Data Science-related roles are among the most talked about jobs in the market these days, fuelled by both buzz around the industry and growing demand from a wide range of companies.
Here’s what 812 recent job postings tell us about what employers in Singapore want, and what job seekers should expect.
BACKGROUND
I scraped the Government’s Jobs bank, MyCareersFuture on April 27, 2019 as part of a class assignment. Here’s the link to my repo for the project, but do note that some links could have expired, as job postings have a limited shelf life.
I whittled down the dataset to 812 job entries after removing outliers and odd roles picked up during web-scraping, which is not always precise. Several companies also posted annual salary figures, which I converted to monthly renumeration instead.
SUMMARY
In a nutshell, employers are predominantly looking for experienced data science professionals who also understand the business needs and can lead a team. These senior roles come with higher pay, particularly in the banking and IT industries.
Experienced data professionals – especially senior data scientists – stand to gain the most from the current job market outlook, while those entering the industry with little or no experience should temper their expectations. Entry level and junior data professionals can expect a monthly pay of under S$5,000.
The median pay for those taking up senior data roles is above S$8,000, though there are companies which are willing to pay more for top talent. Let’s take a closer look at the key trends.
I. WHERE ARE THE JOBS (AND HIGH PAY):
1. Median monthly salary: S$ 6,750
Some employers looking to fill senior roles are prepared to pay over S$20,000 in some instances. But the median monthly salary for data professionals is a more modest S$6,750. About a third (35%) of the average salaries being offered by companies fall between S$4,50 and S$7,500.
2. Banks pay best, most data jobs in IT sector
Data jobs in the banking and finance sectors offer the best median salaries, while most of the openings (40%) are in the IT sector (which also offers competitive salaries).
Among the companies hiring for data roles, DBS Bank offers the highest median salaries, followed closely by accounting firm Ernst & Young and Traveloka, an Indonesian travel booking site.
If you are fussy about where you work, well, you’ll be happy to know that popular office locations in the CBD and Marina Bay are home to some of the best paying data jobs.
Those with the right skills and are prepared to travel a little further out west to Science Park Drive will find two high-paying jobs from Merck, and Johnson & Johnson on offer.
II. WHAT ARE EMPLOYERS LOOKING FOR:
1. Wanted: Data Scientists, Data Engineers
These are the two most common job titles in the 812 job postings, though they account for just about 10% of the postings. This is mostly due to the messy and myriad ways in which job titles are used to describe the same role in the industry.
2. Experience, business knowledge, management skills
"Data" was the top word used in how companies described the responsibilities of potential hires.
But below that, the top key words used were: Business, team work (showed up as two separate words), management, solutions, teams etc. This is consistent with informal feedback from those already in the industry – that companies don’t just want data experts, but also those who can translate their expertise to deal with business needs.
And when it comes to job requirements, employers used the word "experience" most often, followed by terms like "skills" and "knowledge".
This could be a potential stumbling block for newcomers, though "experience" can be interpreted somewhat widely in an industry that sees substantial crossovers from those who are not native to the tech or STEM sector.
3. Technical requirements: Python, SQL, computer science background, engineering skills
Surprisingly, terms used to describe technical skills are not in the list of top 10 requirements. They only appear in the top 20 list of key words used in job requirements for data science roles. In my dataset, "engineering" is the top technical skills term, followed by "Python", "analytics", "computer science" and "SQL".
Terms referencing more advanced skills, such as "machine learning" and "big data", appear further down the list though still within the top 50 most used words in descriptions of job requirements. "Deep learning" is not in the top 50 list of key words, reflecting its relative niche status in the Singapore job market.
WHAT JOB SEEKERS SHOULD KNOW:
1. Experience is your main currency
It is obvious that those with more experience will command higher pay. But it is always good to have visual confirmation and a sense of what the pay gap looks like between experienced and inexperienced hires:
2. Good job mobility for those "in the middle"
Data professionals who are in the middle of the pack – some years of experience but not yet considered a veteran – will be cheered by the fact that most openings (almost 74%) are for mid-level positions. This will give junior data professionals the opportunities to move up, and for those already in the middle category to move laterally across different sectors, or perhaps move up to a senior role.
3. Manage your expectations if you are a newcomer
The two charts above show why newcomers need to be realistic about their job hunt and salary expectations. Of the 812 data job openings, only 90, or 11%, are for junior or entry-level positions. The median salary is also the lowest, naturally, under S$5,000 a month.
4. Internships and temporary work arrangements are rare
Data professionals are expected to work with substantial amounts of sensitive data. This is one possible reason why the vast majority of job openings are for permanent roles, with very few opportunities for internships or temporary jobs. But it could well be that companies don’t officially advertise for these roles as much.
Here are the links to the dataset, and my Jupyter notebook for these charts.
For the project, I also had to build a bunch of machine learning models to try to predict for high salaries, as well as the seniority of the roles offered in job postings.
The exercise was good for honing my skills in building and evaluating machine learning models, though I don’t think the dataset is big or granular enough to generate major insights far beyond what we can discern from these charts. But it will be interesting to apply these models to a larger dataset, if I get my hands on one in the future.
Meanwhile, good luck on the job search!
Source links:
