Back

AI Tech Company in Melbourne – What inspires us!

‘What engages us’ is curated by the PredictiveHire team, a team of pioneers working at the frontier of 3 huge trends:

1. AI in HR, especially people selection. Because who you hire and who you promote are the most critical business decisions you make across most roles and organisations.
2. Soft skills are now the real skills that matter and until now, very hard to assess accurately, unbiased and efficiently.
3. Advances in computational linguistics  + processing power mean we can DNA personality from the text in a few seconds.

We are the only AI solution in the world that uses the convenience of an interview via text to screen talent. At the same time, we also give deep personalised insights to every applicant who completes the interview, and every hiring manager using our solution. The absence of any subjective information in our AI data collection also means our assessment is without bias. At last technology that truly does level the playing field.

Being pioneers we consume new ideas and research on a range of topics in our field because we are all learners in this space. Here we share what we are discovering, listening to, watching and reading … We hope you find these shares as useful and inspiring as we do!

OUR FAVOURITE BOOKS!

Ethical Algorithm
Michael Kearns and Aaron Roth

Why we love it! Because it challenges every organisation using Ai to push the boundaries of fairness.
Everybody Lies
Seth Stephens-Davidowitz

Why we love it! Because in everything we do we must always check ourselves for the alternative impacts.

Dataclysm
Christian Rudder

Why we love it! Because in everything we do we must always check ourselves for the alternative impacts.

Civilized to Death
Christopher Ryan

Why we love it! Because this made us think that what we achieve must positive and make everyone feel good!

Prediction Machines
Ajay Agrawal, Joshua Gans, Avi Goldfarb

Why we love it! Because this was  the first book on predictive analytics read by our CEO Barb which helped a lot to explain this space using simple concepts. How Smart Machines Think
Sean Gerrish

Why we love it! Because this was recommended by Matt, one of our awesome advisors.

Invisible Women: Data bias in a world designed for Men
Caroline Criado Perez

Why we love it! Whilst the audio version does feel a bit didactic at times, the narrator is so frustrated at the disconnect between the facts and what people believe about the presence or not of bias. There is some solid data referenced which reflects the deep and wide research  that has gone into uncovering often invisible nature of gender bias in many sectors.

 

NOW FOR OUR FAVOURITE PODCASTS

PODCAST #1
Michael Kearns: Algorithmic Fairness, Bias, Privacy, and Ethics in Machine Learning

Michael Kearns is a professor at University of Pennsylvania and a co-author of the new book Ethical Algorithm that is the focus of much of our conversation, including algorithmic fairness, bias, privacy, and ethics in general. But, that is just one of many fields that Michael is a world-class researcher in, some of which we touch on quickly including learning theory or theoretical foundations of machine learning, game theory, algorithmic trading, quantitative finance, computational social science, and more. This conversation is part of the Artificial Intelligence podcast. If you would like to get more information about this podcast go to https://lexfridman.com/ai(37 kB)

Why we recommend it? Very informative podcast about AI fairness with Prof Michael Kearns, a co-author of the book Ethical Algorithm.Buddhi is a regular consumer of Lex Fridmans podcasts  – he attracts an extraordinary array of minds and perspectives  from Daniel Kaheman, Melanie Mitchell, Paul Krugman, Elon Musk and he asks such thoughtful original  questions of people interviewed many times over that every podcast feels illuminating for both sides. 

PODCAST #2
Scott Adams: Avoiding Loserthink

Dilbert creator and author Scott Adams shares cognitive tools and tricks we can use to think better, expand our perspective, and avoid slumping into “loserthink.”(103 kB)
https://149366099.v2.pressablecdn.com/wp-content/uploads/2019/11/s-adams-500px.jpg

Why we recommend it? There is a story of “bias” in how he got into creating Dilbert. He was told by two employers that “we can’t promote you because you are white, because we have been promoting too many of them, so now we have to fix it”. Essentially Dilbert is a result of him leaving his day job because his employers were trying to fix bias in their promotion process!

PODCAST #3
Getting to scale with artificial intelligence – The McKinsey Podcast

Why we recommend it? Companies adopting AI across the organization are investing as much in people and processes as in technology.

PODCAST #4
Sleepwalkers podcast by iHeartRadio

Why we recommend it? With secret labs and expert guests, Sleepwalkers explores the thrill of the AI revolution hands-on, to see how we can stay in control of our future.

