We’ve talked about generative AI a lot this year and its incredible potential to change and disrupt talent acquisition. A lot of our industry conversation has been about creating with Large Language Models to generate recruitment marketing content and automate interactions with candidates. However, the impact of AI goes much deeper than this, and its effect on the speed of vendor innovation is highly significant. AI is already changing the way candidates are found, and matches are made and has the potential to make boolean searching and keyword matching obsolete in a very short space of time.
So what is the art of the possible here, and how can TA Leaders ask the right questions of vendors to ensure they are dealing with genuine products that are ethical and legally compliant?
My guest this week is Mark Chaffey, CEO of Hackajob. Hackajob has integrated generative AI into its technology to vastly improve matching between candidates and roles. At the same time, they have a strong focus on using AI to improve diversity in tech hiring and are already getting some significant results.
In the interview, we discuss:
• Every company is a tech company.
• Market differences between the USA and Europe
• The continuing skill shortage in tech
• The impact of generative AI on vendors
• Building technology that actually solves problems
• Moving beyond boolean and keywords with richer LLM-driven datasets
• The questions TA leaders need to ask vendors
• Improving diversity in tech hiring
• Opt-in self, disclosed diversity data.
• Will AI change the way we think about skills and hiring
• What does the future look like?
Listen to this podcast on Apple Podcasts.
Transcript:
Matt: Support for this podcast is provided by Hackajob, a reverse marketplace that actively vets engineers. Hackajob flips the traditional model on its head. Meaning, companies apply to engineers versus candidates applying to jobs, with companies getting an 85% response rate to the candidates they reach out to, as well as exposure to tech talent that directly meets their organization’s diversity objectives. After all, the ability to attract, hire, and retain tech talent from all backgrounds is critical to every organization’s success. Companies such as S&P Global, CarMax, and Sensor Tower are all using Hackajob. So, why not join them? Go to hackajob.com/future to get your free 30-day trial today. That’s hackajob dotcom slash future. Hackajob is spelled H-A-C-K-A-J-O-B.
[Recruiting Future Podcast theme]
There’s been more of scientific discovery, more of technical advancement, and material progress in your lifetime and mine than in all the ages of history.
Matt: Hi, there. This is Matt Alder. Welcome to Episode 556 of the Recruiting Future podcast. WWe’ve talked about generative AI a lot this year and its incredible potential to change and disrupt talent acquisition. A lot of our industry conversation has been about creating with Large Language Models to generate recruitment marketing content and automate interactions with candidates. However, the impact of AI goes much deeper than this, and its effect on the speed of vendor innovation is highly significant. AI is already changing the way candidates are found, and matches are made and has the potential to make boolean searching and keyword matching obsolete in a very short space of time.
So what is the art of the possible here, and how can TA Leaders ask the right questions of vendors to ensure they are dealing with genuine products that are ethical and legally compliant?
My guest this week is Mark Chaffey, CEO of Hackajob. Hackajob has integrated generative AI into its technology to vastly improve matching between candidates and roles. At the same time, they have a strong focus on using AI to improve diversity in tech hiring and are already getting some significant results.
Hi, Mark, and welcome to the podcast.
Mark: Thank you so much for having me, Matt. Super, super excited to be here.
Matt: An absolute pleasure to have you on the show. Please could you introduce yourself and tell us what you do.
Mark: Absolutely. So, my name is Mark. I’m the co-founder and CEO at Hackajob. Hackajob is a suite of software products that help predominantly enterprise companies hire, engage, retain technical individuals. So, our central thesis is that every company in the world is now a technology business. It doesn’t matter what sector or what industry you’re in. If you’re not investing in technology, it’s very unlikely you will be successful as an organization over the next 10 years.
Despite all of the layoffs and challenges in the tech market over the last 12 months, fundamentally, there is still more jobs than there are candidates to fill those roles within tech which creates a really challenging environment for companies, because their go to channels like LinkedIn have a lot less engagement with technical people and it creates this really terrible candidate experience for technical people, because they are often just spammed by everyone, mostly about completely irrelevant opportunities. So, we exist to solve that problem. Headquartered in the UK, but spent the last 12 months expanding into the US. I’m currently sat in New York with our team here.
Matt: Lots of things I want to ask you, but before do, tell people little bit more about your model because it is quite different to some of the other solutions that are on the market.
