Operationalizing Headcount Planning with Real-Time TA Data ft. Ian Jones from HackerOne
Podcast Overview
-
Eric Guidice: Okay. Headcount experts. episode 20, probably the most special episode that we have. We have Ian Jones, the VP of TA and interim head of people at HackerOne, joining us to talk in an extended form about the TA data and the workforce planning process. And with a tremendous amount of experience from Amazon, from HubSpot, now running the show at a very popular startup, Ian Jones. Welcome to Headcount Experts. Nice to have you on the podcast. And I am of course joined by a Boston native MIT alum. MIT alum, headcount expert, Wayfair alum, Chris Mannion. We're going to have a wonderful discussion about
Ian Jones: Thanks, Eric. Really appreciate the opportunity. Excited to chat.
Chris Mannion: haha
Eric Guidice: how to use TA data in workforce planning. We've all had a TA job in the past. That's what makes this cool. So we speak the language. We can get as deep as we want. But Ian, give the audience a little, give us your story. How did you get to where you are today? What makes you a perfect guest for the head count experts?
Ian Jones: Well, thanks. It's been quite a road. I've been in TA for a long time. I got in right ahead of the year 2000 rush. So that's how long I've been at this. an agency and feel extremely lucky to have been trained in the Aerotech era. Learned a lot of the fundamentals that I still use today. So spent the early part of my career, probably the first 10 years in agency, learned a lot there, worked with lot of different companies. Then got a big break in 2008. The break I got was working for a startup. So got to work with an extremely high profile, true internet scale startup. Wasn't easy. It was my first entree into things like headcount planning. And at that moment it was an old org chart on a piece of paper and like four pile of resume pile of like four resumes. And like, this is what we got. So good luck. So, but it was great. had to really elevate my own understanding of how hiring related to the business results, right? So getting an understanding of, okay, here's what we're actually trying to build. This is how long we have to build it because of our burn rate. So we need to hustle and these are the people we need. So it was the first time I really had to look holistically at how people relate to an organization. Fell in love with the startup space. So spent a decade in that space and built some really interesting companies from the ground up. literally everything from companies that were selling t-shirts and e-commerce all the way up to quantum computing. So got to do a lot of companies there and really love that space. And then sort of the latest chapter of my career has been the last decade or so. I really got curious about scale. I wanted to see TA at scale. So I figured I'd go from a solo entrepreneur, which I was at the time and went to Amazon. I figured I might as well go to the biggest company on the planet. to see what scale is all about and had a great adventure there. And then today I find myself at HackerOne, originally joined to grow talent acquisition here as we're going through a transformation from essentially a managed services company to an agent AI C10 platform, elevating our talent along the way. And then through a set of circumstances, I also find myself currently as the interim head of all things people. So as always, I feel like I'm getting into new adventures and learning along.
Eric Guidice: I love it and I'm sure you and Chris will get down on the TAS scale in the Boston scene. That specific combination of TA and running people through a transformation in the AI era is great context for today's topic, which is essentially the TA data and workforce planning. What I like about the workforce planning process for the TA leaders is it broadens the perspective of what TA is to the business, relationship between hires and productivity. in the world of AI, been having, had Reid Gilbert on, he's a VP of Finance. from Reprise, a 20-person startup that raised $80 million and he's talking about from a CFO's perspective, you know, the relationship between tokens and headcount. And so on the people side of the business and the finance side of the business, this is a new relationship happening. So a little later in the episode, we're going to get into what specific data you're looking at and why. But you know, what is the... you know, the general approach to the workforce planning or the headcount process at HackerOne today. And how does it differ from your previous experience where you're at scale with Amazon or HubSpot?
Ian Jones: Yeah, that's a really good question. So I think the nice thing about being with a smaller or mid-size organization is that, one, the headcount process is a lot closer to what we're doing, right? So there's a really clear line between where HackerOne needs to go, the transformation we need to make. and the people we're going to need to get there. So I feel really good about that. think we've done a really nice job of aligning those two things. And it's meant things though, but it's meant challenges. Like we've had to deal with challenges of, hiring, say, AI engineers. Like we didn't have any on staff. So we started making a transformation for something we necessarily weren't prepared for. So we went out and hired a bunch of AI engineers, but now everybody in the organization essentially is becoming an AI engineer. So bringing that group along and making sure we're being really intentional about that. the head count is more than just going and getting talent. It's really, because we're transitioning so fast, it's how about we're developing that internal talent as well. And then the challenge becomes, how do we retain them? Right? Like that's the bigger challenge. So you go and you train somebody in the latest grade. various things, well now you've got to pay them to do that otherwise somebody like me is going to come take them. So I think that's really been interesting to us. But I remember a process essentially like any others, it starts with the dollars, right? So we work and we talk about what do we need to accomplish, what are the dollars we have to do that and then we're able to kind of go in and actually align on what needs to be hired. I don't think we have it perfectly nailed. TA has a voice but I would actually like us to have a bigger voice going forward because I think we can help make it even more targeted and better. you
Eric Guidice: What? Any? We had... go ahead. Sorry, Chris.
Chris Mannion: Yeah, just digging deeper on that, that kind of point that you just made Ian about how you're training and re-skilling and there's almost like, you think about like the prisoner's dilemma of kind of you and the competitors and if you both upskill everyone, then everyone wins. If you upskill and they don't, they just hire your people and kind of likewise the other way around. How do you kind of think about that in terms of like the talent competitiveness and actually being able to actually get that into your head count plan? Like, do you make a for increased attrition? you monitor your competitors or you just kind of focus solely on the internal tile base?
