Real-time labour market information is based on analysis of the millions of online job listings posted every day by employers across the globe. A job description provides crucial insight into what employers see as key requirements for a particular role, and hiring organisations have every incentive to get job descriptions right – a bad fit or drawn-out vacancy can be costly and detrimental for business. Analysis of online postings unlocks the dynamics of the labour market and can provide answers to a multitude of questions: Which sectors are hiring, and how does demand vary across geographies? Which skills carry salary premiums? Which new roles are emerging?
How is real-time labour data collected?
Real-time data is compiled by scanning the internet on an ongoing basis, using aggregators that collect job postings from a variety of sources including job boards, company websites, and government portals. Currently, Burning Glass technology aggregates data from well over 40,000 sources worldwide, with the list expanding continuously.
Burning Glass’s Labour Insight software extracts detailed information about each job posting using patented natural language technology. It identifies and classifies dozens of different variants such as job title, occupation, employer, location, and skills required.
How do real-time data compare to other sources?
Real-time data represents an important complement to traditional survey-based labour market data, such as those produced by government agencies. Most LMI data is collected using surveys of employers, job seekers, or the general public. This data is great for tracking macroeconomic employment trends, and, due to having been conducted consistently for decades, it is a good instrument for long-term comparison and projection.
However, traditional time series compilation takes time. Fresh data is especially important in market areas experiencing rapid change, and while some established labour data reports (e.g. the Vacancy Survey) do provide a timely view on current market conditions, others (e.g. the more extensive Labour Force Survey) are based on data that is already several months old by the time of publication. It can take years for an emerging occupation to make its way into traditional SOC coding, meaning many recent employment trends go unreported in government sources. In addition, traditional reports are often based on broad job categories, and jobs within those categories are treated as identical in terms of the skills and education they require.
By contrast, not only can real-time job data provide a picture of what the market looked like as recently as 48 hours ago, but it is also much more specific, reflecting how jobs differ within and across granular sectors and geographies. Employers are constantly adjusting job descriptions to match changing marketplaces, and real-time data is able to reflect these micro economic changes much more quickly.
Overall, the question of which data sets to use is not one of ‘better’ or ‘worse’, but of big versus small picture. Both elements are needed to access comprehensive insight into the labour market.
Why does the level of detail matter?
Data is only useful if you are able to act on it. In order for that to be the case, it needs to be as specific as possible. It is one thing to know that there is high demand for computer programmers, but quite another to know how many employers are looking for JAVA experience versus Python expertise, and how that demand differs between Bristol and Edinburgh. This level of detail allows researchers, employers, recruiters and academic institutions to make detailed plans about the hiring and training of workers, and enables employees and students to make informed career decisions.
Are there jobs that are missed?
Our analysis shows that a significant majority of UK jobs are now advertised on the internet. However, it’s important to note that certain types of jobs can be underrepresented in this online pool due to being advertised in other ways, such as shop windows or word of mouth. This tends to be the case mostly with small businesses as well as jobs at the lower end of the income and skill scale. Recent research – for example, a 2015 McKinsey report on the transformative potential of digital job platforms – indicates that the online labour market has consistently expanded over the past few years.
What about jobs that are posted in multiple places?
Many employers and recruiters use multiple sites to advertise a vacancy, so it is quite common to find the same posting on more than one platform. In order to ensure that real-time figures provide an accurate picture of the current job market, removing these duplicate postings is an essential part of the data collection process. Labour Insight’s algorithms use a range of factors to identify duplicates, and around 80% of all collected postings are eliminated in this process.
Do employers really mean what they say in job postings?
Employers are generally articulate when describing what they are looking for. Although there are plenty of poorly written or generic job postings, aggregation across millions of postings shows significant consistency across companies regarding the skill and qualifications required for a given type of job. In fact, a 2016 study using US data concluded that job postings have become more skill-specific in the years following the Great Recession – a finding that is likely reflected in UK practice as well.
Nevertheless, it is important to keep in mind that a job posting is not always a comprehensive representation of all necessary requirements; employers tend to emphasise mostly what they feel they may not otherwise find. For example, a job posting for a lawyer might not specify a Bachelor’s degree or Graduate Diploma of Law, as employers often see this as a given.
In other cases, a more general qualification might serve as a proxy for specific skills required by the employer; for example, if good writing, communication, and time management are important, a posting may request degree-qualified applicants on the assumption that graduates are more likely to have those skills.