Why Your Best Candidates Are Scrolling, Not Applying
Author: Niels Geominy
Affiliation: RecruitmentAds.com
Prepared for: IHRE26, London
Date: February 2026
Version: Working Paper v1.2
Access the slides of the London presentation here.
Abstract
This paper argues that modern recruitment strategy must begin earlier than the point of application. The central observation is straightforward: attention precedes consideration, and consideration precedes action. If employers want to influence candidates who are not actively job hunting, they must establish presence where attention already resides. In the United Kingdom, that environment is increasingly mobile, app-based, and feed-led.
Using Ofcom, Eurostat, IAB UK, ONS, and CBS data, together with an original Meta Ad Library proxy study comparing UK and Netherlands employment advertising patterns, this paper develops three findings. First, UK online attention is heavily concentrated in a small number of platform ecosystems. Second, advertising budgets are moving toward video-led feed environments. Third, a substantial gap appears to exist between UK and Netherlands visible employment advertising volume on Meta, raising strategic questions about whether UK employers are underusing these channels.
The conclusion is not that job boards no longer matter. It is that demand capture alone is insufficient for comprehensive talent acquisition. Employers seeking to reach candidates who are not actively searching need a demand-generation layer built around where attention actually sits, supported by feed-native creative and measurement that separates attention, intent, conversion, and quality.
1. Introduction
A candidate cannot act on an opportunity they do not notice. If they do not see it, it has no effect. If they see it but do not register it, it does not enter consideration. If they consider it but encounter friction, they are less likely to continue. This is not a marketing slogan. It is a basic sequence of human behaviour.
The problem with much recruitment strategy is that it begins too late. It starts when a person is already in job-search mode. That captures active candidates, but it misses a larger group: employed people who might move for the right opportunity but are not browsing vacancies at the moment an employer wants to reach them.
This paper examines that gap through a UK-first lens. It focuses on three questions: where attention sits in the UK digital environment, how advertising markets reflect that distribution, and what that implies for talent acquisition strategy.
2. Methodology and Evidence Base
This paper uses two forms of evidence.
The first is external primary and high-trust sources: Ofcom, Eurostat, IAB UK, the UK Office for National Statistics (ONS), and Statistics Netherlands (CBS). These sources are used to establish baseline facts about online behaviour, platform reach, advertising spend, and labour market context.
The second is internal author analysis: an original Meta Ad Library proxy study comparing UK and Netherlands employment advertising visibility. This internal analysis is clearly labelled as such. It is used to generate strategic questions and directional insights, not to make definitive claims about national ad spend, vacancy totals, or hiring outcomes.
This distinction matters. The paper’s main factual claims rest on official or high-trust sources. The internal dataset is used to test whether a visible cross-market pattern exists inside one important advertising ecosystem.
3. The UK Attention Environment
3.1 Online time as baseline reality
Ofcom’s Online Nation Report 2025 reports that UK adults spent an average of 4 hours 30 minutes online per day in May 2025. Adults aged 18 to 24 spent 6 hours 20 minutes per day. This is not niche behaviour. It is ordinary daily behaviour at population scale.
The practical significance is simple: recruitment no longer operates only in moments of active search. It operates in an environment where people are already spending several hours per day online before they ever consider changing jobs.
3.2 Mobile is the dominant environment
Ofcom also reports that 77% of online time is spent on smartphones. This makes the dominant recruitment environment mobile by default. In practical terms, the average first touch is more likely to happen in a feed, on a phone, in a moment of partial attention, than in a deliberate desktop search session.
That shift matters because the medium shapes the behaviour. Mobile environments reward brevity, immediacy, and low-friction action. Recruitment systems built only around long-form, desktop-style intent flows are increasingly misaligned with where attention begins.
3.3 Attention is concentrated
The most important structural finding is not just how much time people spend online, but where that time is concentrated. Ofcom reports that in May 2025, Alphabet and Meta together accounted for 51% of all UK online time, equivalent to 2 hours 18 minutes per day.
Ofcom also reports that YouTube reached 94% of UK online adults in May 2025 and averaged 51 minutes per day among users, excluding viewing on the TV set. The same report notes that the combined Facebook and Messenger service remained Meta’s most widely used service.