Mark: Yeah, totally. So, we’ve got a couple of products, but we’re most well-known for our two-sided marketplace, which is a sourcing product. So, on one side, we have a community of technical people. When they come onto Hackajob, they go through a technical onboarding, which is all fully automated using the tech that we’ve built, which really enables us to understand how good is this individual in the skills that they say they have. So, that could be programming languages, frameworks, architectures, design types, etc.
We then take them through a job fit onboarding where we understand their salary expectations, seniority, visa status, work location preferences, etc. And that gives us a ton of first party data that’s really,really interesting for us to then be able to use to match. So, if a candidate goes through that onboarding process and our system deems them to be relevant for the jobs that we have in the marketplace, they will be visible to companies for a period of four weeks. It’s here where we flip the model. So, rather than the candidate applying to the job, the company actually applies to the candidate. This creates this really magical candidate experience, because they only get pitched by companies that meet their salary expectations, visa status, tech stack, location preferences, etc. Then because the candidate experience is so good, we have exceptionally high candidate engagement, which means companies end up receiving about an 85% response rate to engineers that they end up reaching out to through the solution.
So, we’re predominantly partnering with larger organizations that have super high volume of technical hiring to do and are looking to be far more effective with their sourcing time rather than sending hundreds and hundreds of messages on these sourcing platforms, actually finding really high intent, super relevant candidates that they can then get through their interview process.
Matt: You mentioned a minute ago about the state of the market. And certainly in the headlines for the last, well, almost 18 months now, we’ve heard about tech layoffs and big tech, not hiring as much as they did before, all kinds of disruption in this market. What’s the market really like and what’s it like at the moment? What are you seeing in the tech talent market right now?
Mark:Yeah, it’s super interesting. We raised our Series B earlier this year. I said to our team, “We’re fighting one of the greatest PR stories of all time that like all of this tech layoffs and there’s no tech hiring happening.” I think it’s a far more nuanced discussion than that. So, I think firstly, you have to differentiate between people working in the tech industry and people doing technical jobs in any industry. Actually, a lot of the big tech layoffs didn’t actually impact technical people. But because you were a salesperson working at Google, you were considered a tech layoff. Now, some of the teams did get impacted, but a much, much smaller minority.
Then the other thing that’s happened is that every company is now hiring technical people. I saw a stat recently that Walmart is currently the biggest advertiser of tech vacancies in the US. So, Walmart are not making tech [unitelligible [00:07:04] Actually, Walmart are investing more in technology than they’ve ever done before. I think if you compare it to the boom times of 2021 and early 2022, for sure, that was an incredibly tight labor marketplace and weird and wonderful things were happening. We were seeing some of our users on the candidate side getting 25%, 30% pay rises like twice in a year by moving jobs. That was obviously never going to be sustainable. But I think if you just look at the fundamental market dynamics, there is still incredibly low unemployment here in the US.
There is still an incredible amount of demand, especially from these nontechnical companies or traditionally nontechnical companies. And bluntly, both the US and the UK were still not producing enough technical people. So, whilst it’s not quite as crazy and as tight as it was in 2021 and early 2022, there is still a hell of a lot of hiring that’s happening in the tech world, often actually in companies that set outside of the tech industry.
Matt: You’ve anticipated my next question there, because I’m really interested to know what you’re seeing. Are there any differences between the European market and the US market in this?
Mark: Totally, there absolutely is. It’s really interesting now that we’re both in the US and the UK. I think for the last 18 months, the US was probably three months to six months ahead of the UK. So, when we saw a dip in activity here, it would hit the UK a little bit later. There is still cultural differences between the US and the UK. I think from a recruitment perspective, it is much closer to maybe the UK and France or the UK and Germany, where I think there are much bigger recruitment culture differences or employment culture differences, but there still is nuance between the US and the UK
One thing that absolutely blows my mind about the US is just how expensive talent is here. The average software engineer that gets hired on Hackajob in the US gets about $170,000 base salary. The average software engineer that gets hired in the UK is about an £80,000 salary. So, it’s incredible that the UK has effectively become like an arbitrage opportunity for the US, much like some of Poland and Romania is for the UK. So, that’s a fascinating dynamic to observe as well.
Matt: That’s really interesting. I didn’t know that the salary gulf[?] was that big.
Mark: Yeah. It’s interesting, because ultimately, we work with global organizations, so these are the same people doing the exact same job for the same organization, but getting paid much more in one geo than the others. I think with hybrid work, with remote work, I do think a lot of companies are questioning, do we need talent in these high cost locations? Actually, should we be looking at offshoring, which is funny, because most companies spent the last decade insourcing everything that they offshored in the early 2000s and now conversations around offshoring is definitely happening more frequently than what it was before.
Matt: The other big headline grabber of the year is generative AI. We’ve talked about it a lot on the podcast, obviously, because I think it’s illegal to have a talent acquisition podcast that doesn’t talk about generative AI these days. But we’ve always talked about it very much from the perspective of talent acquisition, how they can use these tools, and what’s available to them, and how they should be thinking about it. I’m really interested from a vendor perspective, not just in terms of how you’re bringing generative AI into your product, but what effects it’s having on things like your speed of innovation and things like that. What’s it like for vendors at the moment when it comes to these massive leap forwards that we’ve seen in AI?
Mark: Yeah, it’s such a fascinating space. I think the first thing that we did as an org is understand how much of this is a hype cycle and how much of this is a paradigm shift in technology. So, if you look at what happened in 2021 and 2022, webfree and crypto was massive. It was a massive hypecycle. There was some recruitment products that tried to build– I hate quoting Steve Jobs, because I feel like you always sound like a bit of a dreamer when you quote somebody that’s obviously so exceptional. But he’s got this brilliant line that says, “The consumer doesn’t care about the technology that you use to solve their problem. They just want their problem solved 10 times better than it’s ever been solved before.” I always felt with webfree and crypto, it was technology in search of a problem rather than it fundamentally solving the problem.
So, the first thing that we did when we saw all of the OpenAI innovation and what happened with ChatGPT was actually run an internal hackathon with our team and say and we ideated across the entire organization, so all of our team could contribute ideas that they wanted us to go away and test with. Then our team went and did a hackathon and thought, is there actually a paradigm shift in the tech here? Does this enable us to do things that we haven’t been able to do before? Actually, really excitingly, we believe there is. We believe that actually there is fundamentally new ways to match talent to jobs than what there has been done before. If you think about what’s happened with hiring since the internet, we’ve effectively still been left with boolean searches and keyword searching. At the end of the day, if you look at any real search product, it comes back to that. It’s a very narrow way to find talent. There are some fantastic sources out there that do some crazy boolean searches. But at the end of the day, you’re still relatively limited.
What these large language models are enabling us to do is effectively infer a job description, infer a CV or work experience, and actually make a human level matching decision based on the inference of full text, not of keywords. So, rather than being like this person knows Java and Spring Boot, you could interpret whole sentences from their experience or a whole paragraph from their experience, which gives you a much richer data set to match talent to jobs. So, we took the decision earlier this year to actually rewrite our core matching engine, which was a brave decision to take. We’re going to be doing a lot of product announcements for our Q4 and early 2024 in this space.
The first announcement that we did make was a new platform for increasing diverse candidates in tech, which we can dive into in more detail. Some of that platform is using some really interesting generative AI to ensure that all of the attraction and awareness content is inclusive. So, I’m really excited by it. I think that if I’m a buyer of HR Tech Company right now, I’m probably very wary of it because I think every organization is going to lean into AI from a product marketing perspective. But actually understanding what are the core competencies, what are they doing that fundamentally hasn’t been enabled before, I think is key when trying to understand the landscape.
Matt: Yeah, and I wonder if you could just expand on that, because I know that lots of people listening, particularly over the next three months, six months, will be bombarded by vendors unveiling AI in their products and all this kind of stuff. There are obviously some fundamentally important things that it can drive in terms of what the products can achieve and how TA moves forward. What kind of questions should people be asking vendors about this to make sure that they’re getting really effective solutions?
Mark: Yeah. So, I think there’s two broad buckets I’d put this in. How much of this is product marketing and vaporware versus actually fundamentally new technology doing something new for the first time? That’s a really hard question to answer candidly, because some of these organizations are exceptional at product marketing. This has always been the case. We’ll lean into whatever the current thing is. I think what I’d be looking for there is really trying to get proof of concepts, trying to get trials of this technology, because until you’re actually using the product, until your recruiters are using the product every day, you won’t really ever understand the impact it can have.
Again, it comes back to that Steve Jobs quote. When your team are using this product, is it making them 5 times, 10 times more effective than they have done previously? If it’s not, then it doesn’t really matter what technology they’re using. Whether they’re using AI or some simple regression model or whatever it is, it fundamentally doesn’t matter. So, I’d be really testing these vendors and pushing them hard to be doing proof of concepts trials with a small subset of your team, and then really closely monitoring the impact that has on your team. So, I think that’s the first thing I would do.
The second thing is all around the legal compliance and ethics side of AI, which we need to be so careful with as an industry. There are already some horror stories in our industry of automated decision making in the hiring process, reinforcing existing biases, etc. So, I would be very wary of any technology that is making automated decisions. I think that’s an area to stay way, way, way away from right now until you can really open up that algorithm and understand why is it making certain decisions and what is the training data that they have used to build those systems.
Where we’ve decided to focus on is in recommendations. So, we’re not actually using AI to take decisions. We’re using AI to make recommendations. And then ultimately, the internal recruiter can make those decisions whether they want to move a candidate forward, interview them or not. So, they’re the two big buckets that I would be looking at as a buyer of HR technology over the next three months to six months.
Matt: You mentioned the long standing issues with diversity in tech hiring. Tell us more about what that problem actually looks like today. What is it that causes that? What are some of the causes of that? What’s your solution? What part of the solution are you to those problems, and how does that all fit together?
Mark: Yeah, totally. It’s a fascinating space. It’s obviously received so much attention. We’ve been doing this for seven yeras or eight years now, and it’s always been top of the agenda of things that people want to discuss. I think what is the fundamental challenge of representation in tech? It’s a systemic issue. At the end of the day, I think Stack Overflow have probably got the best data on this, around 9% or 10% of professional software engineers are female versus the population that is roughly 50-50.
There are just systemic issues and there has been a lot of work done to improve representation in more early careers and junior level roles. And that will just take time to wash through as those people progress in their career and become more senior, but it’s promising to see that the representation data is better in those early career brackets.
I think when you talk to the practical challenge that companies have when actually trying to improve representation in hiring is the vast majority of companies do not get self disclosed opt in DNI data on the candidates they are interviewing. I was speaking to one of the FAANG companies last week, and they get about 6% of their applications will self disclose and terrifying. The internal recruitment team are using visual representation on the rest of the candidates. So, the recruiter is deciding, is that person a man or a woman, what’s the ethnicity, etc., which is terrifying. It’s like the worst possible thing that we could be doing.
So, the first part of our solution actually just opens up the data to a company’s hiring funnel. We get about 80% of our users given us self disclosed opt in DNI data across gender, ethnicity, neurodiversity, disability, sexuality, veteran status, and any reasonable adjustments they need in the interview process. That data is then anonymized and played back to an organization during their hiring funnel on Hackajob. So, they could see that maybe they are requesting 50% men and 50% women for interview, but 80% of men accept their interview request and only 20% of women accept the interview request. They will then be able to find out, because each time a candidate declines them on Hackajob, they give a reason why, why are female candidates declining them at a higher rate than male candidates, and actually be able to take actionable insights to then change that and move the needle.
Because we’re often around 25% of a company’s hiring pipeline in tech, it’s a really meaningful chunk of the sourcing that they’re doing via Hackajob. So, just opening up that data and actually being able to enable companies to measure their hiring funnel broken down by these different characteristics means that they can then start tracking, “Okay, well, is our diverse interview panel working? Is our more diverse outreach working,” etc. So, that’s where it all starts. It all starts from data and insights.
We’re then doing some really interesting generative AI stuff where we will analyze people’s job descriptions, outreach messages, company profile pages, and ensure that all of the language that they are using there is inclusive language, it doesn’t discriminate, etc. And then the final thing we’ve got is on our sourcing product. You can tell the sourcing product your diversity goals. So, if you’re trying to increase female representation in your technical team, our matching engine will then open up the matching criteria to candidates from that background.
So, we’re not discriminating, we’re not excluding men from the search results or anything like that. Instead, if you’re hiring for a senior software engineer that needs to know Java and AWS, we will match you with candidates that might know Java and Google Cloud platform instead and say, this candidate doesn’t have all of the skills you’re looking for, but is from a diverse group that you are trying to increase the representation from.
So, we’re really trying to take a holistic approach to DNI. It is a very thorny issue, it’s obviously a very emotive issue, but I’m really proud of the work that the team has done. So, in our eyes, build a product that we think can genuinely move the needle.
Matt: Have you seen any results from that? Is there anything that your clients are reporting back in terms of what’s happened?
Mark: Yeah. We work with one of the biggest pharmaceutical businesses in the world, both here in the US and in the UK. They were one of the first adopters of this product, and they’ve seen the female to male candidate pipeline go from around 21% when they first started using this product. Now up to 34% of candidates that are accepting their interview requests are female. That’s because they were able to do a lot of A-B testing on the content that they have on Hackajob. And then because we measure what they do, we’re able to close a loop on that and come back to it and say, “Actually the variant ABB and off we go.”
