FullStory VP of Recruiting Jim Miller Interview on Shally's Alley

I stopped doing that. I stopped trying to track it. So anyway, here we go. And going live. All right, welcome everybody. It's 1 PM and it's Friday, so this must be Shelly's Alley. Well guys, I am super excited today to introduce my first leadership guest. With me I have Jim Miller, and I hope you can see him on the screen. Let's see if that's possible. I think I see just myself. Hmm. Not really seeing you on the screen there, Jim. Well, while I figure this out, let me tell you a little bit about Jim. So Jim and I go way back. He was at Google for 15 years. I have to say that's probably some sort of a record. I don't know any other recruiter or, you know, in the talent acquisition sphere that stayed there that long. So 15 years at Google, you ran tech sourcing for EMEA for a while, then you ran global talent channels that included the online strategy, branding, content, job advertising, all of that for Google. And you moved from London to California Is that right?

Yeah, that's right.

Okay. And then you, um, recently— I don't know how recently, but I hope it's okay to share— you recently moved to Atlanta and you are now the VP of recruiting for a company called FullStory.

That's correct.

So without telling me the full story, tell me a little bit about that job and that move. So what do you, you know, what are you doing now?

Um, so yeah, for just over 3 months I've been the VP of recruitment here at FullStory. FullStory, if you don't know, are a digital experience company started by a bunch of Googlers from the Google Atlanta office a fair number of years ago. And I've joined to help them with the hypergrowth that they're going to be going through in the next few years. So very exciting.

So now you're in Atlanta, my hometown.

Yep.

That's exciting for me as well. You know, there's a lot of good talent here in our industry. For a number of reasons. So glad to have you here, and I hope I did justice to your intro. But before we get started with the, with the very invasive questions that I have planned for you, I had one question that I don't know if you were prepared to answer, but I'm just really curious. Your LinkedIn vanity URL is rather unique. It's different. For those of you that can't, you know, obviously you can't go find them right now. If you type in Jim Miller, and full story on LinkedIn, you'll probably see him. And you'll see that his LinkedIn URL is, uh, you know, the typical linkedin.com/in/ and then it's J M and 20 M's. So it's like J—

it's J-I-M and 20 more M's or something to that effect.

Is there a reason for the— I counted 20. So is there a reason for the number?

I used to be, um, jim.miller@google.com, written longhand. And then some stuff happened with privacy and things like that, so I had to change it. And of course, Jim Miller is a very popular name. There's a lot of us out there. There were 5 of us also at Google when I was there. So I just changed it, and I just put my finger on the M, and that's what it produced, and stopped at some point. I didn't even count.

Wow, so you ended up accidentally hit 20. I was thinking maybe that's like one M for every year in recruiting, you know, like stripes on a soldier's uniform or whatever.

No, nothing, nothing at all. It's funny how many times I've been asked what the meaning behind it is, and I just happen to put my finger on the M and just keep going and, uh, lifted it up, and that's what I got left with.

Perfect number, like just perfect round number.

Yeah.

Now you're going to have to edit it every year and add another M.

Now you've told me that it coincides with the 20 years I've been in recruiting. Yeah.

That's just me, man. I try to find patterns in everything. I'm like, I don't know, it's my— I've got ADHD and OCD and together they're really, really bad for each other. So I obsess and hyperfocus, right? And so I always try to find patterns and stuff. And That may be good if I was like an astronomer, but it's not good for like most normal day-to-day things. But that's why I picked up on that. So let me jump into— you said something very interesting, and I want to— I almost want to just focus on that unless we exhaust the other topic. You said when I sent you a couple of questions that I wanted to start with, you said you do not have a sourcing team. And you do not intend to have one. Having been a sourcing team leader, founder, creator, the whole thing, tell me more about that. I'm very curious, and I think I know why, and I think I agree with you, but I want to know the full story behind—

Absolutely.

Why you intend to not have one.

Um, so yeah, you're right, I have led sourcing teams across all sorts of different disciplines before. And now I'm leading full lifecycle recruiting across the board. I thought really deeply about this, and the, the thing that I've experienced over time is that you lose some of the connection with what the job you're hiring truly is with the number of handoffs of responsibility in the recruitment lifecycle. And I've observed this kind of finger-pointing and conflict going on behind, you know, the recruiter who says, I haven't got enough pipeline, sourcing let me down. And I'm like, all right, well, can we think about this in a different way? So my intention and my thinking is to hire recruiters who are amazing sourcers. Now, the other part of the recruitment life cycle that I'm not talking about here is the online inbound you know, advertising responses and referrals, that I do intend to split out and have separate so that any— if I have more than one recruiter on a particular role, I'll be able to distribute the inbound equally across that workforce. And then each recruiter can define their own success through their ability to source and find the talent for the role they're working on that they know so well from their relationship with the hiring manager, and they understand deeply, rather than that person who understands the role, handing over to a sourcer who might have a disconnect, and they then have to work really hard to get calibrated and all the conflict and the tension that that produces. If I just hire the same skill set in one person, they go and find, they elevate themselves, they're in control of their own destiny, and my intention is that it will be that much sweeter.