This concentration matters because it turns “channel choice” into a visibility question. Employers that choose not to operate in these environments are not merely selecting a different media mix. They are reducing their presence in the places where a large share of daily attention now sits.
3.4 App reach and everyday behaviour
Ofcom’s smartphone app reach data helps make this more concrete. In May 2025, the top smartphone apps among UK online adults included WhatsApp at 92% reach, Facebook at 85%, Google Maps at 77%, YouTube at 72%, Instagram at 67%, and TikTok at 46% overall, with TikTok reaching 65% among 18 to 34-year-olds.
These are not “recruitment platforms” in the conventional sense. They are attention platforms. That is precisely why they matter for recruitment. They describe the digital environments people enter before they are in job-search mode.
4. The European Context
The broader European context reinforces the same pattern. Eurostat reports that in 2024, 97% of people aged 16 to 29 in the EU used the internet daily, and 88% of that same group participated in social networks. Eurostat also notes that the comparable social network participation rate for the total population was 65%.
The point is not that the UK and the EU are identical in every platform behaviour. The point is that emerging talent across Europe lives in an environment where online participation is nearly universal and social network usage is normal. The relevant strategic distinction is therefore not “online versus offline,” but “visible versus invisible” within the digital spaces where attention already exists.
5. Advertising Market Signals
Attention is one market signal. Advertising spend is another.
IAB UK reports that UK digital ad spend reached £35.5 billion in 2024, up 13% year on year. More significantly for recruitment, video display reached £8.3 billion, up 20% year on year, and 64% of all online display spend is now invested in video.
This matters because it shows where the wider market is putting money. Advertisers are increasingly buying feed-based, mobile-native, video-friendly environments. Recruitment advertising operates in these same auctions, competing against consumer brands, creators, and entertainment content for the same limited moments of attention.
The strategic implication is difficult but clear: many recruitment teams are still using creative formats that were designed for vacancy communication while competing in environments optimized for thumb-stopping video. In that sense, the problem is not only media allocation. It is environmental mismatch.
6. Original Research: UK vs Netherlands Meta Ad Library Proxy Study
6.1 Research design
To investigate whether feed-based recruitment behaviour might differ across markets, I conducted a Meta Ad Library proxy study.
The study used:
- Platform: Meta Ad Library
- Filter: Employment ads
- Search terms: exact phrase matching, UK = “apply”, Netherlands = “solliciteer”
- Period: 13 rolling monthly windows, from 15 January 2025 to 15 February 2026
- Metrics captured: total results (all media), video-only results, and resulting video share
The study is intentionally narrow. It is not a census of recruitment spend, unique vacancies, or hiring outcomes. Keyword choice affects the results. The findings should therefore be interpreted directionally.
6.2 Results
Across the 13 monthly windows, the totals were:
| Metric | UK | Netherlands |
|---|---|---|
| All media (13-month total) | 115,800 | 444,000 |
| Video only | 32,480 | 49,000 |
| Video share | 28.1% | 11.0% |
Using rough population estimates of approximately 69.5 million (UK) and 18.2 million (Netherlands), this implies:
- UK: ~1,666 results per million population
- Netherlands: ~24,396 results per million population
On that basis, the Netherlands shows approximately 14.6 times more Meta employment ads per capita than the UK in this proxy dataset.
6.3 Vacancy sanity check
A reasonable challenge is that the Netherlands may simply have more open vacancies per capita. Official statistics do show a tighter vacancy position in the Dutch labour market, but not at a scale that appears sufficient to explain the size of the Meta proxy gap.
The UK ONS reported 729,000 vacancies in September to November 2025. CBS’s seasonally adjusted vacancy series for the Netherlands shows 380.2 thousand vacancies in Q4 2025.
These two national series are not a perfectly harmonized international comparison and should not be treated as one. But as a rough sense check, they point in the same direction: the Dutch vacancy rate per capita appears materially higher than the UK’s, yet still far below the 14.6x per-capita differential in the Meta Ad Library proxy. That suggests vacancy levels alone are unlikely to explain the entire observed gap.