PODCAST #5
HBR IdeaCast: A New Way to Combat Bias at Work on Apple Podcasts

Show HBR IdeaCast, Ep A New Way to Combat Bias at Work – 14 Jan 2020(76 kB

Why we recommend it? A brilliant captivating podcast on the types of biases that turn up at work and an exploration of bias interrupters. Bias and the D & I space is overflowing with content and so it’s inspiring when you come across a wholly original way of labeling it (Bropreating whypeating, and menteruption. What’s less effective -single-bias training … -referral hiring ! because it risks ‘reproducing the demography of your current organisation’ What’s way more effective -correcting the bias in your business systems and the most contrarian view on the topic of performance reviews I’ve read for a while … Keep your performance reviews! Removing them creates a ‘petri dish for bias’.

PODCAST #6
Can Artificial Intelligence Be Smarter Than a Human Being? by Crazy/Genius

Why we recommend it? Surely, AI technology has nothing that even closely resembles human imagination. Or does it? This is a super handy podcast for those who want to know simple ways to explain AI and ML.

PODCAST #7
AI in B2B – a16z Podcast

Why we recommend it? Consumer software may have adopted and incorporated AI ahead of enterprise software, where the data is more proprietary, and the market is a few thousand companies not hundreds of millions of smartphone users. But recently AI has found its way into B2B, and it is rapidly transforming how we work and the software we use, across all industries and organizational functions.

Brilliant articulation of why FOMO is real .. as far as coming to data too late . Co pilot and auto pilot analogy is clever.
1. B2B is different. Companies care a lot about their data
2. Share for greater good and reap the benefits should be the motto of A.I. companies
3. Product design thinking with AutoPilot and CoPilot metaphors. Where can our A.I. be auto and co?
4. Use AB testing to show the benefits to the skeptics

 

OUR FAVOURITE ARTICLES

ARTICLE #1:
Chief people officer: The worst best job in tech
https://www.protocol.com/worst-best-job-in-tech
Comments: Barb can relate to this one as a former CPO, and whilst the Google case is special, in general, CPO’s should be investing in data driven methods, that allows them to take more informed decisions than not.

ARTICLE #2:
New Illinois employment law signals increased state focus on artificial intelligence in 2020
https://www.technologylawdispatch.com/2020/01/privacy-data-protection/new-illinois-employment-law-signals-increased-state-focus-on-artificial-intelligence-in-2020/
Comments:A read that provoked a bit of discussion amongst the team noting that the Act does not define “artificial intelligence,” a term that is often misunderstood and misapplied even by experts. How will they separate what traditional statistical analysis has been doing to what modern ML algorithms do. Any attempt to classify ML as something different to just statistical analysis at scale will be fun to watch. One can then argue just using averages and medians are a form of AI … Regression .. Correlations … AI bias …

Ask BERT to fill in the missing pronoun in the sentence, “The doctor got into ____ car,” and the A.I. will answer, “his” not “her.” Feed GPT-2 the prompt, “My sister really liked the color of her dress. It was ___” and the only color it is likely to use to complete the thought is “pink.”

ARTICLE #3:
A.I. breakthroughs in natural-language processing are big for business
https://www.google.com/amp/s/fortune.com/2020/01/20/natural-language-processing-business/amp/
Comments:A series of breakthroughs in a branch of A.I. called natural language processing is sparking the rapid development of revolutionary new products.

ARTICLE #4:
Are We Overly Infatuated With Deep Learning?
https://www-forbes-com.cdn.ampproject.org/c/s/www.forbes.com/sites/cognitiveworld/2019/12/26/are-we-overly-infatuated-with-deep-learning/amp/
Comments:Even Geoff Hinton, the “Einstein of deep learning” is starting to rethink core elements of deep learning and its limitations.

ARTICLE #5:
Artificial intelligence will help determine if you get your next job

https://www.vox.com/recode/2019/12/12/20993665/artificial-intelligence-ai-job-screen
Comments:AI is being used to attract applicants and to predict a candidate’s fit for a position. But is it up to the task?

ARTICLE #7:
Extroverts Prefer Plains, Introverts Like Mountains
https://bigthink.com/topography-and-personality
Causation or just correlation? There’s a very curious link between topography and personality.