Yeah, so I knew we agreed on this, just because basically the way you answered the question before, and I knew that this is where you were going with it. Interestingly enough, I've always, you know, I've been doing this for 25 years. I've always maintained that sourcing is a discipline within recruiting, that to be a sourcer, you must first be a recruiter first, with the emphasis on first. In other words, where I diverge from the majority of people in the industry is I don't believe that a sourcer is a baby recruiter that wants to someday grow up and be a full-cycle recruiter. I believe a sourcer is someone who's already done the recruiting and is learning to specialize. Like in any other industry, we have specializations in intellectual property, in marketing and sales. You don't have an HR person anymore. You have people that do employee relations, compensation, payroll, right? So I believe recruiter first, then sourcer, because recruiter first can also mean then social media, then recruitment marketing, then, you know, a talent advisor, all these other specializations that exist. So I totally agree with you. And I also think that we've created this— I don't know what you want to call it, but what you described as tension artificially because of that environment where organizations hire people that are junior and say, oh, this is an entry-level job. You know, you'll be a recruiter someday. And the recruiters are challenged by that because they don't want to use these leads that the sourcer generates because that means they're not doing a good job. And it just artificially creates this tension. So I totally agree with that. However, here's where we diverge. We diverge in that I believe that there's a way to manage that so that there isn't that tension. And I've solved for that, but it's not easy to do. And it requires that you have a leader that understands both sides, the sourcing side and the recruiting specialty, and the mainstream full-cycle recruiting. But it also requires an organization that's capable of absorbing the leads. And that's where there's a problem. In other words, you can be a great leader that inspires sourcing and recruiting and gets them to work together, but then the organization just simply doesn't know what to do with, with half of the talent that's coming in that's being nurtured. And then we end up treating passive candidates in a less than ideal way. So there's— you're solving that problem one way, which is let's just not fight about it. You source for yourself. And you're also solving that problem another way, which is basically what I call pipeline development. And that's one of the things that I've been chartered to create here at Aspirant. Our sourcing-as-a-service solution that I told you about before is research in the triage sense, right? Like panic button, we need 100 leads now. We can do that. We can jump in, we can get it, we hand it over to the recruiter. We're not recruiting, we're just sourcing, just research only, right? But the pipeline that you're describing, that is a long-term investment. So if you actually are able to build something like that where it essentially is a self-subsisting sustainable advantage that runs and brings your, your, you're generating inbound traffic to recs. So the recruiters don't even really understand or know that that's happening. In other words, that sourcing team, which I don't know if you have a name for it, the pipeline team or whatever, is practically invisible to the recruiters, right?

The recruiters should always be agnostic as to where the candidate is coming from. They shouldn't have any preference over channel, whether it's inbound, you know, referral or anything else.

Obviously, to them it shows up.

A lot of recruiters have a preference over referrals, which is a whole nother problem. But yeah, I hear what you're saying and I appreciate the value of the research generation in feeding a group of recruiters who are doing that.

That's right.

Full life cycle recruiting. Um, that I agree with. The engagement with the candidate should be the person who owns the position, in my opinion.

What do we call that then? So, you know, so we've, we've— it's, it's too late now. The, this, the job title sourcer has been permanently damaged. What do we call this job that you're describing, this pipeline You know, invisible, agnostic to the recruiter, invisible to the recruiter where they're bringing in leads for all roles of the organization. What is that job title or what is that team called? We're gonna come up with a name today.

That's not my forte to coin anything in that space. The first things that came into my mind were talent attract, and that was because of something I'd seen recently. But yeah, there's so many nuances to it. And it just has to meet your needs of your business. And it depends where you are in the scale and the volume and your type of hiring. It doesn't necessarily work very well if you have a lot of unique positions.

Yeah, it doesn't. That's right.

Profiles that you're hiring at scale, then that model can work very well.

