Every year, salary guides from Hays, Reed, Robert Walters, and others land in inboxes and LinkedIn feeds.
They're free, nicely-designed, and full of numbers on ‘average salaries’ – so it's natural to reach for them when you need salary data.
But the reason these guides exist and the reason HR teams need salary data are two different things.
Salary guides are published by recruitment agencies or HR organisations as content marketing – they run a survey with the employees and job seekers in their database, aggregate the results into average salaries for each role, and release it as a report.
They're useful for getting a broad sense of the market – for employees wondering whether their pay is in the right ballpark, or for managers wanting a quick read on market trends for certain roles.
What they're not designed for and should never be relied on for is compensation benchmarking – when you need accurate market insights for setting pay structures, building salary bands, or making decisions about individual salaries.
This guide explains the difference, walks through what's available for accessing salary data, and helps you match the right source to the right need.
Salary guides or salary reports are publications produced by recruitment agencies and other professional bodies.
The major ones include:
There are also industry-specific versions – like the RIBA salary guide for architects, or IT salary guides produced by tech-focused recruitment firms.
The data behind these guides varies by publisher, but typically draws on one or more of the following:
Published once a year (‘Reed salary guide 2025’, ‘Reed salary guide 2026’ – and so on) they're designed to give employers and jobseekers a broad read on the market, to be able to go into pay conversations more informed.
They're free to download, usually in exchange for an email address – because lead generation for the organisation is a primary purpose.
If you're wondering whether your salary is roughly in line with the market, or preparing for a pay conversation with your manager, a salary guide gives you a directional sense of where your role sits.
It's a reasonable starting point for setting expectations before an interview, or for understanding broad trends in your industry.
But, it’s worth understanding that your company won’t (or shouldn’t) use the data in these free salary guides and reports to inform decisions on pay for you and the rest of your colleagues – because it’s not reliable data.
Instead, they’ll be drawing on validated benchmarks from a compensation data provider – which might look very different from the data you see in an industry report.
Salary guides are useful for:
They're a useful sense-check. They're not a robust foundation for compensation decisions.
Salary benchmarking – that HR and Reward teams use to set pay structures, run compensation reviews, and make decisions about individual salaries – requires current, validated data matched to specific roles and levels.
Salary guides aren't built for that.
Which means if you rely on them as a source of compensation data, the pay decisions that follow are built on a foundation that can't be verified, defended, or trusted under pressure.
Here’s a summary of why salary guides fall short for compensation benchmarking:
Salary guides and reports are built on self-reported surveys: the agency or professional body asks respondents for their job title and salary, then aggregates the responses.
The problem starts there.
Self-reported data is prone to error – and in some cases, respondents have reason to inflate figures.
More fundamentally, these surveys are run by organisations whose core business is recruitment or professional membership, not compensation research.
There's no robust methodology behind the survey design, no controls for how respondents interpret questions, and no independent validation of what comes back.
For benchmark data to be usable, roles need to be standardised across respondents – a consistent framework that defines what a "Senior Engineer" or "Marketing Manager" actually means in terms of scope, seniority, and responsibilities, regardless of how any individual company titles that role.
Without this, it’s impossible to ensure a like-for-like comparison, which leads to inaccurate benchmarks.
Compensation benchmarking companies do this through formal job mapping: matching each role to a defined level framework before any analysis runs.
Salary guides skip this step.
Responses are typically taken at face value – if someone says they're a Senior Backend Engineer, they're counted as one.
The result is that the figures for any given title are an aggregate of potentially very different roles, levels, and scopes.
Typically, the teams behind these salary guides will take those self-reported responses and average them by role title to create an average salary per role.
There are several problems with this if you use the data for compensation benchmarking.
Firstly, averages are easily skewed by outliers – which is why the compensation industry uses medians and percentiles (25th, 50th, 75th) to give a meaningful picture of where the market actually sits.
Secondly, those percentiles also make it possible to align with your compensation philosophy on market positioning – salary guides don’t include that detail.
Thirdly, there’s also no detail on the differences in typical market rates by location, industry, company size, or funding stage – unless you look for niche salary guides (like RIBA’s for architects, for instance).
You're looking at an averaged view of "the market" that may bear little resemblance to the talent market you actually compete in.
Salary guides are usually published once a year, based on a survey run in the months prior, as an annual marketing campaign.
By the time the guide reaches you, the underlying data is anywhere from six to 12 months old – and it won't be updated until the next edition.
That means it’s not a useful view of the current compensation market. And that’s especially problematic when you’re dealing with fast-moving markets like the tech industry, where roles and pay can shift a lot within a year.
This isn’t a theoretical problem – many HR and Reward teams do rely on free, unverified sources for salary benchmarking.
Take TestGorilla, for instance. Before switching to Ravio as a reliable source of compensation benchmarks, TestGorilla’s People team were manually scraping free European reports to build company-wide salary bands – taking up huge amounts of team time, to produce results they weren’t at all confident about.
Similarly, Tiqets used to conduct compensation reviews using Glassdoor and LinkedIn data, and found that line managers didn't trust the figures enough to have productive pay conversations.
If salary guides aren’t cutting it, here's an overview of where to find salary data, what each is based on, and what it's appropriate for.
Examples: Ravio, Pave
Data source: Live data from HRIS integrations (and ATS / equity software integrations, in some cases) updated continuously.
Good for: Reliable compensation benchmarking – though coverage and benchmark methodology varies by provider.
Modern salary benchmarking tools pull compensation data directly from company HR systems and use that to build accurate benchmarks that reflect what companies are paying right now – with no risk of human error, and no outdated surveys.