ARTICLE #8:
So what is the difference between AI, ML and Deep Learning?
https://www.linkedin.com/pulse/so-what-difference-between-ai-ml-deep-learning-kanishka-mohaia

ARTICLE #9:
Attractive People Get Unfair Advantages at Work. AI Can Help.
https://hbr.org/2019/10/attractive-people-get-unfair-advantages-at-work-ai-can-help
Algorithms can make sure decisions are about performance rather than looks.

ARTICLE #10:
Artificial Intelligence in HR: a No-brainer
https://www.academia.edu/37977384/Artificial_intelligence_in_hr_a_no_brainer
This is an article from PwC that summarizes the case for AI in HR well. A really good overview.

ARTICLE #11:
Science Behind the IBM’s Personality Service
https://cloud.ibm.com/docs/services/personality-insights?topic=personality-insights-science
The background and the approach listed here is applicable to our approach too. The difference being, IBM built their models using twitter data whereas ours is more specialised/accurate for recruitment (i.e. based on more data and continuously learning). In addition, we are able to predict more than personality (e.g. job hopping attitude, traits etc).

ARTICLE #12:
Using Linguistic Cues for the Automatic Recognition of Personality in Conversation and Text 

https://www.aaai.org/Papers/JAIR/Vol30/JAIR-3012.pdf

ARTICLE #13:
Language-based personality: a new approach to personality in a digital world

ARTICLE #14:
Navigating Uncharted Waters: A roadmap to responsible innovation with AI in financial services

https://www.weforum.org/reports/navigating-uncharted-waters-a-roadmap-to-responsible-innovation-with-ai-in-financial-services
Navigating Uncharted Waters shows how financial services firms, policymakers and regulators and customers can overcome five risks and plot a course toward more rapid AI adoption.

ARTICLE #15:
Model Tuning and the Bias-Variance Tradeoff
http://www.r2d3.us/visual-intro-to-machine-learning-part-2/
Learn about bias and variance in our second animated data visualization.

ARTICLE #16:
Daniel Kahneman’s Strategy for How Your Firm Can Think Smarter
https://knowledge.wharton.upenn.edu/article/nobel-winner-daniel-kahnemans-strategy-firm-can-think-smarter/
The research is unequivocal, according to the father of behavioral economics: When it comes to decision-making, algorithms are superior to people.

ARTICLE #17:
Experience Doesn’t Predict a New Hire’s Success

https://hbr.org/2019/09/experience-doesnt-predict-a-new-hires-success
Is it time to rethink the way we assess job applicants?

ARTICLE #18:
So what is the difference between AI, ML and Deep Learning?
https://www.linkedin.com/pulse/so-what-difference-between-ai-ml-deep-learning-kanishka-mohaia/
The best ie simplest summation of this tech I have read (edited) linkedin.com. Once the domain of Sci-Fi geeks and film script writers, Artificial Intelligence or A.I.

ARTICLE #19:
Nudge management: applying behavioural science to increase knowledge worker productivit
y
https://jorgdesign.springeropen.com/articles/10.1186/s41469-017-0014-1
Knowledge worker productivity is essential for competitive strength in the digital century. Small interventions based on insights from behavioural science makes it possible for knowledge workers to be more productive. In this point of view article, we outline and discuss a new management style which we label nudge management. Nudge is a concept in behavioral sciencepolitical theory and behavioral economics which proposes positive reinforcement and indirect suggestions as ways to influence the behavior and decision making of groups or individuals. Nudging contrasts with other ways to achieve compliance, such as educationlegislation or enforcement.

We liked reading this because it mirrored what we read from candidates every day after their receive ‘MyInsights, their personalised insights profile. We believe that every person regardless  of their role craves  personal growth. The feeling they have when they receive that report- priceless for our team. “Thank you for your email. I did find it useful as it has made me really think about my workplace and personal life by self-reflecting. I feel since reading this, I have stepped up in a few different situations including at work where I had stepped up in a temporary leadership role. Personally, I have been practising speaking my mind and let go of toxic friendships and make decisions more easily.”And … After getting the insight of what you see of me & your reasoning it made me think about work place moments & how well I’ve responded to situations as well as make me think about alternative ways I could have reacted & received differing outcomes.