So, we’re already starting to see the top of the funnel impact and obviously, Hackajob is a top of funnel tool. It’s then up to that organization to run a fair hiring process to ensure that the best person or the most suitable person is getting the job, but there’s already been some really, really promising results from some of our big enterprise accounts.
Matt: Just picking up on a couple of things that you said, because what was interesting to me there, you were talking about presenting candidates who didn’t have the exact skills that the company was looking for, but had adjacent skills that could be relevant. And also, earlier on, you were talking about using AI to infer more about someone’s background. Do you think that these kind of developments in technology are going to see us think very differently about hiring being much more open minded and more equitable about how candidates are judged, particularly in a tech environment where articles can be seen as the main criteria?
Mark: I really hope so. I really, really hope so. I think not only because we can now match talent to jobs in different ways and we can infer far more about an individual, but I think the actual technology we’re talking about with these LLMs, I think is fundamentally going to change how most knowledge work is done. I think we’re going to end up optimizing or valuing school skills such as problem solving, the ability to learn far higher than your proficiency in a specific programming language, framework, or tool.
I think this has been true for a long time. I think if you go and speak to the best CTOs or VP of engineering, they will tell you that somebody that is really passionate, that has a great problem solving mindset, that has the ability to learn, is far more important than if they’re an expert in Java or Python or C Sharp or Go or whatever the language might be. I think you’re going to see that applied to just about every knowledge domain, every knowledge profession. So, I really hope so. We’ve been trying to push this case for more skills based hiring and aptitudes based hiring for a long time. I think not only because of the way that we will be able to match talent to jobs, but because of the fundamental shift in the nature of the work that I think all of us will be doing. I do think there will be an opportunity to fundamentally change how hiring is done.
Matt: Which leads nicely on to my final question, which I suppose is in some ways a summary of what you’ve been saying. But what does their future look like? What do you think talent acquisition is going to look like in three years to five years time?
Mark: Yeah, really interesting. That’s an interesting time frame to put on it as well. So, one hype that I don’t believe is I don’t believe we’re going to have a global distributed workforce. I think that the number of companies or the percentage of companies that will be fully distributed will be higher today than what it was in COVID or before COVID should I say, but I still think it’s going to be the minority of organizations that operate in this fully distributed environment. So, that’s one thing I don’t believe.
What I do believe is that, every knowledge worker is going to be operating with some form of AI based copilot that is going to make them far more effective in their day to day job. Now, I do not believe that’s going to replace humans from the recruitment process. At the end of the day, the human in the recruitment process is there to deliver an exceptional candidate experience. And until it’s AI, hiring AI, which might happen in our lifetime, who knows? I think the humans are critical in the loop.
I think what we are going to see is that a lot of the work that a recruiter does today will be automated. So, I don’t think you’ll spend so much time doing outreach, so much time doing sourcing. What I don’t think will be automated is those human to human relationships that human to human contact that I think is incredibly, incredibly challenged to automate. And so, what does that mean? I think that you’re probably going to see smaller TA teams working across more requisitions than what they have done previously. I think you’ll see more leverage in those teams, and I think you’ll see a lot of tools and technologies being used, and I think that’s going to be true of pretty much every knowledge work and not just TA professionals. So, if I was a gambling man, that would be where I’d place my bet over the next three years to five years.
Matt: Mark, thank you very much for talking to me.
Mark: Matt, this has been a pleasure. I really, really enjoy it and I’m a big fan of everything you do. So, thank you so much for having me on.
Matt: My thanks to Mark. If you’re a fan of the Recruiting Future podcast, then you will absolutely love our newsletter, Recruiting Future Feast. Not only does it give you the inside track on what’s coming up on the show, you can also find everything from book recommendations to insightful episodes from the archives and first access to new content that helps you to understand where our industry is heading. Sign up now and also get instant access to the recording of my recent webinar on the future of talent acquisition. Just go to recruitingfuturefeast.com/webinar. That’s recruitingfuturefeast dotcom slash webinar.
You can subscribe to this podcast on Apple podcasts, on Spotify, or via your podcasting app of choice. Please find and search all the past episodes at recruitingfuture.com and don’t forget to sign up for the newsletter, Recruiting Future Feast. Thanks very much for listening. I’ll be back next time and I hope you’ll join me.
[music]
