Solving the People Puzzle · 2026-04-08 · 20 min
Key moments - from our scoring
Substance score
47 / 100
Five dimensions, 20 points each
Steven Roberts (Senior Talent Consultant), Mornay Bester (Account Manager), and the host - all industrial psychologists at Wamly - unpack what scientific hiring actually means in practice. They challenge the common pitfall of prioritizing speed over fit, and the misconception that digitizing a broken process with technology will magically improve outcomes. The core principle they emphasize is falsifiability: hiring practices must be provable, testable, and backed by research rather than gut feel or tradition. Wamly enforces this by requiring job analysis before any vacancy launches, then collecting structured data through video interviews, psychometric assessments, and role-specific knowledge tests. Critically, they stress that the real validation happens post-hire, by measuring actual performance, retention, and engagement to confirm the process works. They also discuss AI's potential to surface non-obvious patterns across multiple data signals, while warning against bias creep and emphasizing the need for human-in-the-loop oversight. For B2B operators stuck in speed-driven or gut-feel hiring, this episode provides a concrete framework: start by questioning whether each hiring practice is falsifiable and research-backed, use technology to enforce consistency and collect evidence, then continuously improve based on actual employee performance data.
A falsifiable hiring practice is one you can prove works by showing it predicts job performance, and where someone else could prove you wrong if evidence contradicted your approach. Rather than relying on tradition or gut feel, you must base decisions on testable research and measurable outcomes.
Wamly enforces structured job analysis before launching any vacancy, then uses asynchronous video interviews (single-take, anxiety-reducing), psychometric assessments tailored to job competencies, and role-based knowledge tests - all standardized so every candidate experiences the same process.
Research shows anxiety increases with repeated attempts, making candidates perform worse each time. A single, time-limited interview reduces this anxiety bias and ensures fairer, more consistent assessment across all candidates.
You must track actual post-hire outcomes like job performance metrics, turnover, engagement, and retention rates among employees hired through your process versus traditional methods to validate that your scientific approach translates to business impact.
AI can identify non-obvious patterns across multiple data inputs to make better predictions than traditional analysis, but only if it's trained without bias, continuously validated for fairness, and kept under human oversight - technology without ethical governance can amplify discrimination.
Our reviewer’s read on each dimension, with quotes from the episode.
The episode covers foundational hiring science concepts (falsifiability, job analysis, structured processes) that are well-established in I/O psychology, but presents them as relatively novel to the listener base. While the core ideas are sound, there is significant filler including introductions, repetition, and throat-clearing ('Sounds weird, two firsts'). The substantive content - job analysis requirements, psychometric practices, validation through performance metrics - is distributed thinly across 20 minutes with considerable padding.
the core principle of science is essentially it has to be falsifiable. Right. So you need to be able to like, you can't just say this is the truth and not be able to show why it's the truth.
we cannot, for example, launch a project or open a vacancy without going through the steps of already deciding what are the questions we're going to ask, what are the tests, um, we're going to use for these candidates.
The framework of 'falsifiability' applied to hiring is presented as original thinking, but the underlying concepts - structured interviews, psychometrics, job analysis, predictive validity - are standard I/O psychology practices documented extensively since the 1980s. The AI discussion touches contemporary trends but remains surface-level. The episode lacks contrarian arguments or first-principles rethinking; it primarily packages existing research as best practice without novel synthesis or counterintuitive claims.
the core principle of science is essentially it has to be falsifiable
we stand on the shoulders of giants. We need to look at what already exists in our research landscape and we try them as a first step.
Steven Roberts is a senior talent consultant and Mornay Bester is an account manager at Wamly, a hiring software vendor. While both hold psychology credentials and work in talent, neither demonstrates practitioner-level scale execution (founding a company, building a function at a major firm, or publishing research). They are internal employees speaking about their own product, which creates an inherent conflict of interest and limits the independent credibility expected of standout guest interviews. They lack evidence of having built or scaled hiring operations outside their vendor context.
Steven Roberts, uh, senior talent consultant here at Ramlee
Mornay Bester. I'm the account manager at Wamney
The episode mentions research and academic work but provides almost no named studies, data points, or concrete metrics to ground claims. References to 'research' and 'articles and journal writings' are vague. A single anecdote about anxiety in video retakes hints at evidence but is not cited. No client case studies, performance metrics, retention rates, or financial ROI figures are provided despite claims about validating impact. The South African legislative context is mentioned but not exemplified.
there's actually cool research around. Well, your anxiety then goes up and so every time you attempted to do it again, you're actually worse off than the first couple of sl.