Ian Jones: It's a great question. It's a little bit of everything. So I think we have accounted for increased attrition in this market. I was literally just on with one of our talent strategy partners today. And I was equating back to that when I got into recruiting. In the late 90s, I was literally pulling COBOL programmers out of retirement. Not kidding. Because their rates were going up so fast. You're talking people who had never made more than $50,000 a year in their entire careers. And suddenly I'm calling them, offering them 50, 60, 70, 80, name your price per hour to go change date fields because that's what the market demanded. And I think one of the challenges we're dealing with right now is how do we keep our compensation approach updated with how fast the market's moving? And yes, so we are thinking about attrition. We're expecting attrition, especially for these in-demand roles. Like we know somebody is making those phone calls and going after our folks. We know that for sure. Internally, we have to think about how do we keep our team that is upskilling from picking up the phone. Because at end of the day, if you don't pick up the phone, if I keep you happy enough that you don't pick up the phone, and whether it's you, Chris, on the end, Eric, the end, me on the end, we can convince somebody to make a move, especially once we start dangling some dollars out there. So we need to keep our team very engaged with our mission. We need to reward them and keep them moving forward, both skill-wise and appropriate compensation-wise, so that we get to the point where we focus that they don't look elsewhere. Because when the market's this hot, it gets a little scary.
Eric Guidice: Any new positions or any changes to the structure now that you're in the age, whether we like it not, sounds like Hacker One's going through a bit of a transformation. In our previous podcast, we're commenting on a new position that we saw, think, at Townful, which was a head of people in AI, which is somebody to monitor what is being built or to consolidate all the different builders into a singular strategy. So as you think about decentralized building or the ability for a team to add productivity, that's not only headcount, right? Now teams can add productivity or production without adding heads. Anything new coming in your conversations or people thinking about data security or consolidation of building into a single strategy? Is there a position around it? Does it fall under you to own? Is it someone else? How are you as as the TA people leader kind of working through that workforce planning conversation with this in the age AI.
Ian Jones: Yeah, it's a great question. I think for me, it's been I've always found that that head counts one of those functions that falls between the cracks a little bit. Like everybody wants it. Everybody wants it deeply. Nobody necessarily always wants to own it. It's every organization I've been with, right? Like there's a lot of really important stakeholders, leaders, finance, people, operations, T.A. I think actually what we have the opportunity with with products and with AI, we have the opportunity to not worry about having a person on it. And I think that's very much a talent position going forward is in the past, we've always thought about like who owns this. Now we get thought to think about like what process, what agent, what automation owns this. And I think actually, I've been playing around with some stuff internally. We have access to Claude here and some other products. I'm thinking about workflows that can increase communication and visibility because as you know, the biggest challenge in headcount, it is a It is a angry and ever-changing beast. And it changes every day. And to think you're going to put this plan, here's our headcount plan for the year. And we need to have that. It's a great starting point. But the moment you put it out there, it's dated. It changes. And that communication, feels like that's where most things fall out. It falls out around the communication, the visibility, all those things that are becoming challenged. I feel like AI is an opportunity to... to put that into a more sound process. But then the flip side is everybody's building their own one-off product and data security becomes something that I'm super nervous about because whether it's our ATS is putting out an MCP that we can start building against, but like. Okay, but who's now got access to that data? Because as you know, people data is by nature high risk, right? It is viewed as high risk AI data. So how are we balancing this new flexibility with the controls that we don't do something silly that exposes us and the data of the people who trust us the most.
Eric Guidice: Chris, I know you have reaction to that. Chris is building a pretty awesome, you know, he has a few, they're all SHRM accredited courses, but he has a how to build with TA data into Claude lesson that's pretty awesome. So I know you're thinking about something now as it comes to the data security piece. What are your thoughts on what Ian's saying about the people data?
Chris Mannion: Yeah, I'm kind of in this exploration phase right now, because you have a few different kind of frameworks that you've kind of got operate within. You've got the kind of legal requirements, especially if you operate in Europe or California, the data requirements are much stricter. And I think it makes sense just to adhere to whatever the strictest standard is as default. But then there's the issue of actually not giving enough flexibility. And we just talked about retention of key employees, especially people that like to tinker with AI. So if in your organization, you're not given the tools to allow people to do that, are they going to go somewhere else that does have the tools? so I Ian's job is especially difficult because you've kind of got to, you've got to like draw the line between your risk and reward and retention of key employees. And I'm just very curious actually how to do that. I'm building that next module right now in our course. just to try and give some advice, but actually frame it in the kind of legal groundwork as well, which is something that, you know, I'm not trained in and we're kind of trying to explore. So I'm curious, do you think about that? How do you think about the firewall around the people data, who gets access, and then how you monitor that and the potential for, you know, someone in a team outside of HR to get access to payroll data or all kind of privileged information that shouldn't be released somewhere else.
Ian Jones: Yeah, mean, it's a great question. And I think that's something we're working on internally here. I mean, at an anecdotal level, we've already seen it happen, right? I've already had engineering managers build their own GPT to evaluate resumes and I had to shut it down. Not because I don't value their innovation, but like you can't do that. Like that those are against the current laws are saying that you can't inform hiring decisions. Now on the flip side, one of the things that I'm building now, which is really cool is I'm basically rebuilding the Amazon Barraiser program with an agent. which is to me is a huge win because it doesn't judge candidates, it judges the interview process. I'm legally, cool. It never sees the candidate's resume. doesn't really care about it. says, what did we ask? What are we going to try to interview? Did we interview against it? And did we, do we have enough data to support our decision? Literally the Amazon Barraiser program, which is a massive investment of time, we can turn into a bot. So I think there's opportunities there to innovate without touching that data Chris, cause it scares me to death about that. So I think we're to have to be really thoughtful about do it now. I know technically, Claude is supposed to inherit the privileges of whatever app it's pulling from. I'm not willing to give that up. And so there may be some things that we choose to not get into AI, but can we automate all the processes and communications around that so that that data is, at least people are aware that, you need to go check this data, you need to double check this. And because I think we do lose a lot of momentum in that communication. So there may be some limits to things we're willing to put out there. Because to your point, I don't want to release really sensitive data to. folks that aren't attuned to it. Now we have a nice firewall, which is me and our legal team. We review every AI that comes in, but once that firewall is gone, what do we do? So we may just choose to not expose some things. I mean, I'm okay with that.