Roles are matched using a standardised levelling framework, which makes the data meaningfully comparable across companies – though approaches to this mapping vary across providers.
For HR teams making regular benchmarking decisions, that combination of data freshness and methodological rigour is where the difference is.
Ravio also produces an annual Salary Trends report using its compensation dataset – a free resource just like the salary guides by Hays and Reeds, but providing a much more accurate view of the market.
Examples: Mercer, Radford (AON), Willis Towers Watson (WTW)
Data source: Employer-submitted compensation data, collected via annual or biannual surveys and statistically validated.
Good for: Salary benchmarking for large enterprises and regulated industries where the job market is relatively static – not suitable for fast-moving sectors like tech.
Traditional compensation surveys are the long-standing standard for salary benchmarking.
They collect data directly from HR teams via detailed salary surveys, apply statistical validation, and cover a wide range of roles, levels, and markets.
They're credible – but expensive to participate in, slow to update (data is typically months old by the time it reaches you), and require significant manual effort to map roles and apply to your specific context.
Today some HRIS platforms also include access to third-party compensation survey data as part of their offering – Workday and HiBob, for instance, both integrate Mercer benchmarks. This can be tempting for tool consolidation, but check how current and relevant the data is for your specific roles and markets before treating it as a primary benchmarking source.
Examples: Hays salary guide, Reed salary guides, Robert Walters salary guide, Robert Half salary guide, Glassdoor, Indeed salary calculator, Totaljobs pay tool, RIBA salary guide, HRDataHub.
Data source: Varies – placement data, job ad analysis, self-reported surveys, or job ad scraping depending on the publisher.
Good for: Directional market awareness, employee sense-checking, understanding broad trends.
Not appropriate for: Compensation benchmarking or any pay decision.
This category covers a wide range of sources, but they share the same fundamental limitation: the data isn't reliable enough to build compensation decisions on.
Agency salary guides (Hays, Reed, Robert Walters) and professional body reports (RIBA salary guide, IT salary guides) are based on self-reported surveys with no formal job mapping or validation – as covered above.
Free self-reported tools like Glassdoor, Indeed, and LinkedIn Salary have the same problem: crowdsourced, unverified figures with no levelling framework and no way to filter for your actual talent market.
It's also worth flagging a newer category of tool – platforms like HRDataHub that scrape job advertisements to produce salary benchmarks.
The problem is that advertised salary ranges are unreliable by nature: they reflect what companies are willing to say publicly, not what they're actually paying, and they're often wide ranges designed to attract candidates rather than accurate market midpoints. The same limitations on benchmark methodology also apply – no job mapping, no validation, so not a solid foundation for pay decisions.
A salary guide is a publication usually produced by a recruitment agency or professional body that provides indicative salary ranges for a range of job roles. Major UK salary guides include those from Hays, Reed, Robert Walters, and Robert Half. They're typically free to download and useful for getting a broad sense of the market – but they're not the same as validated compensation benchmarks.
For their intended purpose – general market awareness and directional sense-checking – yes. For compensation benchmarking and pay decisions, no. The data behind agency salary guides reflects the agency's placement market rather than the full workforce, updates typically once a year, and doesn't include the levelling frameworks or statistical validation that reliable benchmarking requires.
A salary guide is typically produced by a recruitment agency or professional body as content marketing, drawing on placement data or candidate surveys. A salary survey – from providers like Mercer or Radford – collects data directly from company HR departments, applies statistical validation, and is designed specifically for compensation benchmarking.
A salary report or a tool like Glassdoor will give you a ballpark, but is not a reliable source of salary data. For compensation decisions – building salary bands, running pay reviews, benchmarking individual roles – you need validated data from a compensation benchmarking provider, collected from real payroll data rather than self-reported surveys or job ads.
For most growing tech companies in Europe, real-time benchmarking software like Ravio gives you the most relevant and current data – you can search three free benchmarks to get started, or download the Compensation Trends report for a broader view of where the market is heading.
Treat salary guides as directional rather than definitive. Use them to get a broad sense of where the market is, not to anchor specific pay decisions. When reading them, check the methodology: what data source is the publisher drawing on, how many respondents, and how current is the data? Be especially cautious with figures for highly specialised roles, where small sample sizes can make averages misleading. For any decision that will affect an individual's pay or your organisation's pay structure, cross-reference against a more rigorous compensation data source.
Salary growth in 2026 is more targeted than in previous years. Strong demand and premium pay continue for specialist roles in AI, data, and cybersecurity. Generalist roles are seeing more conservative movement. Ravio's Compensation Trends report covers 2025-6 salary trends across European tech in detail.
Salary bands should be refreshed at least once a year, typically ahead of your compensation review cycle. But compensation decisions happen year-round – new hires, promotions, retention conversations, budget planning – and each of those moments needs current market data to be defensible. The teams that get compensation right tend to have continuous access to up-to-date benchmarks from a provider like Ravio, even if the formal band-setting exercise happens once a year.
Your monthly dose of market insights and expert perspectives
Join Sunny Chatterjee and Alistair Fraser for an honest conversation about benchmarking the roles that don't fit neatly into a survey – the ones where the data is thin, the peer group is unclear, and you still need to make a call.
From free salary calculators to real-time benchmarking platforms, the options are wide – and the quality varies enormously. Here's every type of salary benchmarking tool compared for 2026.
Bigger datasets tend to be broad but not always reliable. Learn how Ravio prioritises data quality over quantity to give fresh, accurate, and relevant pay benchmarks.