We engage with a bunch of clients and kind of get that a bit of information back from them to kind of almost look at a little bit of thing about return on Investment?
Speaker A (the host) asks open-ended setup questions and allows guests to speak, but rarely pushes back, challenges claims, or digs deeper with follow-ups. Questions are largely softball prompts ('What are you seeing?' 'Your initial high level thoughts?'). When guests make claims about AI predictive power or bias mitigation, the host affirms rather than interrogates. No genuine disagreement or productive tension surfaces. The host occasionally redirects but does not use Socratic method or critical questioning to test the guests' assertions.
Specifically, both of you keep on using the word practices. My mind goes, okay, so can we highlight or talk about some of them?
Yeah. Amazing. Stephen, thoughts?
Computed from the transcript - who did the talking, and the words that came up most.
Stop betting on "vibes" and start hiring with certainty. In this episode of Solving the People Puzzle, we’re hosting a first-of-its-kind "Industrial Psych Panel." We’re stripping away the Metric Lie that speed is the only KPI that matters. If you can’t prove why you hired someone, you aren't using science - you're just following a workflow. We’re geeking out on how technology "forces" objective hiring by baking falsifiable research directly into the process. From automated job analysis to AI-driven performance signals, we show you how to turn your recruitment tech from a digital filing cabinet into a high-performance prediction engine. Inside the Episode: The Speed Trap: Why prioritizing "bodies in seats" is breaking your quality of hire. The Falsifiable Framework: If your hiring data can’t be proven wrong, it isn't science. Tech as a Guardrail: How Wamly "forces" a proper job analysis before you ever see a candidate. The AI Pivot: Using split-second data signals to predict long-term job performance. South African Context: Navigating legislation and removing bias through structured practices.
Transcribed and scored by The B2B Podcast Index.
Speaker A: Hello and welcome back to another exciting episode here on solving the people puzzle in the beginning of the year. Uh, I said to you that this year we're going to be trying new things and we're going to be bringing you exciting new content. And today I'm so proud to announce two firsts. Sounds weird, two firsts. Number one, it's not just me in the studio, but it's also not a guest. I have two of the Whamly colleagues joining me who we'll be introducing in a second. So it's the first time that we're three on solving the people puzzle. And then secondly, um, the privilege of having an in house panel, an in house audience all from Wambly discussing something that is really cool and that is the science and how science and technology can work together. Gentlemen, welcome. It's great to have you here in studio, uh, maybe intro yourselves and uh, just briefly share with the audience what you do here at wamly.
Speaker B: Joseph. Yeah. So my name is Steven Roberts, uh, senior talent consultant here at Ramlee. Obviously as you know Fran so typically involved with strategic uh, initiatives as part of the talent team. But also obviously then that mix between having data driven talent analytics, but also then using that for best practice implementation at our clients.
Speaker A: Lovely. Mornay.
Speaker C: Mornay Bester. I'm the account manager at Wamney. Today at the time of filming is actually my one year work anniversary. Love it. So, uh, yes, lovely time here so far. Um, as an account manager I'm obviously working with uh, a lot of our clients and trying to bring our technology and our science to them to, to improve their hiring process also on a strategic level, make sure that they're going in the right direction in terms of hiring better people faster.
Speaker B: Yes, yes.
Speaker A: So if you're a client listening, you might recognize both Mona and Steven as your account managers or your consultants serving you and helping you. Gentlemen, um, we are all three industrial psychologists.
Speaker B: Right.
Speaker A: So this can potentially become a geek out session. But don't stress, we won't go too technical. But I do believe on a more serious note that this is a really important strategic drive for wamly to ensure that we are not just building tech that can serve a purpose, but that we're also ensuring that how we are building it, our thinking behind the technology and all the modules that we build in webly and actually serves the ultimate purpose, which is how do we help our clients hire better people faster. Right, so let's unpack this in the 20 minutes that we've got by starting off Just maybe illustrating the landscape that you observe when we onboard a client who's not yet using a system like Wably. What, what, what are you seeing? Uh, which we could potentially label as not necessarily scientific kind of the, the
Speaker B: first thing that we're seeing is really people are almost kind of prioritizing the speed of getting someone into the seat instead of getting the right person into the seat. Especially for those high volume roles. Uh, we've kind of typically seen that almost kind of with that prioritization of the speed that we potentially break process, we kind of take those objective decisions and then turn them into a little bit more. But I think I have a good feeling about what this candidate will do and they'll succeed kind of. That's what I've seen in the high volume.
Speaker A: Love that speed. Speed.
Speaker B: Speed, exactly.
Speaker A: Yeah.