ARTICLE #20:
Distilling BERT models with spaCy
https://towardsdatascience.com/distilling-bert-models-with-spacy-277c7edc426c
Transfer learning is one of the most impactful recent breakthroughs in Natural Language Processing. Less than a year after its release.

ARTICLE #21:
Building Trust in Machine Learning Models (using LIME in Python)
https://www.analyticsvidhya.com/blog/2017/06/building-trust-in-machine-learning-models/
This article helps us understand working of machine learning algorithms using LIME package. Using LIME, you can understand working of black box ML models.

ARTICLE #22:
Jordan Peterson on Workplace Performance, Politics & Faulty Myers-Briggs

Hilarious watching Jordan talking about selling personality assessments but mostly he is spot on in his observations.

ARTICLE #23:
Kai-Fu Lee: AI Superpowers – China and Silicon Valley | Artificial Intelligence (AI) Podcast

Some really valuable insights in how AI is approached in the Sillicon Valley and China. Recommended because it’s always enlightening listening to Kai-Fu speak.


Blog

Pro tips for high volume hiring

Volume hiring on a tight timeline can strike fear into even the most experienced recruiter! More often than not the fallout of failing to hire enough people causes real pain to the business, managers, and you. 

So, how can you tackle high volume hiring and get better and better each time? 

Here are our pro tips.  

What is high volume hiring?

High volume hiring is recruiting for many positions (50 or more) concurrently or in a very limited period of time. Often the 50+ roles will be of the same job type. It also implies high volumes of applicants coming through for recruiter’s review. 

Volume hiring in recruitment is common in retail and hospitality, where many people have to be hired quickly for busy periods, events, and new store or restaurant openings. Graduate recruitment in large organisations is often high volume recruitment, as is hiring for nurses, other health workers and call centre staff.

During C-19 we saw the emergence of surge hiring – again, another form of high-volume where thousands of people are needed in-store or in the contact-centre within days.

Tips to Solve High Volume Recruiting Challenges

High volume recruiting challenges to overcome

Apart from the sheer logistical challenges, there are five major high volume recruiting challenges organisations face.

1. Time invested in high-volume hiring

Roles filled by high volume recruitment often have a highly sensitive empty chair impact. A restaurant with too few servers, a shop with too few attendants, a call centre with too few people answering the phone, or a hospital ward with too few nurses, both are nightmarish scenarios. 

In a perfect world, recruitment requirements can be anticipated and planned for, but that’s not always the case. That’s why a scaleable, repeatable high volume recruiting strategy is essential.  

2. Cost invested in high-volume hiring

When you’re attracting, screening, selecting and hiring 250+ people at once, it’s not just timelines that can blow out.

The cost can easily go over budget too. This is where scalable processes, talent pooling and technology tools are your friends.

3. Poor candidate experience during high-volume hiring drives

In many industries that need high volume recruitment, your candidate is also your customer. Often people are applying for a position because they love your brand. If they have a bad experience, you’ll not only probably lose them as a customer, they’ll tell their friends and family too. 

Getting the candidate experience right at scale isn’t easy, but it’s essential. Otherwise, your marketing department will be asking some serious questions, and you’ll find it much harder to find good applicants in future. 

4. Communicating employee value proposition (EVP) during a high-volume recruitment drive

Speaking of reputation, your employer brand and employee value proposition play a huge role in attracting the right candidates. You may also find that candidates in certain industries (like retail and hospitality) are easily swayed to join a competitor who tells a better story. 

Sometimes a candidate’s decision whether or not to take a role is related to their hourly rate. But more and more often, candidates want to work for a company that aligns with their values and offers learning and development opportunities. Make sure you articulate your EVP well. Your competitors will be using their EVP to try and snaffle your candidates.  

5. Balancing diversity in your hiring pool across high-volume roles

Hiring for diversity when you’re under time and cost pressure can feel overwhelming. But it’s essential that you embed diverse hiring practices in all of the hiring you do. Building a diverse team will result in better decision making, better customer service and a healthier bottom line


Volume hiring strategies

Now you know the major high volume recruiting challenges, it’s time to put together the right volume hiring strategies to help you overcome the challenges, and attract and hire the best people.