Chris Mannion: I think that's not a problem, actually segmenting things out. And I heard recently at a conference that
Eric Guidice: Yeah, shameless plug.
Chris Mannion: the focus is kind of almost shifting away from the, like the individual and automating the work that they're doing. which, which normally involves working with personal data to actually thinking about managers and how, how much of the workflow of a manager can be automated away so they can actually spend more time doing the things that are critical that need a human eye that are kind of personal. And that's potentially the right path to leverage AI to improve capacity across a team, not necessarily having AI make hiring decisions on behalf of a hiring manager. So I think that's a really smart move that the workflow process automation is definitely the way to go there.
Ian Jones: Yeah, that's a great call out. I think one of the things as I go through, I was challenged by my CEO recently because when we got Claude, she was challenged just to think about things. I was like, man, most of my day is eaten up by people-to-people communication. I think people-to-people stuff is really hard to AI. I think If you have data on one side of it and people on the other, you can do it. If you have data and rules, like that's great. That's where that bar razor program is just data and rules, right? So I think it just, gets increasingly hard when there's people on both sides of it. So I'm okay with that. Like let's automate comms. Let's automate things with rules and data as much as we can, as long as it's not scary. But I very interested in your course because if you can give us some guardrails on it that would be amazing. So that we have more time for the people to people things because I'm not convinced that we're going to solve people to people through AI in my the time left in my career.
Eric Guidice: Has this AI change influenced, as you're the interim head of people, the people-to-people values that you've installed? Have you taken a look at the company values and said, of this, we might actually be changing the way we expect the people to interact with each other? And is that on your mind if it hasn't?
Ian Jones: So good point. mean, from a, so we look at it in two ways. Our core values are unchanged, right? Like the idea that when, when together, the customer obsession, like those are the things we care about. Those don't change how we do those. we, have talent and leadership principles. So how we do that has changed dramatically and will continue to evolve. So AI first absolutely is critical, right? Data-driven decision-making. If anything, think AI is helping us deliver on those principles. better, but our values remain unchanged. We are not willing to sacrifice our values for speed or any of that. Like, if you don't have your values, what are we doing? So yes, we have thought about that, but for instance, it's a great call out. Like our AI first talent principle is being updated and we just rolled it out in September. But we're like, you know what? need to, like our expectations are changing that fast. So we need to update that. And on the flip side, like we need to set guardrails around like discernment, right? Like we're getting some internal AI slop, right? Where people are like, okay, I gotta use AI. Let me do it to do this. And then they dump it in and the person on the receiving end is getting stuff that they have to run through their own bot. So now we have like. So like those are the we're still working through some of those working norms that we have to do. But yes, absolutely. We are having to update our talent principles to keep pace with change and they will they will change over time. I fully believe that.
Eric Guidice: Internal AI Slabs. The first time I've heard about that, I like it. I'll say that because I have everyone's experience. Like you ask for details and you get the backstory, the context, all the involved stake and you're just like, I just want the plan. What do we do? What is the three-step plan? And I have a 10-page diagram. We all have been there. Well, let's switch. I know we're kind of at the midpoint. Let's talk the workforce planning process and Ian's perspective on workforce planning. I'm not sure, you know, when's your fiscal, when do you go through the process, what's your style? Just to set a little bit of context, I'm sure it's different Amazon HubSpot to HackerOne, but you where you mentioned earlier, some companies might find some level of base in what the annual plan is. I'm sure there's a rolling forecast or an update or a re-forecast time there. So I guess, when's your fiscal year? What's the cadence of planning? And then let's kind of dive in. the process so that we can learn a little bit about the TA information that you've been delivering to help that process along. But give us give us an idea of your annual planning process as it is today, if it is any.
Ian Jones: Yeah, mean, we're our fiscal year study offset. start February 1st. we were in February 1st, January 30th on our fiscal year. Workforce planning process starts like late summer. we start to kind of get a feel for like, okay, you how is the business going? What are the, you know, what are we think we're going to accomplish next year? Those types of things. And so we start those, those early conversations and then start, you know, working, working from that. And then along the way, yeah, we do get to have, ideally we would have even more TA input. Some of the things that we've been, we've been like really focusing on is getting more realistic understanding of like how long it takes to get a hire, right? It's a critical data point that often doesn't get put into headcount plans. So you could put in, know, that, you know, we're going to start the fiscal year on February 1st and that, you know, all your hires are going to start on March 1st. Like you can do that. It looks great from a financial perspective because you're like, oh, look, we're running under our, our, our expected spend. But the reality is, like, you, you, like, you're not going to get those people in seat. So we've worked really hard to understand, like, how long does it take from the time that we say we're going to do a headcount, we start that process, right? So we start going through, to the time that person's in seat. by role, by level, so that we can actually deliver for the business as expected, both financially. So I think we didn't quite get there this time, but now we actually have that data, so we're gonna be a little more informed on that, I think, the next time of the year. So then, yeah, we go through that approval process, like everybody else, you build it, it goes to the board, everybody buys in, and then it gets launched, but we're still, you know. Like most companies, mean, I don't know how many companies actually, what percentage of companies actually deliver their headcount plan at the start of the fiscal year. Most tend to trail and we still, we still get there a little bit. So we're working on it. It's a work in progress, but for the most part, we, think directionally our TA team knows where, we need to do and what we need to hire. And if anything else, my, my guidance to my team is go with what you know and don't worry about what you don't. If you feel everything we know about, then we'll worry about the things we don't. And let me worry about that.
