietary datasets rather than generalisable effect sizes. Four in five UK jobseekers report being less likely to apply for a role if the salary is not shown. Sixty-four per cent, according to one analysis of nearly 10,000 listings, will simply not apply unless salary information appears.¹⁷ "Competitive salary" is the most common form of this failure. It is not a salary statement. It is the absence of one, dressed in language that implies the opposite. For a senior passive candidate with options, it functions as an early indicator that the process ahead may be opaque - and that impression, once formed, is difficult to recover. Experience The Experience dimension carries the lowest weighting in AiDE because candidates tolerate process friction when they genuinely want the role. But this does not mean process design is unimportant. Certain elements in the advert itself communicate process quality before any application has been made. Research into how candidates respond to disclosed selection processes is instructive. A study published in Computers in Human Behavior (2021) found that job-seekers reported significantly lower intention to apply when advertisements described AI-based, automated selection compared to human decision-making.¹⁸ The mechanism was reduced expectation of control and fairness - a feeling that there was no human on the other side of the process who would read what they sent. This finding translates directly to the unnamed, contactless advert. "Click apply below" with no named contact and no stated timeline communicates, accurately or not, that the application will enter an automated queue. For a passive candidate with options - who has no urgent need to apply and is evaluating whether the process is worth their time - this is often sufficient reason to move on. The persuasion knowledge literature offers a complementary explanation. Research in this field shows that when readers can identify a communication as self-serving, they activate scepticism and become harder to persuade.¹⁹ An anonymous process reads as employer-serving: it withholds information that would benefit the candidate while asking for information that benefits the employer. A named contact, a stated timeline, and a commitment to reply are not merely courteous. They are evidence that the process is designed for people, not for volume. The evidence on fit - and why getting it right matters The research literature on person-organisation fit is extensive. A meta-analysis by Kristof-Brown and colleagues (2005) across hundreds of studies found that higher levels of fit were consistently associated with greater job satisfaction, stronger organisational commitment, higher engagement, and meaningfully lower turnover.²⁰ The connection to advertising is logically direct: a specific and honest advert gives candidates the information required to assess fit before applying, and filters out those for whom the role is not right. The cost of getting it wrong is substantial. The REC's Perfect Match report calculated that a poor hire at mid-manager level with a £42,000 salary could cost a business more than £132,000. The same report found that 85% of HR decision-makers admitted their organisation had made a bad hire - and a third believed it had cost them nothing.²¹ CIPD figures put the average direct cost of filling a vacancy at £6,125, rising to £19,000 for management roles; Oxford Economics' modelling put the cost of replacing a single employee at approximately £30,000 once productivity loss, training, and transition are included.²² Against these costs, the investment required to write a more specific, more honest advert - one that attracts a more considered applicant, supports better self-selection, and communicates clearly about what a candidate would be joining - is negligible. A note on when standard adverts work - and why better always helps The evidence assembled here should not be read as a claim that standard job adverts never work. They do, under certain conditions: when candidate supply is abundant, when applicants are highly motivated and will process information regardless of quality, when the role is sufficiently distinctive that even a generic description generates interest. The Schmidt data makes this explicit. For administrative roles where the candidate market was less constrained, ad wording had no measurable effect on applicant quality. Standard adverts, in those conditions, are adequate. The offer-oriented premium appears where it is most needed: in competitive talent markets, for roles requiring specific experience, when the target candidate is employed and content. Is adequacy what we need? The 2026 UK job market is characterised by high volumes of AI-generated applications from candidates applying at scale without genuine engagement, and by increasing difficulty reaching the experienced professionals most likely to succeed in senior roles. A standard advert in a competitive sector does not produce adequate results - it produces noise. There is also a rising-tide argument that applies regardless of market conditions. A better-written advert helps all three candidate types. The active candidate gets clearer information and can self-select more accurately. The unsuitable candidate gets honest insight that allows them to step away without wasting anyone's time. The passive candidate gets a reason to consider a move they were not already planning. Better writing does not disadvantage the candidates a standard advert would have reached. It simply reaches further. The Greg Evidence Since 2020, a 99.1% fill rate on retained mandates has been achieved in this practice, excluding only those vacancies cancelled due to changes in economic conditions.²³ Average retention of successful placements runs at 4.1 years, with consistent feedback they deliver above expectation and enjoy their roles. The proportion of applications that warrant a conversation consistently exceeds 30%, against an industry norm where unsuitable applications are estimated at 95% or more of the total. The adverts that produce these results are not sophisticated in any technical sense. They are specific, honest, and written from the candidate's point of view rather than the employer's. They name the salary. They describe the actual texture of the work. They say something true about why a satisfied professional might consider the move. This is, in practice, what the research predicts. The Gaucher findings on language and belongingness in the Definition dimension. The RJP research on honesty as a mechanism for retention. Schmidt's evidence that offer-oriented adverts attract measurably better candidates. The passive-candidate data on what motivates a content professional to reconsider. The salary transparency evidence on what withholding information costs. The conversion copywriting tradition on how opening in the reader's world changes response rates. The person-organisation fit literature on why precision in attraction produces better retention. The research does not prove AiDE. It establishes the conditions under which a candidate-first framework addresses real, evidenced problems. The chapters that follow make the case for the specific approach.  References Barry, T.E. & Howard, D.J. (1990). A review and critique of the hierarchy of effects in advertising. International Journal of Advertising, 9(2), 121–135. Jones, D.A., Shultz, J.W. & Chapman, D.S. (2006). Recruiting through job advertisements: The effects of cognitive elaboration on decision making. International Journal of Selection and Assessment, 14(2), 167–179. Petry, T., Treisch, C. & Peters, M. (2022). Designing job ads to stimulate the decision to apply. International Journal of Human Resource Management, 33(15), 3019–3055. The phrase "identify a small number of early favourites on the basis of very limited information" appears in Allen et al. (2007, p. 1697), cited in this study. Attributed to Wiebe, J. (CopyHackers). Rehab clinic headline split-test. Control: "Your addiction ends here." Challenger (VOC-derived): "If you think you need rehab, you do." Result: >400% increase in CTA clicks, >26% increase in form submissions. Reported in CXL (2024). cxl.com/blog/voice-of-customer/ Palmer, R. (2025). Conversion Copywriting: How to Write Copy That Turns Visitors Into Buyers. robpalmer.com . Headlines derived from VOC research consistently outperform headlines written from the marketer's vocabulary in A/B tests. Schmidt, J.A., Chapman, D.S. & Jones, D.A. (2014). Journal of Business and Psychology, 30(2). Higher-quality applicants tend to be more selective and more sensitive to recruitment messaging (Schmidt et al., 2014). Earnest, D.R., Allen, D.G. & Landis, R.S. (2011). Mechanisms linking realistic job previews with turnover: A meta-analytic path analysis. Personnel Psychology, 64(4), 865–897. Meta-analysis of 52 studies. Primary mechanism identified: perceptions of organisational honesty. Premack, S.L. & Wanous, J.P. (1985). A meta-analysis of realistic job preview experiments. Journal of Applied Psychology, 70(4), 706–719. See also Wanous, J.P. (1973). Effects of a realistic job preview on job acceptance, job attitudes, and job survival. Journal of Applied Psychology, 58(3), 327–332. LinkedIn Talent Solutions (2014). Active vs. passive candidates: The latest global breakdown revealed. Talent Trends 2014 (n=18,000 across 26 countries). Gaucher, D., Friesen, J. & Kay, A.C. (2011). Evidence that gendered wording in job advertisements exists and sustains gender inequality. Journal of Personality and Social Psychology, 101(1), 109–128. Archival analysis of job advertisements plus three controlled experiments. Reduced appeal was not explained by women perceiving themselves as less qualified. It was explained by a lower sense of belongingness — whether a candidate felt they would fit and be welcomed (Gaucher et al., 2011). Participants did not explicitly attribute their responses to gendered wording; instead, effects on attraction and application intentions were driven by implicit perceptions of belongingness and fit (Gaucher et al., 2011). Seong, M. & Parker, S. (2024), in a replication and extension of Gaucher, Friesen & Kay (2011), report that gendered wording in job advertisements can deter women from applying to start-up roles, consistent with prior evidence on belongingness effects. Roberson, Q.M., Collins, C.J. & Oreg, S. (2005). The effects of recruitment message specificity on applicant attraction to organisations. Journal of Business and Psychology, 19(3), 319–339. See also Feldman, D.C., Bearden, W.O. & Hardesty, D.M. (2006). Varying the content of job advertisements: The effects of message specificity. Journal of Advertising, 35(1), 123–141. Careerminds UK (April 2026). UK Salary Transparency Report: analysis of 18,283 full-time job listings across 27 UK cities. Only 13% of listings included numeric salary details. Reed.co.uk analysis (2024), cited in Astute People (2025). Roles disclosing salary received approximately 60% more applications. Liberty Hive dataset showed 67% more applications, response times 50% faster. D&I Leaders campaign (2024), citing Reed.co.uk research: four in five jobseekers report being less likely to apply if salary is not shown. Reward Gateway | Edenred (2026) analysis of 9,646 UK listings: 64% of jobseekers will not apply unless salary information appears. Wesche, J.S. & Sonderegger, A. (2021). Repelled at first sight? Expectations and intentions of job-seekers reading about AI selection in job advertisements. Computers in Human Behavior, 125, 106931. Eisend, M. & Tarrahi, F. (2022). Persuasion knowledge in the marketplace: A meta-analysis. Journal of Consumer Psychology, 32(1), 3–22. Kristof-Brown, A.L. et al. (2005). Consequences of individuals' fit at work: A meta-analysis. Personnel Psychology, 58(2), 281–342. See also Hoffman, B.J. & Woehr, D.J. (2006). A quantitative review of the relationship between person–organisation fit and behavioral outcomes. Journal of Vocational Behavior, 68(3), 389–399. Recruitment & Employment Confederation (REC, 2017). Perfect Match: Making the Right Hire and the Cost of Getting It Wrong. A poor hire at mid-manager level (£42,000 salary) calculated to cost over £132,000. 85% of HR decision-makers admitted a bad hire; a third believed it cost nothing. CIPD (2022) reports average recruitment costs of approximately £6,000 per hire, rising to around £19,000 for managerial roles, depending on organisation size and sector. Broader economic modelling, such as estimates from Oxford Economics, suggests that replacing an employee may cost in the region of £30,000 when productivity loss and wider disruption are included. Wyatt, G. (2020–2026). Bircham Wyatt Recruitment internal data. 99.1% fill rate across retained mandates, excluding cancellations. Average candidate retention 4.1 years. Suitability rate consistently above 30%. Okay I'm citing myself, you're welcome to contact me to challenge, and you should always be sceptical of unverified statements.