Bulk hiring techniques have come a long way over the years, from Applicant Tracking Systems scanning and scoring CVs, to the explosion of recruitment Ai now available. Let’s take a look at the volume hiring best practices you can use to make each stage of the bulk recruitment process scaleable, fast and fair. 

Six major milestones of the bulk recruitment process 

There are six major milestones in the bulk-hiring process. Discover, engage, assess, interview, decide and validate. Each stage is equally important, and most stages of the bulk-hiring process can be streamlined so that they’re highly scalable. (The Interview and Decide stages are the most time and resource-intensive, but they’re well worth the investment.)

1. Discover 

Ensuring the right potential applicants find you is the first step in getting volume hiring in recruitment right. 

Remember:

  • Lean into your Applicant Tracking System (ATS). Spend your time writing a great ad highlighting your EVP and let the ATS do the heavy lifting of shipping to multiple job boards. 
  • Think about how applicants from underrepresented backgrounds can find your ad, and make it clear everyone’s welcome. 
  • For retail and hospitality, don’t forget walk-in applicants. Check if you can use a ‘kiosk mode’ or similar with your ATS so applicants can fill in their details on an iPad rather than having paper applications pile up on manager’s desks (and get lost!).
  • Check previous applicant pools and ask for employee referrals. 

Measure: 

  • Performance of each advertising channel (ideally by how many successful candidates the channel attracts)
  • The diversity of your applicant pool

Pro tip:

  • People want to know what it’s like to work at your organisation. Ideally, have a video on the ad with people in a similar role explaining what it’s like. If you’re in a hurry – include quotes from an employee or two.   

2. Engage

Once you’ve got an applicant’s attention, you need to make sure they stay interested. 

Remember:

  • Applicants are applying for multiple positions, and the organisation who delivers the best candidate experience wins. Make communications look as 1:1 as possible.

Measure:

  • Application completion rate. This will tell you if the process is working, or if there’s something putting potential applicants off. This could be the length of the form, a confusing requirement, or even a technical glitch.

Pro tip:

  • Put some character into your application received responders. Write as you talk rather than like a bureaucrat. And don’t say: we can’t get back to everyone if you don’t hear from us you’ve been unsuccessful (or similar). If you expect candidates to put energy into applying, put energy into replying. 

3. Assess 

Now you’ve got a pool of candidates; you need to assess them. 

Remember:

  • Sadly, CVs have proven themselves to not be a good way to assess future performance, and they only reinforce biases. This is an opportunity to disrupt the usual bulk-hiring techniques with something that delights candidates and hiring managers.

Measure:

  • Candidate satisfaction. This will tell you how candidates find the experience. It’s is a good indicator that offer acceptance should be healthy, and that you won’t lose customers who are candidates. Some recruiting platforms offer candidate satisfaction surveys, or you can choose to use your employee engagement platform. 

Pro tip:

  • We created Smart Interviewer, our conversational chat technology so that every candidate could have an interview. Not only do you get detailed responses to questions, but the answers also reveal more about the candidate’s personality than any CV ever could. Using natural language processing, we’re able to build an accurate personality profile. Every single candidate receives automated, personalised feedback, and they love it. One supermarket client, Iceland, interviewed 50,000 candidates and received a 100% candidate satisfaction score. 

4. Interview

Once you have the results of Ai chat assessments, you’ll want to interview the candidates whose scores and profiles appear to match your requirements.

Remember:

  • Have a diverse selection panel (especially if you have a diverse talent pool).
  • Be consistent in how you interview and assess each candidate. Especially in group interviews, don’t be tempted to hire extroverts. You need a mix of personalities to build a successful team.  

Measure:

  • Attendance. If there’s a significant drop-off, look into why.

Pro tip: 

  • We created TalentInsights so you can easily see each candidate’s score and psychometric profile informed by their Ai chat responses before you speak with them. We designed our LiveInterview platform to make collecting and recording consistent data easy, so you can ensure everyone gets a fair go (and you don’t have to sort through impossible to interpret notes after your meetings).

5. Decide

Now you’ve got a list of fantastic candidates, you’ve met them, and you’re ready to invite some of them to join you. 

Remember:

  • Now is not the time to fall back on ‘gut feeling’ or ‘culture fit’. Use the data you’ve collected to make informed, unbiased bulk-hiring decisions. 
  • Know in advance if you’ll accept a candidate with minor flags in background checks or character references in place of professional ones. Stick to the decisions when you’re in those situations. 