Yeah, I like to refer to them as multi-incumbent roles. So where you've got the same, you know, either the same position or very similar positions where it's more like a profile that works really well. And it is, there's a long tail to it. You know, you don't just turn it on, turn it off. It doesn't start today and then tomorrow, you know, it takes weeks or months to really develop that, but you have to have roles for which you're hiring fairly regularly, if not constantly, for that to work. So, um, I'm— I suck at naming things too. When I first started doing this whole thing that is now called sourcing, I called it cyber sleuthing. I remember. And I had the Cyber Sleuth newsletter.

You did.

And, um, when the, you know, when the internet was just a baby and they had just put fresh diapers on it, um, cyber sleuthing was, was a very unique term that if you put it on a search engine, you wouldn't find a lot of cyber sleuthing out there. But fast forward a few years later, you know, come 2000, 2002, 2003, you would type in cyber sleuthing and you would get all these people that are doing like detective work and skip tracing. It's completely not what I was doing, so I stopped using it. But by then, this sorcerer term had already sort of evolved. And I just went with that. I think it's time that we— I think we need to, you know, sign a petition to change the name or evolve the role. So that's really interesting. Another question that I wanted to ask is talking a little bit about the horizon of technology. Do you see, or what do you see is gonna be disrupting technology for us in our industry? It doesn't have to be a CRM or an ATS, or, you know, what technology do you see that is going to disrupt the way things are in recruiting and sourcing today?

I think machine learning is gonna be front and center of it. Having come from where I worked for the last 15 years and observed some of the work that goes on and, some of the ways people are thinking about the future implementation of it. And I wasn't privy to the internal intellectual property there. So I can talk about—

Big picture.

—how I see it being used. Yeah, the big picture in recruiting. I think it is very effective at crunching huge amounts of data accurately. That's right. And at some point, that is going to be applied effectively to big corpuses of data. Be it a product, a standalone product that will come to your own applicant tracking system and look at your database. You and I both remember where a company's competitive differentiation was the strength of their database and how many candidates they had on it, right? Or whether it's a company like LinkedIn who invest heavily in it so that instead of sitting and writing your search, you're pumping in some information. It might even be a raw job description. And it starts to throw you back the talent that it can intelligently figure out is right for you, the role of this will very much change at that point.

I would really like that you called it— Yeah, exactly.

—the most important role within recruiting.

Writing the job description, which unfortunately is really not, in a lot of places, a function that they allow recruiters to do. So, um, you know, and you end up having to live with whatever job description you get, and you can advise them on rewriting it, but very often I see recruiters don't have the— they're not given the power to actually really rewrite it, which is sad. I really like that you said machine learning, and I'm really glad you didn't say artificial intelligence, because artificial intelligence is a bunch of bullshit. It doesn't exist right now. We call it artificial intelligence, but it isn't. It's what you just described. It's machine learning, it's deep learning, it's neural networks, you know, those kinds of things. So I really like that that's what you called it because that's exactly what it is. And I, and I see the applications of machine learning pervasive throughout a lot of the things that we do, more so with the engagement piece in my mind than with the identification piece. The reason is that When you create a job description, often the language that's in the job description is the language that's described by the hiring manager regarding what they want to see in someone. So I want some person who has these experiences or skills or has this, you know, on their profile versus when we look at someone's profile, especially now that LinkedIn has essentially permanently changed resumes, that that's written in a different language, same words, different sentences. Different phrases, different, et cetera. So the pattern identification that's on a job description is always going to have a bizarre disconnect with the pattern identification that's on a resume until the machines can actually detect that difference, which isn't happening right now but will in the future. But right now, the machine learning bit can be done in chatbots and micro chatbots, which I think is pretty cool. Sorry, you were going to say something.

Yeah, no, I mean, I think to a certain extent you're right, but the smartest companies are taking that job description, yeah, using a specialist team to actually do what we've always done as sourcers, which is synonyms of the words, the key terms, the experience that the hiring manager has stated, and turn those into Boolean logic searches, right? That's how we— that's right. Cost, programmer, software engineer, developer, coder, et cetera, as an example. But if you're developing your job description with the thought of, I've got to attract all of the talent based on how they write their experience into their searches for a job, which is ultimately exactly the same skill set that's— then that attraction engine that you're creating within your organization to write your job descriptions effectively becomes the thing that a tool could read and then take and run across a database in order to find the information.

That's how you connect the ontology with— and the lexicon with the job description. And unfortunately, that's the piece that's missing now is that the lexicon is different. But that's beautiful. That's what I would say.

The thing that has always been pushed that I've been seeing a lot of is the resume screen. Being done by a machine. Yeah. And that's actually what I think will be the last part of recruiting. Yeah. And that may well have to wait until true artificial intelligence.