Eric Guidice: I just took a quick glance at your careers page. It looks like you're predominantly a US based hiring, but I saw some in Pune, India as well. you one of the things that we try to give to our TA customers is the idea that there's a different notice period in different locations. That might be relatively similar between India and the US, but when you start expanding it to Europe or parts of Asia, the notice period could be equally as impactful as the time to fill. And then before that, like whether or not the role is approved or unlocked. especially when it comes to backfills. then Chris has, you know, two things that I really love. One is Little's Law about prioritization, and then, you know, to the idea that there's a ramp to productivity that is, you know, it's starting on one type. Starting is great, and I think that's the TA metric, but the business is backing into when, especially, I'll use sellers because it's the easiest. This person needs to be producing revenue on this date. means they need to ramp. They need to start on this date. They need to quit their job on this date. They need to be in the ATS on this date. So, you know, I don't know, Chris, you want to share a little bit about the Littles Law stuff that you've been teaching me and get Ian's feedback on it? I'm curious to listen to two head counter experts go at it.
Chris Mannion: Yeah, I'm actually curious and it's kind of more of a framework for people that have not gone through the prioritization process. But the theory is that if you have, you can actually measure how many roles you should have active in a given point in time using the kind of monthly hiring capacity and time to hire measured in months. So if you, if you can hire 10 roles a month and it takes two months to hire an average role, you should have 20 roles prioritized at a given point in time. And that way you don't wind up with 60 roles and everyone's trying to figure out which one is the one that should get work. at a given point in time. I think it's good in theory that comes from supply chain where you... kind of like have to forecast how much inventory you should have in progress at a given point in time. But I've applied it in the past to TA and I found it works quite well for kind of assigning prioritization and not overloading some recruiters with more recs just because we know where they can handle them. I'm curious how you think about prioritization and especially given some of the things you said about shifting the values and shifting the specs that the recruiters have to hire for. with more people requiring stronger AI skills, how you prioritize and how you manage that kind of re-prioritization throughout the year. And do you use anything like kind of a rule of thumb or a Lethal's Law or anything like that to kind of figure out how many roles you have prioritized at a given point in time.
Ian Jones: it's a fantastic question. I would love to see like a, model that you have, because I think it be, it would be fantastic, but, we, do. So when we go through our, our rec approval process, we prioritize things in the approval process. So by the time they get to TA, we know, and we, kind of have a couple of different levels when we consider business priority. So we didn't want to go with like high, low, cause nobody wants to feel like their thing is less priority. But we talk about business priority. So business priority is saying, Hey, this is really important to the business unit. The business unit cares about this role, it's critical for them. Then we have a strategic urgent, which basically says, hey, like this is really important to the larger piece of HackerOne. This is a high leverage role. So we get those. The challenge is like, we always seem to have a significant number of those and we'll basically have different engagement levels for our TA team. have different engagement levels for each of those. So there's a couple of levers with our strategic urgence, whether it gives me increased flexibility that I don't have to ask if I'm going to go get external help on that. Like I can just go and do that. I don't have to go through approvals. It also means our recruiters priority. that role and then with the business priority stuff that gives our recruiter we have a different sort of SLA and we say, as a hiring manager, we have increased expectations for you if you want this to go fast, you are going to have to take some action and we're very clear on what those actions are, whether that's resume reviews or, you know, doing. just doing the initial screens yourself, we obviously have a different engagement model depending on the strategic, whether it's strategic urgent or business priority. So that's a critical one. And I love the idea of, think the biggest challenge that comes with the forecasting is the forecasting is really useful for us when it comes to repeatable positions. So if you look at our Pune roles, we have a couple of roles that we hire a lot of, or even like engineers for the most part. Like we generally know how long it takes us to hire an engineer. We hire a lot of them. So the prediction models work where we tend to struggle a little bit is like the one-off. Because we don't have that predictability and I could tell you it's going to take me 90 days to hire, I'm hiring a director of UX. So we have a really good pipeline and we can go all the way down that and I can be close to meeting my 90 days and that offer is 50-50. It gets rejected or accepted. And if it gets rejected, my whole timeline gets blown out. So think that's where we struggle a little bit. think that's the harder, but with the repeatable roles, like we feel really comfortable about what we're able to deliver on those. And over time, even though like, sure, some role I predicted 90 days might take 30 and then another one might take 120, but I love the sales one. And I think the part we've gotten nailed Chris is we know how long it takes for us from the time we open our sales role to the time that person's in seat. What I don't know quite honestly, if we're working backwards in the headcount plan to say, Hey, we need this person to be at a million dollars in revenue. Because honestly, if we need somebody to be at a million dollars of revenue this time next year, we should probably be hiring them now. And that's, that's the part where like the headcount plan is hard because ideally you're looking at it from a multi-year standpoint, but in this current environment where our industry is changing, like we had a key metric in our organization just explode, you know, 40 % month over month because of AI. And we've had to adjust our headcount plan to it. Like nobody's thinking about that two years ago. So that's the hard part for us is how do we be really strategic about planning it and, and, and, know, have a good model and be methodical use data. And then, like this thing that we didn't know about. I mean, who knew meet those was coming out? Like that's blown our whole world out. Like, okay, we have to now we have to change things. So I think that's always the hard part about it, but I'd rather start from a good plan and then have something that gives us visibility to the changes. Cause that's what crushes us. Right? If you don't start with a good plan and then you don't have visibility to the changes, you spend so much time on comms and alignment that that eats your ability to actually deliver.
Chris Mannion: We actually spoke a little bit about a concept called the OODA looping comes from military aviation of how quickly can you react to changes when they come in and how does that compare to your competition and the quicker you can react and repurpose roles or kind of reposition and actually get ahead of the OODA loop of the kind of the time taken to go through that whole reaction cycle of your competitors that the companies that react fastest win.