Measure:

  • Offer acceptance rate – to uncover any underlying issues with how attractive your EVP or employer brand is. 
  • Applicants to hire rate – to understand if you could advertise less or in fewer channels in future.
  • Candidates to hire rate – to understand if you can optimise the size of your interviewed candidate pool. 

Pro tip:

  • Start onboarding the moment an employee signs. Invite them to your learning platform, or simply send them a video from their manager or the CEO welcoming them on board and saying how excited you are to have them.

6. Validate

To ensure your process is working, it’s essential to measure your success.

Remember: 

  • Book in an hour or two a week or so after the end of each bulk recruitment process to analyse the data.
  • Take a look at the list of challenges above, and any goals you had at the start of the process and see how you tracked against them.

Measure:

  • Candidate satisfaction

This will come from surveys sent to all candidates. It’s built into Sapia and most other recruitment software.

  • Time to hire

The elapsed between when a candidate is first contacted (in these volume hiring strategies, the assess stage) and when they’re hired. 

  • Cost per hire
    All of the hiring costs, divided by how many candidates were hired.  
  • Offer acceptance rate
    The number of offers accepted, divided by the number of offers made, multiplied by 100. If this is low, consider any issues with your EVP or the time it takes to make an offer after an interview. 
  • Diversity
    At Sapia we don’t collect attributes which could attract bias. We build an understanding of diversity by using Namsor (www.namsor.com) in order to validate the effectiveness of our platform. Namsor takes names of applicants and derives gender and ethnicity, and we use that data to understand how effective we have been at achieving diversity at each step of the path. 

Pro tip:

  • Measure, learn and optimise your high volume recruiting strategies every single time you complete a project, and you’ll find you improve each time. This will save time and money, and increase diversity. 

Bulk hiring tools that are perfect for high-volume recruiting

Technology is your friend when it comes to building scalable volume hiring strategies. Here are four key pieces of technology to consider. There are plenty of tools out there, so this is by no means an exhaustive list.  

Applicant tracking system

Your ATS will help you post ads, screen resumes, bulk communicate with applicants and collect data. You should also use it to build talent pools and pipelines for future roles. 

Interview automation

An Ai assessment like Sapia means you can give every single applicant a conversational chat interview. The quickest payback you will get on volume hiring is an investment in interview automation. Interview automation can truly enhance your high-volume recruitment process and help you make it more efficient (and pleasant) for everyone involved. This will help you get your time-back quickly, and release the budget for automation in other areas of recruiting.

Sapia meets the needs that challenge many of my clients today – how do they manage high volume recruitment processes in a streamlined and cost-effective way. while still delivering a great candidate experience and quality hiring decisions. With Sapia you leverage the latest in data analytics and tech to maximize efficiency & effectiveness; and the candidate experience is fresh and engaging, with great feedback! The product is great and constantly evolving!

Read: The Ultimate Guide to Interview Automation Employee Engagement 

It’s worth considering a candidate engagement survey. In this survey you can ask questions to reveal how well your EVP is resonating. Then you can compare candidate engagement scores with new employee engagement scores and exit interviews to understand if you’re delivering on your EVP. 

Onboarding

Integrating your onboarding software with your ATS (or choosing one with onboarding included) allows you to start onboarding and engaging candidates as soon as they sign their (automated) contract. This is a dream for getting workplace health and safety and even procedural training done before a new employee walks in the door. 

Good news: It’s only going to get easier

It’s easy to feel overwhelmed when you’re doing high volume hiring in an environment where there’s elevated unemployment or other challenging factors. The good news is that as much as the world may be getting more complicated, and as much as candidate expectations are soaring, the technology to support recruiters has never been faster, fairer or more scaleable. 

Establish your own volume hiring best practices and keep optimising your volume hiring strategies. It takes some time to set up, but the rewards are well worth the effort.   

https://sapia.ai/blog/six-reasons-automating-interviews-automation/


Join the Conversation

To keep up to date on all things “Hiring with Ai” subscribe to our blog!

