O
Catalyst | On Topic
Insights from AI can also be used to attract
candidates in a different way – to tailor email
marketing or adapt communication style, says
Jonathan Novich, vice president of product strategy
at recruitment software company Bullhorn: “A lot
of email recruitment marketing comes across as
conversational but candidates see it as junk mail.”
Having said that, there is no doubt that AI systems
can save a considerable amount of time. At UK-based
delivery company Yodel, for example, ‘candidate
experience bots’ deal with candidate queries, suggest
relevant roles and gather real-time feedback from
applicants. This has reduced time to hire by 38%.
AI supporters argue that using chatbots can boost
candidate engagement and therefore retention;
analysis of these interactions can even suggest
whether someone will be a good cultural fit, which
saves recruitment costs in the long term.
Jobs Intelligence Maestro
The Development Bank of Singapore (DBS Bank)
took a similar approach when it needed to hire 40%
more wealth-planning managers for its wealth-
management franchise. It’s a high-volume role, with
recruiters typically spending around a fifth of their
working day collecting information and responding
to emails before they meet candidates face to face.
In 2018, DBS created ‘Jim’, or Jobs Intelligence
Maestro, a virtual recruitment bot powered by AI.
“With Jim, we’re able to automate the pre-screening
process, which involves posting screening questions,
conducting psychometric assessments and tests,”
explains Susan Cheong, head of recruitment.
“This enables recruiters to spend more time
sourcing for candidates and interviewing them.
Candidates have told us that Jim is easy to use, and
the application process is intuitive and seamless.”
While candidate-screening time has reduced
drastically (from 32 minutes to eight), a key
“Candidates want the process to be
as frictionless as possible”
advantage has been Jim’s ability to answer 96% of
all candidate queries. Cheong has been impressed.
“This has helped us to manage workloads better and
free up employees’ time, so they can focus on more
strategic tasks and higher value work,” she adds.
Democratising hiring
A clear downside to automation is its potential to
add traffic to overloaded recruitment systems – and
increase work for hiring teams. However, Novich
points out that taking a “programmatic approach”
can help candidates de-select themselves, reducing
the volume of unsuitable applicants.
“Introducing screening questions, perhaps
followed by a personality test… it’ll become clear
through these stages if someone is not being
themselves. You see the people who really want the
job, and they’re selected in an unbiased way.”
For example, IBM’s Watson AI system has a
job-matching capability, where candidates can
engage in a discussion with a chatbot and get
job recommendations based on their skills. The
application ‘learns’ based on user feedback and
interactions to make better matches in the future.
While high-profile ‘mistakes’ made by AI systems
will continue to make headlines as they did for
Amazon, there’s no denying that the technology has,
to some extent, democratised hiring; people who
might have been overlooked in the past are now
getting interviews. In addition, AI offers benefits such
as easier candidate rediscovery, identifying ‘hidden
gems’ among past applicants.
Matching capabilities will only get more nuanced
and successful, argues Hipps from Oleeo. “In the
future, candidates will expect us to tell them where
their skills will be most recognised. AI will be able
to do the same analysis it does for applications but
pinpoint positions in the business that might suit
them better and where they have a better chance
of success.”
In that sense, it might be that AI, like a considerate
friend who knows us better than we know ourselves,
helps us find a job that suits us, even if it wasn’t the
one we applied for in the first place. The worry is, of
course, that when it comes to our closest friends, it’s
much harder to hide our weaknesses.
Issue 3 - 2019
39