6.4 Interpretation
The cautious conclusion is not that the Netherlands is “better” at recruitment. That would overstate what this dataset can support.
The defensible conclusion is narrower: within one major attention ecosystem, the Netherlands appears to use Meta employment advertising at substantially higher visible per-capita volume than the UK.
Several explanations are plausible:
- UK employers may still concentrate more heavily on established job boards, search, LinkedIn, or public-sector channels.
- The Dutch market may have adopted feed-based recruitment formats earlier or more broadly.
- Sector mix may differ in ways that favour Meta usage.
The strategic question is not whether the Netherlands is “ahead.” It is whether UK employers may be underusing an important attention environment.
7. Strategic Implications
The evidence supports a distinction between demand capture and demand generation.
Demand capture includes job boards, search, and other channels that work when a person is already looking. These remain important.
Demand generation begins earlier. It is concerned with entering the candidate’s field of view before they are actively searching. In practice, that means operating in the environments where attention already exists and where familiarity can build over time.
The main strategic implication of this paper is that demand capture alone is incomplete. Employers who rely only on channels that activate after job-search intent appears are likely missing part of the candidate journey, especially for passive or high-quality candidates.
8. A Simple Measurement Framework
A common failure mode in social or feed-based recruitment is to measure only final outcomes, such as hires, without diagnosing the stages in between.
A practical framework is to separate four stages:
| Stage | Question | Example indicator |
|---|---|---|
| Attention | Did they stop? | Viewability, video starts |
| Intent | Did they engage? | Landing page visits, time on site |
| Conversion | Did they act? | Application starts, completion rate |
| Quality | Did it matter? | Interview rate, quality of hire |
This does not claim to be a formal academic model. It is a practical diagnostic structure.
Its value is straightforward: “social did not work” is not a useful conclusion unless it is clear where the process broke. Weak attention, weak intent, weak conversion, and weak quality are different problems and require different fixes.
9. AI as Production Economics
AI matters in this context for one practical reason: it changes the economics of producing creative variation.
Historically, video production has been slow and expensive, which limited how many variants recruitment teams could test. In feed-based environments, that matters because one static or generic asset is often not enough to compete for attention against better-funded consumer advertisers.
The operational change is not that AI makes perfect creative. It does not. The change is that AI can lower the cost of generating enough options to test, compare, and improve. In practice, this shifts the main bottleneck from production capacity toward strategic judgment: which messages, hooks, formats, and proof points actually earn attention and produce quality outcomes.
10. Conclusion
The central conclusion is simple.
The best candidates are often not actively applying when employers want to reach them. They are employed, distracted, and spending their time inside a small number of digital ecosystems. In the UK, Ofcom shows that a large share of attention sits within a few platform families, and IAB UK shows that advertising budgets are moving in the same direction, especially into video-led environments.
That means the modern recruitment problem begins earlier than the point of application. The relevant question is not only “how do we get more applications?” It is also “how do we get seen, remembered, and considered before a person decides to search?”
Job boards still matter. Search still matters. But they are demand-capture tools. Employers that want to influence people who are not actively searching need a demand-generation layer built around the way attention now works: mobile-first, feed-led, video-shaped, and measured in stages.
Or more simply: your best candidates are not ignoring you. They are busy watching something else.
References
Official and statistical sources
Ofcom. Online Nation Report 2025. Published 10 December 2025.
Eurostat. Young people – digital world (Statistics Explained; data extracted May 2025) and 97% of young people in the EU use the internet daily (news release, 15 July 2025).
IAB UK. Digital Adspend 2024: UK’s overall digital ad market hits £35.5bn. Posted 29 April 2025.
Office for National Statistics. Vacancies and jobs in the UK: December 2025. Published 16 December 2025.
Statistics Netherlands (CBS). Vacancies; seasonally adjusted. StatLine table, updated 30 January 2026.
Internal author analysis
Geominy, N. Meta Ad Library Proxy Study: UK vs Netherlands Employment Advertising (2025 to 2026). Employment ads filter, exact CTA keyword search (“apply” in the UK; “solliciteer” in the Netherlands), 13 rolling monthly windows, manual result-count extraction. Directional proxy, not a market census.