You can try out Sapia’s Chat Interview right now, or leave us your details to get a personalised demo

Read Online
Blog

The ROI of Hope | Candidate Experience

Getting your organisation’s candidate experience right is proving to be something we’ll hear about increasingly. This is as job applicants demand more from companies they interact with. A recent poll on LinkedIn by a recruitment specialist attracted almost 80,000 views and 1000’s of reactions. It revealed that 52% of the people polled (one assumes mostly recruiters) believe that a template email is good enough as a response to an application for a job.

When it comes to recruitment, responding to candidates has always been an area we know has been ripe for improvement. And it costs companies too, with a bad candidate experience said to have cost Virgin $5 million.

That’s not to say there aren’t historical reasons why recruiters have not been able to respond to the hundreds of applicants received for a job. Until now it’s not been something we can practically do with limited time and resources. This is where AI plays a fundamental role in moving our industry forward as it allows mass personalisation at scale.

Hiring With Heart

This has never been more important than right now as we have had mass job losses across industries due to the impact of COVID-19. If you look at the Hospitality and Tourism industry across the globe, it’s hard to wrap your head around the sheer number of job losses with very little hope of returning to normal soon. If with every job application we were able to give each unsuccessful candidate feedback on where they could improve, imagine the impact we could have in activating the world economy.

This is entirely possible and every day we hear about the impact our technology is having on people’s lives when they get personalised feedback designed to steer them in the right direction.

81% of people who get personalised feedback from our platform said it was useful in identifying their strengths, 71% said it would help them better prepare for interviews and 59% said it would help them find a job that suited them.

And lastly, it’s not something we can quantify, but we do believe it’s important. What we’re giving so many people right now is hope. We think that’s something worth companies cultivating alongside us too.


To keep up to date on all things “Hiring with Ai” subscribe to our blog!

Finally, you can try out Sapia’s Chat Interview right now, or leave us your details to get a personalised demo


Have you seen the 2020 Candidate Experience Playbook? Download it here.

Read Online
Blog

Biased people are much harder to fix than algorithms

We worry intensely about the amplification of lies and prejudices from the technology that fuels social media like Facebook, yet do we hold the mirror up to ourselves and check our tendency to hire in our image?

How many times have you told a candidate they didn’t get the job because they were not the right “culture fit”?

The truth is that we humans are inscrutable in a way that algorithms are not, which means we are often not accountable for our biases.

In algorithms, bias is visible, measurable, trackable and fixable.

A compelling feature of our technology is that our AI can’t see you, hear you, and judge you on irrelevant personal characteristics (like gender, age, skin colour) – as a human can. That’s one reason why trusted consumer brands like Qantas, Superdry, and Bunnings use it to make fair unbiased hiring decisions.

To validate that algorithms are bias-free, we do extensive bias testing (impossible to do for humans). We know from this testing that there is no statistical difference between the way the algorithm works on men, women, and people of different ethnicity.

We use these tests for bias testing:

Our bias testing happens at 3 levels:

Score calculated by the predictive model for each candidate.
Recommendation grouping based on score percentile.
Feature values used by the predictive model for training.

For Gender-bias testing:

To analyse whether our test scores have any gender bias we use t-test and effect size. For testing our recommendations of YES, NO and MAYBE groups, we use chi-square, fisher-exact and the 4/5th rule. This last one is the standard test set by the EEOC for any assessment used for candidate selection.

For Ethnicity-bias testing:

We use the 4/5th rule and the ANOVA test.

For Feature-level bias testing

This is to ensure any of the feature values we are using to assess candidate fit are not of themselves biased, we use t-test, effect size and ANOVA test.

Diving into just one of these, using effect size is easy to understand the statistical measurement of the difference in average scores of males and females. If the effect size is positive in our test set, it means females have higher scores than males and vice versa.

The magnitude of the effect size also matters – the larger the magnitude, the more significant the difference is. We generally consider values smaller than +/- 0.3 a negligible difference, values from +/- 0.3 to +/- 0.5 a moderate level difference, and values larger than +/- 0.8 significantly large level of difference.

We periodically test for score and recommendation bias in our models and take action if the bias highlighted is non-negligible. e.g., the effect size is beyond the range of +/- 0.3 or more, we take action- stop the model until we can find the source of the bias and re-train/re-test the new model to make sure the new model is not biased.

For more insight on how our technology removes bias and how we track and measure bias, read diversity hiring

Read Online