Eric Guidice: You are speaking my language.
Chris Mannion: And it sounds like that's how you're kind of approaching it with the prioritization and the forecasting and actually not necessarily having fixed plans that you will adhere to regardless of what changes on the outside and recognizing that the market's changing so quickly and it's the quicker you can make the decisions, the more likely you are to win on head count overall. Does that ring true or is there anything you're doing that has helped you to speed up that decision-making process?
Ian Jones: It does. mean, that means when we go back to our leadership, talent principles, again, change agility is one of the ones we hire against, right? Because we expect people to be able to change quickly. It's necessity. The hardest part of that is like, I can adjust quickly on things. I'm literally having to do that now. was on a call this morning where we were having to do a hiring sprint and we weren't anticipating. The problem is that part's easy. The hard part is the things that are going to have to get put aside to do that. That's the harder part of it. We have to go back and say, these things we were supposed to deliver, we actually can't do those now because we have to refocus there. How do you abandon the things that you've already committed to that were already part of the plan and tell somebody like, hey, you were part of the plan. You did this. You focused on getting your plan in a good place. You did it six months ago. Yeah, all that has now gotten sidelined. I think that's the hard part of agility. It's not the adjustments. It's... making the adjustments relatively easy. like, how do you deal with what you were already committed to? I think that's the hardest part of adjustment, is your point. Like, I think we can move quick, but the slowness is not being able to walk away from things you're already committed to. That's the really hard part. That's a great, call out.
Eric Guidice: So I think time to fill and a little bit of predictability about what's to come helps the business understand what is possible when they're building their future workforce plan. What are some of the other data points that you find the team has responded to from the TA world specifically when it comes to either what you plan to do in the future or consolidating or reporting on what you've done in the previous fiscal year?
Ian Jones: Yeah, I think a couple of really key things that have been helpful for us and one of them seems so silly, but it's important is taxonomy. Just like getting people aligned on the right words to use internally has been, I've found that to be super useful. So I think that that is the foundation. Before we start talking data points, people need to understand what, you know, what headcount means, what a position idea is, what a rec is, because we spend a lot of time doing that. So think that was step one. Definitely time to fill is a critical one for us because that's how people can plan. I think the other thing that I think we're having to figure out how to adjust to is on compensation side. I think traditionally and now that I've been on the people side understanding more about how a lot of the big salary surveys work, they're really good, they're really thoughtful, they're global, but man, they're lagging indicators. How do we deal with that? We're doing our survey now. We're sending our stuff into Radford and it's great. They'll take a couple of months, they'll do it then they'll put out a report and that'll be informing next year. But the problem is, especially with some of these roles in an AI world, that market's moving so fast. It's back to those COBOL programmers again. Radford would have told you what they made last year. I need to do it in the now. I think one of the things we've done is we've actually leveraged AI again where we've built a market mapper that's pulling real-time data from Glassdoor. That's one reality from Indeed, from LinkedIn, with the current salaries that people are hiring. What our competitors are hiring for. What the real data is behind that today. And we're using that to match that to help us adjust quickly so that we can plan in real time as things are changing. Like that's been a game changer for us to build that in. So we put in, you we put in everything from our leveling guides and job descriptions and said, Hey, what's this going to cost us in the market in this geography? And we use that to match that to help us make decisions about where we hire to adjust levels. We actually can't afford that. Go ahead and take that from an E1 to an M6 and tell me what the differences are. That has just been something that's been an absolute game changer for us because we're using real-time data. think that data point more than anything else is going to be really helpful in making sure we're being thoughtful about how to plan for the upcoming year.
Eric Guidice: I love it.
Chris Mannion: Yeah, I'm taking a note and I'm going to make a video showing how to do that. So incredible idea. think the kind of Radford point is really critical. It's almost like there's an opportunity for an AI first company to come in and actually almost have a real time compensation mapping that actually just anonymizes and updates in real time and gives you that, you know, what is this person going to need to make next week in order to retain them?
Ian Jones: Yeah. Yeah, and then what if you could build like, I mean Chris, you're infinitely smarter than me and more technical than me. So like, what if you could build something predictive, right? Says this is the trend, like this is where it's going. And so we've gone from Radford will tell you like what's happened and they do so in a very structured way and I respect what they do. We're now building tools that are telling you here's what's happening today. But what if you could then build something and say, hey, here's where we think this is where this market's going. How do we get ahead of that? Not only just for hiring, but for internal retention and things like that. That would be amazing. Because you've ever been through a comp planning process, it's like, here's your percentage that you get based on the budget and go use it. But even that, it's not as market driven. And the reality is we're all in the employment market all the time. You can think you have security for folks and then somebody picks a phone call, meets somebody, gets a call, and they'not. So we have to be thoughtful about retention. And I think the only way to do that is now we have unprecedented access to process all that data quickly and hopefully predictively.
Chris Mannion: That's fascinating. There was also another point I want to just touch on because we covered it with Reed as well. So you mentioned taxonomy and headcount. And one of the things we were talking about was headcount versus FTE. And then the bigger question is when you have agents within the workforce, they headcount, are they FTE, or are they software spend? And how do you think about tokens? That's probably a rabbit hole that I know Eric has some reactions that we're probably going to go through. Just in kind of simple terms, is there any way that you thought about that?
Eric Guidice: It's an active conversation in the space right now. I don't think anyone has it right. think there's a... It's productivity without people, but you still need the people. And so there's like, what is it? yeah, I'm more interested in the conversation too. Like I want your opinion. also am interested in like, is that even a conversation? Cause I think there's just like a general pulse out there. Go ahead. Sorry. But that is that, that's that Chris got me. He got me right in the, right in the buzzer.
Chris Mannion: Ha
Ian Jones: Yeah, honestly like The limit of the thought I put in was like, spent a lot of tokens. That's about it. So how is that considered like a capital expenditure? To me, that's R &D. To me, that's R &D. You're putting in research and development, you're spending tokens with the expectation you're going to build something that's going to have a huge yield. And it almost feels like one of the most amazing things about Amazon, which was fascinating, was I've been in so many organizations where they worry about like, are we doing a project that someone else should be doing? coordinate with them and get it right. Amazon had the totally opposite view. They're like, no, we want 20 projects going on to solve the same problem because the right one's going to win. And I feel like companies that are really AI forward are adopting that saying, yeah, we're going to spend a lot of tokens and probably build some stuff that isn't great. But along the way, we're going to build something that's absolutely transformational for our business. And that's an R &D expense. So I would say tokens are R &D expense, but I'm not. mean. I'm barely, I'm an interim head of people. I definitely don't know finance and how to categorize that. But if you were asking me, that's how I would do it. It is necessary R &D that hopefully will save us headcount spend in the future. Cause it's just, it's been fascinating to watch like. I came through the startup space in the late 2000s where it was just hire as many people as you possibly can as fast as you can. It was great for business on the people side, but it's like the opposite now. People are looking to hire, how can we hire as few people as possible and make them super duper zinc, but you're going to have to do R &D investment into the artificial intelligence stuff.
Eric Guidice: You just gave me two things. One, there's still what reads take was is that rush is still happening, but it's with tokens. And then the second thing I just tried to ping my finance guy on Slack while we're here is like, can you write off tokens is because there's an R and D tax credit. So if you're a, if you're a workforce planner, now this is a live thought by the way, but if you're a workforce plan, you're trying to decide where you invest dollars. There's a tax incentive from an R and D headcount spend that I'm not sure exists on the token side. More news at 11. We have about 10-15 minutes left. Chris and I do a segment where we pull up some of the stuff from the past week, past month, and we try to react to it live. You want to play that game with us? Okay. I'm going to pull something up here. Maybe you've seen this. I'm going to start with the big one.
Chris Mannion: Okay.
Ian Jones: That sounds like fun.
Chris Mannion: You
Eric Guidice: Maybe you've seen it, maybe you haven't, if you're in my zone window. Here it goes. Article number one. Work days last work day. Have you read this article from A16Z? It's about the decline in their stock position. It's about the change in the way work is happening and what is the future of work day. Chris, you see this one yet?
Ian Jones: I have not.
Chris Mannion: I saw George LaRocque's take on it. He runs WorkTech and it was interesting. I have a strong opinion on this one, but I'm curious to get Ian's before we kind of dig into that.
Ian Jones: I might have a little more conservative approach than other folks on that. I learned a lot of management from my dad and my dad ran backups, backup recovery systems for Merrill Lynch. And back in those days, everybody was excited. Linux was just coming on the market and he ran big Unix systems. And his response was like, Linux is fantastic until it breaks.
Eric Guidice: This is the
Ian Jones: like, I use Sun Solaris and I use it I have a vendor that can support that. So, he had to think about things from a security standpoint. He'd think about things from an uptown standpoint because at of the day, when 9-11 happened in New York, all his primary servers got fried. They were in World Trade Center 7. He flipped a switch and he unfortunately wasn't able to retire because they kept him for another 10 years after that because it worked so well. But kind of being like, I feel like of all the systems that are going to change, I'd be surprised if Workday will be one of the last ones because of the the nature of the data that's in it because it's so like integral to your organization. I'm not saying it shouldn't change. I'm just saying I think it's going be one of the harder ones to change because you're going to be a little, people are going to be a little less, more, more, I'm going to be risk averse when it comes to that.
Eric Guidice: I think the conversation, the take on LinkedIn is that this enterprise behemoth now has competition. The barrier to compete is now drawing against the number of customers who may or may not stay with Workday or add into Workday services. I think we've noticed the sasspocalypse of some of these bigger players. And the article is super interesting. I highly recommend a read. It's from a partner at Andreessen Horowitz. His name is Joe Schmidt. Follow Joe Schmidt on Workday. He's got, from Andreessen Horowitz, he has a bunch of interesting content. He's an operating partner. He's a venture partner and it's a very interesting perspective, especially when it comes into our organization to see I think the people and talent world is relatively insular, and so when you get an outsider's perspective, it's always fun to hear the reactions from the real world. The next post is from, I want to pronounce it, Luchi Gomez. sorry, Chris, I...
Chris Mannion: Yeah, quick one on that. Yeah. just like a point that I feel strongly about this. I think the thesis of A16Z is that their whole business is investing in software companies, make them really big and then get their money out. Right. So their thesis is a new software company is going to replace Workday. I think they're not really thinking big enough. I think that what Workday does is it solves the job that is handle all my HR admin for me. And I think that the startups we're seeing now and not necessarily buying software to do that. They actually want to outsource that. So I almost think that the opportunity is not an AI driven SaaS, which is what their thesis is. I think it's an AI driven agency or HR as a service so that companies can focus on their go-to-market and their product and actually have all that work done for them by essentially a back office. might be wrong there, but I think if you're going to replace Workday, you probably... It's probably not a new software platform. You probably, to Ian's point, going to keep it for as long as possible. And then when an alternative comes along, you'll just hand that over to this company, kind of like what we've seen with a lot of IT. A lot of IT is offshored or outsourced. That's my take, at least, but I feel quite strongly about it.
Eric Guidice: Right. as the software, you know, trying to nip at their heels and prevent people from moving to an ERP to connect their systems, I hope that maybe that is the case. You know what I mean? That said, next article, Luchi Gomez. I'll give a share here so that you guys can see. Why won't it let me share? Hold on. This will be edited out. There we go. Okay. Alright, okay, so the next article from Luci Gomez. She is the VP of talent acquisition over at GitHub, formerly at Microsoft and ADP. Great content. Luci, if you're listening, excellent content. I followed and liked a bunch. What she's saying is there's a metric at the intersection of TA and the business that no one really owns and it's time to productivity. I think this is just calling back this workforce planning concept of like when does that person in the seat actually, you to come available for the business and making a hiring decision off that. Ian, what are your thoughts?