That's right. Because you can't—

The learning piece is based on your historical hiring. And very few companies have got unflawed historical hiring practices. So bias is being built in from the beginning.

Your bias, the machine is in fact biased.

And the machine picks up on that. And it's going to be the lawyers who stop the machine learning screening happening because you can never get it to 100% unbiased. And that's a positive thing in my mind, that it will still enable humans to do it. And we can work with humans to coach for the unconscious bias that exists. That's right. Work on the language and building out the profiles that we talked about to enable people to bring in transferable skills and hire across a much broader labor pool.

That editing of the language is also happening at the level of gender neutrality and ethnic neutrality, and I like that as well. So that's going to have an effect, a sanitizing effect, a hygiene effect on our job descriptions, making them more accessible to people that have been marginalized. So I like that. Interestingly though, what you just said about the last thing to go. I, I encounter these kinds of examples all the time, but let me give you one that literally just happened today. So I had a recruiter that asked me, hey, Shelly, I'm stuck. This hiring manager, I've sent them a bunch of resumes. They don't like any of them. The resumes look exactly like the job description. You know, can you help me? This is not a rookie. This is someone that knows what they're doing, but they're just pulling— they're frustrated. So they come to me and they ask me this. And I say, okay, send me a couple of resumes of people that they like, that they you know, that went to like final or semi-final interview. Let me pull those together and see. So they sent me 3, and this is for a DevOps position. Interestingly enough, as I looked at the profiles, job titles all over the place, keywords all over the place, the only pattern that I found that a machine would not have found is all 3 of them had come from economics. They all had econ in their background. So they were either undergrad, you know, bachelor or master's in economics, not computer science. But the job specs say computer science all over the place. Must have a computer science degree, blah, blah. The candidates that they liked, all three of them, had made major changes from economics to computers, but not DevOps. One of them had gone from econ to like cost analysis. And another one had gone from econ to sales to systems analysis. You know, these bizarre career paths, which to a human makes sense because if you know anything about DevOps, you know that it's kind of a mishmash field, right? There's not a clear path like product managers. There's not a clear path to, you do this, you get a degree, you do that, you know, you're a product manager. No, I mean, you know, so that is something that if the machines identify, they'll go off in the wrong direction because now all of a sudden they're only going to grade econ degrees. And if they don't identify it, they completely miss it, their machines are going to be providing false positives just as much as humans do.

I would argue that the recruiter has got it wrong in the in-between stage. So the hiring manager, I mean, how many places are we dealing with which have got hypergrowth, right? So you're in a company, hypergrowth going on, most of the managers are relatively new and don't really know what they're supposed to be hiring for. You don't know how robust the hiring manager training is and so on. So the hiring manager says what they want and they think, but they pick up a job description that somebody else shares with them and they don't necessarily go into the detail of it and change the, the language and add in their own personalization to it. It's to what they truly want. There's a little bit of laziness going on there. The recruiter, if they pick up I get a new job description from a relatively new hiring manager, to me, the first task I want them to do is to go and analyze everyone that's been hired before for that role.

Yeah, I love that. Good.

And to do pattern matching that you're talking about, because that's going to throw something up. And I'll give a simple example, the MBA. Hiring manager says, "Candidates got to have an MBA." Right? Okay, fine. You go off, you do your analysis. No one we've previously hired has had an MBA for this role. Why did you say they have to have an MBA? Oh, really? I thought everyone at the company we work for had to have an MBA. And there you have the surprise moment. So they clearly don't. Here's the data, here's the evidence to change your mind. Brilliant, we don't need the MBA, thank you. You've just made your job easier as a recruiter in hiring this role because you've taken out one of the key requirements, which truly wasn't needed.

Which was misleading. Exactly.

So you go through all of this and then you compare your results and you talk it back through with the hiring manager with a review of the job description. You find that half of the job description is incorrect and you can correct that through the power of data. And that persuades the manager to your point earlier of the recruiters not having the opportunity to write the job description.

To write the job description.

Now you can, because you've got the evidence and the power to change their mind. You get the job description right and it's accurate. You then put that out there. And of course, by doing that, you're going to have the inbound candidates. You might not even need to source on it. Let alone start thinking about machine learning or another methodology further down the line. So that's how I think about a problem along those lines. Your recruiter had the right instinct in going out and asking for help and doing some analysis. But actually do it before you start working. And you will have such stronger results. Totally agree. Yeah.

Work with a bit more purpose. Yeah. And that is something that probably requires a lot more work upfront. As you start doing that. But once these job descriptions have been, you know, modernized or whatever, that work is reduced because the newer, you know, more modern ones will have that already built in.