Speaker C: Morning speed and gut feel. Uh, we often see them uh, rather going with what they think is a good candidate, not basing it on any uh, prior analysis of what a good person might look like in that job. We see them using hiring practices that they haven't tested out in their environment. Things uh, that they think might probably work but have no real basis for making that um, decision on and just sort of putting a process together to, to get someone hired rather than thinking about all right, how do we make sure that the person we hire is someone that is really a fit for the role and would exceed our expectations.
Speaker A: Yeah, I love that because when I think of that question my uh, my mind immediately goes to, well firstly, do you have tech or don't you? Right, because we know if it's manual, it's extremely difficult, a lot of excess practice and in a spreadsheet.
Speaker B: Right.
Speaker A: Secondly, if you then do have tech, I often meet uh, clients and clients that we also engage with where, who they believe that by just adding technology or digitizing the process is going to solve the hiring problem or predict better quality candidates or predict better job performance. And uh, I'm hoping in today's conversation that we can just guide the audience around. Well, adding tech is maybe the start, but it's definitely not the be all end all silver bullet to ensuring that you're improving the quality of your hires. You gentlemen recently had the privilege of speaking at the acsg. You had a beautiful panel discussion and before we double click on that, just maybe your thinking around the wamly way in terms of science and then we can start unpacking. Well, practically what are the things that we are doing and we don't get everything right. And we are also still learning and developing and taking feedback and testing. Um, but just your initial high level thoughts around. Well, if someone in the audience is listening around a science process, what are the things to consider?
Speaker C: Yeah. So I think maybe let's first quickly define science. Right. Because it's easy to think science is chemicals that you're putting together or these abstract equations. Um, the core principle of science is essentially it has to be falsifiable. Right. So you need to be able to like, you can't just say this is the truth and not be able to show why it's the truth. And someone should be able to falsify that. So looking at that, we need to be able to say that, okay, if we follow these sets of practices or this process, then it will likely lead to a good performing employee. And we need to show in a falsifiable way that these practices actually do lead to that. And then we have a lot of very intelligent academics and people out there doing very cool articles and research and uh, journal writings on. We tested these practices and we saw that they do lead to these outcomes. So our first step is to look at like we stand on the shoulders of giants. We need to look at what already exists in our research landscape and we try them as a first step. So let's make sure that we use things that were tested in the real world out there to, to predict these performance outcomes or retention or whatever and put them into our technology in a way that suits our clients as well. It's also about making sure that we are solving their problems.
Speaker A: Yeah.
Speaker C: Um, but then after it's already in the app, then we look at how can we then build on our site and improve. How can we check whether there are some things that are actually better predictors than the science out there that maybe not for the South African context or not within the industry we're working for, and then improve and put some more science out there based on that grape point made.
Speaker B: They buy Monet specifically with the context of South Africa he brings in. So it's great to have the research and to know what's working. But obviously in South Africa, with the legislation that we have, is also to make sure that those practices like Monet mentioned can be proved and can be verified and that we don't bring in anything that potentially could lead to additional biases.
Speaker A: Specifically, both of you keep on using the word practices. My mind goes, okay, so can we highlight or talk about some of them? Because I'm hoping that some of the audience members are also going, well, what are these practices that you're talking about. So can we uh, uh, just elaborate a little bit on what they are? Yeah.
Speaker B: So the first one, uh, I assume you'll touch on a little bit later as well is the video interview. So the asynchronous um, video interview that's specifically structured, having almost kind of that first step as well, but then including things like the psychometric assessments, the job based knowledge questions, which you can really then contextualize for each role to make sure that you're measuring the right types of competencies or behaviors that's needed for each individual role. Yeah.
Speaker C: So pretty much anything that you include in your process and in your decision making for making this hire would be considered a practice. Right. And there are good and bad practices like just looking at someone's hair color could be a hiring practice, but it would be potentially be a bad hire,
Speaker A: not a great one.
Speaker C: And we can then with science rank order these to look at which are the practices or the ones that are most likely to uh, result in us hiring someone that fits the job well and will perform well.
Speaker A: Now Steven, you started off by talking about video interview. If you're listening to this episode and you don't know what the gentlemen are speaking about, inside the WAMLEY Technology, our all in one hiring system, the candidates in their application journey will go through these modules or practices and the system is collecting information of the candidates and then allowing HR and line managers to make job based performance decisions or hiring decisions off of this data. Now you started off by saying video interview. That is where WAMLY started. Right. So when we as a business started, this was before you joined. We were video interview only. And often I got asked questions around, well, why can't candidates redo a video interview? Or why is it time based? Or uh, why is the interview guide locked down when the project is launched? And every time my answer was a scientific one. And so looking back five years later, it's always been our way building out our modules, um, or methods in WAMLY while collecting information from candidates to ensure that we keep the science in the back of our mind. Because you can imagine if I can redo an answer of a video interview of I'm going to practice it 27 times and there's actually cool research around. Well, your anxiety then goes up and so every time you attempted to do it again, you're actually worse off than
Speaker B: the first couple of sl.