Ian Jones: Yeah, mean, Chris has got the right idea on this, right? He's thinking backwards from that. Yeah, I think it's unspoken. I think we don't talk about it, like we don't bake it in, but she was like, yeah, I need somebody by X date. And they're probably doing the math as to like, they need them to be productive by then. But I don't think it's as organized or as structured as it should be. But if you want to have a good workforce plan, that's what it should be. We have X deliverable in sales, it's super easy to understand. but an engine product, have roadmaps too, right? So it's really like workforce planning should start from the roadmap and then you build the people into it. I think that would be an ideal situation.
Eric Guidice: Chris, any thoughts before we move on to our next?
Chris Mannion: Yeah, think just concurring with Ian there, I think he mentioned it earlier in the podcast. You want a sales rep to be at a million in room rate by a given point in time. If you want them next year, you probably have to start hiring them this year. And I think that the annual headcount planning cycle actually misses that whole point. I was thinking this kind of rolling. cycle that we've discussed a few times where you're kind of forecasting ahead, but because you don't look in the budget and the revenue forecast and everything else, these kind of like 15 month ramp periods just get missed in this process. So I think there's, it's a good push.
Eric Guidice: I love it. Okay, so let's get to the next article. Hold on, I'm gonna give a pause here and then I'll start my little speech. We're here. Okay, our next article is from Roald Harvey. He has a background in cognitive psychology. He's also been a head of people at various startups across the health tech sector, and his take is AI agents in the org chart is one of the most controversial moves in the past, but now it's a necessity. What is this idea of synthetic capital? and it's the total cognitive capacity or maybe even productivity of your head count and looking at the digital worker. So I'll give you guys a second to take it in. It's one of the more complex articles that we've shared or posts that we've shared. What's your take?
Ian Jones: Here's my hot take is head counts already complex enough. To me, I would rather just bake that into the expectations of workers, right? Like the expectations of workers is increasing what we're asking them to do. We're asking you to let be less of a doer and more of an orchestrator. So I'd be much more interested in, you hiring somebody on the TA side who orchestrates all of our agents and doing that rather than building the agents into the plan. It's hard enough and I don't know how to do that, but maybe Eric, your software can do it for us, but I don't know.
Eric Guidice: There's a concept of productivity. think what we're seeing in our world is fewer people create more outcomes. so tying the outcome into the forecast is a way to keep the flexibility of how much headcount changes with a higher performance standard. And so it's just this relationship of productivity plus cost. once you start just like, know, I might be approaching it as a generalist, rolled in cognitive psychology and being a director of people having this focus insight. Now you're getting into the nuance which, Chris, you have to have a thought. Normally I do this just with Chris. He loves his stuff. What do you think? What's your thought?
Chris Mannion: Yeah, I get confused easily. when there's a lot of different kind of terminology and a lot of variables, I try and condense it to as minimum as possible. So I actually think I agree with Ian's point. Like I think capacity is a good enough metric for most of us and the automation and agentic workflows and everything else that go into increasing capacity for the individual, I think is like the perfect way to think about this. So what is the capacity of like how many widgets can a person make in a week? And that is how you create your capacity plan. think thinking about synthetic capacity, adding on top of that just complicates things and makes it harder, especially given almost on like a monthly basis, the models change or the workflows change or something new comes along. So you're to have this variable capacity that's going to increase over time, but you don't know by how much sitting on top of your human capacity, but you can only unlock that synthetic capacity if the humans can actually do it. So. I think it comes back to like, what is the capacity of the individual, the unit, the single unit person? And then how does that change over time? I don't think we need to muddy the waters anymore. You're muted. You're muted, Eric.
Okay, my last post here. I just don't want to hear the clicking of me pulling up the next thing. I mute my phone. Anyway, back in the character. All right, the next and last post, I know we're coming up at time, is from Bremlee Villasenor. He is a software engineer who landed in the people space. He took... the meanderer, the people analytics, the certification, the applied AI for HR taught by none other than Chris Mannion, fan of the show. I wanted to find a cool way to bring him into the podcast, him a shout. Appreciate all your fandom, Remley, but Ian, what do you think of engineers and talent and taking courses from Chris Mannion's academy to better their skills?
Ian Jones: I think it's going to be a must, right? Like not trying to do a plug, I have two of my, two of people on my people team just took courses. Chris, I'm sorry. I don't know that they took your course or a different course. I apologize for that. If I had known we'd have done it, but yeah, I mean, again, this is the expectation, right? Like this is the expectation going forward that you are, there are AI fluent. You are using this to increase your capacity 100%. I'm all for engineers coming in, right? They're solutions folks.
Chris Mannion: Ha
Ian Jones: It's one of the things I tell my TA team all the time, if you work with the engineering team, if you don't give them clear instructions, they're going to make up their own solution. Yours is probably going be better, so you better do it. And it makes them be really clear. So I'm all for diverse perspectives in the space. Give me the engineers, give me the finance folks, give me the creatives in the space. Let's make it all better. Because I think we need to rethink people in general. And I think as a people person, we have certain perspectives on it. But man, this has never been a time to put more people in the conversation because we can blow the whole thing up. I think we need to do it. But Chris, yeah, let's talk through the courses because this is must for everybody in people.
Eric Guidice: Chris? We really the courses are great. And if you haven't seen them, they're summer free summer are paid for. They're all shurm accredited. I think it's awesome to have creators in the space helping to elevate. We've seen a couple of job postings. know Joe Atkinson from seed or purple wherever you know him from has been really hot on tracking the talent engineer. This kind of evolution of recruiting operations. So as you see engineering and talent converge into a space where they need to AI in these real time situations to help leaders like Ian, better workforce plan. It's awesome to have a provider in the space at your level creating this type of content. So we thank you for it. Ian. We're right at time. This was an awesome conversation. I'm glad you got to have the floor for as much depth as you did here. And where should we send people who are interested in learning more about you or learning more about HackerOne? Where do you want people to go to learn more from you?
Ian Jones: Yeah, I mean, I'd love for them to go to HackerOne and check us out. It's such a really interesting space. Cyberspace is moving so fast, so you can check us out there. I'm probably not as active as I should be on LinkedIn. So yeah, if anyone wants to get an insight, they want to ping me on LinkedIn, tag me out, I'm than happy to chime in. I just haven't had a lot of time given the space for some original thought, but I probably should build an AI bot to create some content. So let me...
Chris Mannion: Hahaha.
Eric Guidice: This was awesome. I love the conversation. Chris, close us out.
Chris Mannion: That's awesome. Yeah, I appreciate that. Thanks for confirming, I think a lot of the things that Eric and I have been talking about for the last nine months. I this was really helpful and best of luck with Hacker One It sounds like there's some exciting times coming ahead.
Ian Jones: Absolutely. And thank you both. This has been great. I genuinely enjoyed it and got some good takeaways, some things I need to follow up on, some things to learn. Cool.
Chris Mannion: Awesome.
Talent Acquisition’s Perspective on Workforce Planning
Traditional headcount planning frequently relies on lagging financial models and static compensation benchmarks, introducing severe payroll variance and delayed hiring cycles into the business.
In Episode 20 of the Headcount Experts Podcast, Ian Jones, VP of TA and Interim Head of People at HackerOne, joins hosts Eric Guidice and Chris Mannion to break down how to inject real-time talent acquisition intelligence directly into the workforce planning process. From building automated market mappers to deploying compliant AI orchestration frameworks, this episode reveals how forward-thinking talent leaders protect their talent management strategies from volatile market shifts.
Moving Beyond Static Financial Models
Workforce planning often collapses when top-down financial spreadsheets ignore the operational realities of the talent market. Ian Jones highlights that standard headcount plans often assume all hires miraculously start on Day 1 of the fiscal year, creating immediate friction between budgeted spend and actual hiring velocity. To build an effective headcount management model, organizations must integrate true recruiting pipeline data directly into their corporate strategy. Empowering Talent Acquisition Leaders with a clear line of sight into business goals is the first step toward reducing forecasting error.
Real-Time Compensation Mapping
Incorporating Pipeline Realities: Headcount plans must account for complex variables like localized notice periods and historical time-to-fill metrics broken down explicitly by role and level.
Overcoming Lagging Salary Surveys: Traditional compensation benchmarks, such as Radford, act as lagging indicators that fail to keep pace with fast-moving modern tech markets.
Deploying Real-Time Market Mappers: To counter data drift and protect talent retention, HackerOne built an internal market mapper pulling active salary data from Glassdoor, Indeed, and LinkedIn to adjust leveling guides in real time.
Capacity Orchestration and Little’s Law
During the episode, co-host Chris Mannion introduces a supply chain principle, Little’s Law, to optimize open requisitions and prevent recruiter burnout. By calculating monthly hiring capacity against average time-to-hire, Recruiting Operations teams can mathematically define the exact number of active roles a recruiting organization can successfully manage at one time. This disciplined framework prevents companies from opening downstream requisitions that outpace their actual capacity to execute, preserving the accuracy of their single source of truth.
Tiered Prioritization Matrices
To execute an agile headcount planning framework, HackerOne moves away from arbitrary "high vs. low" designations, opting instead for a clear, two-tiered business governance model:
Strategic Urgent: High-leverage roles critical to the broader corporate transformation that automatically unlock increased structural flexibility and rapid external resource allocation without friction-heavy approvals.
Business Priority: Standard roles critical to specific business units where hiring managers must meet elevated Service Level Agreements (SLAs), such as conducting initial screens themselves to maintain hiring velocity.
AI Guardrails and People Data Security
As organizations transition toward an AI-first talent structure, protecting highly sensitive data across disconnected systems of record is non-negotiable. While engineering managers may attempt to build unauthorized custom GPTs to evaluate resumes, talent leaders must strictly enforce legal compliance boundaries to prevent biased models from informing hiring decisions. Real efficiency is achieved by automating management communication workflows rather than exposing privileged candidate and payroll records to unsecured models. To streamline these operations securely across your enterprise, explore our comprehensive Workforce Planning & Strategy framework.
Featured Headcount Content Creators
Joe Schmidt - a16z | Workday’s last Workday
Workday has historically maintained an unassailable moat through deep operational lock-in, a massive AI-driven replatforming cycle is opening up a historic window for true AI-native HR systems of record to displace legacy, forms-and-approvals software infrastructure.
Luci Gomes - VP of Talent, Github | Time-to-productivity
Time-to-productivity" is a critical yet overlooked workforce planning input that talent acquisition must formally own, leveraging AI to bridge hiring and performance data to accurately forecast hiring cycles and optimize employee ramp times.
Roald Harvey - Sr. Director, Athennian | AI Agents in the Org Chart
Formalizing autonomous AI agents on the corporate org chart has shifted from a compliance controversy into a management necessity, requiring leaders to track "synthetic capital" within the HRIS to accurately measure total cognitive capacity, enforce human accountability, and optimize resource allocation.
Bremly Villasenor- Data & Reporting Specialist, Manulife | Applied AI in HR Certification
Announcement of Meander’s Applied AI in HR certification from Chris Mannion’s Talent Academy, Meander.
Sign up for Meander’s Applied AI in HR Certification Course Here: https://learn.meanderhq.com/courses/applied-ai-for-hr