Speaker A: Um, but okay, so those are the instances. You mentioned video interview. You mentioned the skills test that we've got, the psychometric Assessments. Are there any others in your research that you presented that you touched on or were those the main three?
Speaker C: One other one that came to mind is how wamly, uh, enforces a job analysis in a way. Right. So we cannot, for example, launch a project or open a vacancy without going through the steps of already deciding what are the questions we're going to ask, what are the tests, um, we're going to use for these candidates. Are we going to use psychometrics or not? What competency profile are going to use for the psychometrics. So we, we essentially force the hiring practitioner to already decide, uh, beforehand what are the steps they're going to take. And they're technically doing a job analysis of what does this job require and making sure that if it's all set up, um, beforehand, we know that every candidate goes through the same process, but we also already know that it is based on what the job needs and we're not figuring it out as we go along.
Speaker A: Such a foundational principle in HR and in hiring, job analysis and a proper job profile that we to this day see organizations neglect. Right. It's this piece of paper that no one really refers back to, but if you think of it, it's the foundation of everything that flows in the employee lifecycle is from this thing that is describing job success. What does good look like? My dear friend Fred Guest always said, um, so, okay, we've got a job analysis basically as part of our project setup. Then the candidate, when they apply, go through these steps. We are collecting information. How, in your scientific approach last week, were you then communicating the impact of this to the audience at the acsg?
Speaker B: Yeah. So kind of when we look at the process and we look at everything that we've planned, uh, it's all good and well that we think that we're following research and that we know that the process is going to work, but we then need to prove that the process works. So looking at what does the outcome of this all say? So then you can typically look at research, uh, things about turnover or productivity or kind of how engaged the candidates are throughout the process or when they've been onboarded. So we do this type of research to make sure that what's the actual impact of not just the candidate going through the process, but then when they get to the place where they actually need to perform as well. So then we engage with a bunch of clients and kind of get that a bit of information back from them to kind of almost look at a little bit of thing about return on Investment?
Speaker A: Yes.
Speaker B: What's the, the return that I'm getting looking at all the efforts and the time that we're putting in about getting these candidates? Yeah, yeah, yeah.
Speaker A: So what you're saying, Stephen, is post the hire, the, the link, the golden thread towards performance and performance data and other employee data as actually where the real magic sits. It's not in the isolated selection decision.
Speaker C: Mon. Exactly. Yeah. So we, we can always make our signs better by trying to predict further into the future by, by trying to, to specify what the outcomes are supposed to be like. We, we might just look at. All right. Doesn't improve the number of hires. And we might say yes, but all those hires, actually better hires. Right. We want to dive deeper into things like their job performance and their turnover and retention and engagement. And we, we have the privilege of working with clients who give us these information to, to solidify that their processes are in fact scientific and they are indeed making a difference in their workplaces.
Speaker A: That's incredible. If we peek into a, uh, lens for a split second, a crystal ball. It's impossible these days to shoot any sort of content without using the word AI. Right. Is there in your mind an exciting opportunity for AI given this context of the conversation? Science, technology, AI.
Speaker C: Absolutely. Uh, one of the great um, advancements we've seen in AI is the ability to look at multiple uh, signals and multiple points of input and make even better predictions than a normal deterministic, uh, statistical look can, can do. Right. Because we might, let's say we're developing a new test. We might say, okay, that we see that candidates who perform above 50% do better in a job, but the AI can look deeper into those inputs. Things like what is the split second differences between how long candidates take to answer question three versus question five? And we don't know whether that makes a difference in the job performance. But with AI, we can actually make those types of uh, predictions. And um, having a system like wamly allows you to have all these inputs going into a single system, into a AI black box, to then have a much better predictor, uh, of how well someone can perform, how long they will stay at the company, um, and then in the end make a better hiring decision.
Speaker A: Yeah. Amazing. Stephen, thoughts?
Speaker B: I think you alluded almost kind of to it a little bit earlier as well. With the inclusion of technology, does that automatically improve the process? And, and I would say it's almost kind of a twofold answer. It's yes but no as well. It's great to include these types of process and Systems to get all the information together. But you need to make sure that when you train something like the AI, um, model that you're using, that it doesn't include things like bias towards specific candidates, making sure that you verify that is consistent. Um, kind of in the industry we're speaking about human on the loop or in the loop or out of the loop. And with all those process that you're including, it's just really important to make sure that throughout that, that there's no kind of golden thread that we're potentially missing 100%.
Speaker A: And I think that just ties it all back to what we've been saying all along. And that is that AI is not necessarily going to replace jobs, but it's going to replace people not using AI, with people using AI and systems and technology carefully, ethically considering the impact and the use case for AI, uh, because I agree with both of you, I think there is no future in any hiring software and any piece of technology for that matter that does not include some form of artificial intelligence. It's our responsibility to govern that ethically and to make sure that we do test and that we do follow guidelines and best practices and South African legislation and different contexts, uh, around the world. Um, gentlemen, in closing, if you could leave the audience listening now going, why? I don't have any of this and I think I'm missing the boat. I'm focused on speed or I'm um, focused on pleasing my line managers or actually I don't know where to start. What piece of advice would you like to give them from your perspective?
Speaker C: Yeah, so I think, um, one of the things I frequently ask myself before I had the, the, the privilege of working with a, uh, science based technology solution is um, how, how is what I'm doing to hire someone actually, um, based on something that I can prove. And so when I'm sitting in an interview, am I asking a question that I've seen in the past or can see through research actually does give me an answer that would make a difference in my decision. Um, how am I giving them an assessment that would actually predict their performance? If you just start by thinking in a way that am I basing my hiring practices, as we mentioned, on science, you will get further than most other companies in terms of who you hire. Um, and then a system like Lamley just makes it easier to implement that at scale.
Speaker A: But that foundational thinking, regardless of what tech you use, is the starting point.
Speaker C: Think about, is this practice falsifiable? Can someone prove me wrong on this?
Speaker A: Yes.
Speaker C: And Do I have a good argument to prove myself right again, Stephen?
Speaker B: Yeah, 100% on the nose there. I think we've gotten so accustomed to kind of following the process that we've been following. Just because all that we got given. Exactly. It's tradition. So that's just how we do things. And I think we should get out of that loop towards getting away from. That's how we've done things and going back and what's the new research saying? Because it's all good and well that we're basing. Let's our uh, hiring practices of what worked in the past. But things like. Things like the improvement of technologies, additional things that we need to start considering especially for positions that are becoming specialized or in basically being invented now.
Speaker A: Yeah.
Speaker B: Uh, as the new kind of revolution comes, I think it's about number five or six now at the moment.
Speaker A: Yeah.
Speaker B: So just really making sure that you're continuously on the search for what we. What we can improve.
Speaker A: Yeah. Gentlemen, I want to thank both of you, firstly for the amazing things that you are doing, how your minds are working, how you are helping us pushing this business forward and uh, for sharing your thoughts here on the episode. Well done. This was your first episode. I think you did great. To the audience out there, leave us a comment, leave us a question. Share with us how you are doing this. Are there other scientific approaches that you have in your current process that can help this conversation? Give this video a like a share a thumbs up and thank you so much for being on this journey with us here at Wamley and solving the people puzzle. Until next time, thanks so much for listening to solving the people puzzle. If you have any questions around how to create the ultimate end to end hiring solution, head over to WAMLY IO so that we can help you hire better people faster. Catch you next time.
Other episodes covering the same guests and topics, from across The B2B Podcast Index.
Episode 590 - ‘References Available on Request.’ A statement that is complete Garbage. with Lee SeymourTaPod · on Talent acquisition82 / 100
Talking Psychometric Assessment | Peter RhodesLeadership BITES · on psychometric assessments77 / 100
Episode 57: Kshitij "KJ" Jain - Unleashing the Power of Data with JoveoTalent Acquisition In The Trenches · on Talent acquisition70 / 100
Success Leaves Clues: Ep 302 - The Moneyball Strategy Changing How Companies Hire with Andrew Tai, Co-Founder at Deeply Talented Success Leaves Clues with Robin Bailey and Al McDonald · on Talent acquisition
Ep. 317: Creating a Long-Term Career Strategy in I-O PsychologyWorkCookie · on psychometric assessmentsThe Power of a Chance: Why Corporate SA Needs to Hire Hungry Youth. Featuring Lesego Shelembe, Specialist recruiter, Nedbank.65 / 100If managers resist hiring tech, is it about change… or control?46 / 100Marko Stavrou on Solving the Gen Z Puzzle: How to Lead, Hire, and Connect with the Next Generation59 / 1007 Ways Candidates Commit Fraud (And the 5-Layer Defense to Stop Them)62 / 100From Prompting to Executing: The Shift to Agentic AI in 2026