It pays itself back really quickly, I find, as well, because one of the outputs of an analysis exercise like that is, here are the companies we've hired from before into this role, and here's the job titles those people used. And immediately you can jump onto LinkedIn and run those pattern searches and LinkedIn will then throw up all of the skeleton profiles, the people who don't have fully fleshed out information that you wouldn't find when you're running your Boolean logic searches. That's right. Those profiles, especially in technical roles, because tech stacks are consistent across a company, those are your gold because there's no other source or recruiter finding those candidates. So those folks with no information, they're not getting reached out to all the time. They, they're possibly going to be surprised that someone has contacted them on LinkedIn because they have very little info. But you can see from the context of the resume of the person that was historically successful in getting that role exactly what the job— the person doing that job title at that company does. So you already know their skill set and you know the relevance, and you can dive into that pool, find candidates, contact them, and you've got an immediate head start on where you would be if you started working blind without doing that kind of work and analysis upfront?

Ironically, that is— well, maybe not ironically. Maybe, you know what they say, right? Fools seldom differ, right? It is exactly what I do. And it's exactly what I teach my people to do is to go out and do that analysis, look for the companies and the job titles of those companies. And then, you know, where they came from, what was the job title before that and the company before that. There's cool tools out there for that, but there's actually also a human— just think it through and you'll find that process that people just forget to do.

One other little pattern that I find is absolute gold with that methodology is what was the job title of the referrer of any referral candidates that got hired? Because you'll start to see a pattern of the job title. So it's not necessarily a software engineer referring software engineer. It's the designer, the UX designer referring the product manager. It's the brand person referring the marketing person or the salesperson. And then you can just go to that team inside your organization and say, here's the data that suggests that you people have in your network the people that can do these open roles. Give me your referrals, please. It becomes compelling. And again, you're working smarter, not harder.

That's smart. Yeah, I love it. Tit for tat, I'll give you another one. So, you know, when, um, when LinkedIn went to Microsoft, Microsoft quickly got started working on how are we going to leverage this technology outside of what, you know, they originally intended to buy it for. And they, they actually included in the, uh, Microsoft Word infrastructure. So on the— I don't know if this is the most recent update or the second most recent update, but I just recently noticed— if it's older than that, I just recently noticed because I don't do this very frequently— when you go into Word and you start a new template, one of the choices that they give you for templates is resume. And now across the top of the— what do they call that— the ribbon In Word, in Microsoft Word, there— this used to be a template, but now there's a button that actually says LinkedIn Resume Helper. It's no longer, um, buried inside of, you know, like a typical Microsoft, right? They create these great features and then they put them in a menu under a menu under a menu under a menu that you have to like search for. Um, so, but now there's actually like a resume button. Anyways, here's my freebie for the day. If you go in there and you click that button and you pretend to write a resume for the job description that you're trying to post, it brings all of that data back because it shows you all of the samples that match, and it's actually grouping those together to give you recommendations. So you see the bullet points that LinkedIn is recommending for that job title. How cool is that?

Copy, paste, plug it into LinkedIn and go and find all the people that have got it in there. It almost makes me want to go out and get a Windows computer.

Well, you don't have to have a Windows computer for Microsoft Word. There's Microsoft— I have a Mac and I have Word on the Mac.

It's been a long time since I've had a Mac as well.

What are you using then? Good question.

I've been, I've been on Chromebooks for years.

Oh, okay. And, um, the Chromebook has, uh, all of the built-in Google Docs. Yeah, yeah, right. I— that's I'm glad you mentioned that. My kids actually have Chromebooks because the school provided them with these bizarre, I don't know, they're laptops that are very— let's just say they look expensive, but they're not. All right? And the kids just hated using them. And so I was like, all right, what can we do? And I went out and I bought $250 Chromebooks for both of them and they love them. But the school uses Google School or whatever it is with Google Sheets for School, Google, you know, it's all like a whole suite. So it's just tied in together. And I've been using Mac for a while. I think I should probably buy a Chromebook to have just to kick around, like travel with and everything, because I don't want to travel with my heavy laptop. Yeah. Yeah. Well, it's been fun. I told you we were gonna stick to 30 minutes, so let's wrap up. Is there anything that you wanted to talk about, or what keeps you up at night about what's going on right now in the industry that you wanna just close with?

The market is really hot right now. Yeah. You have to look at what your competitive differentiators are as a company. You might be a company that can pay a fortune, so compensation is your competitive advantage. Good luck to you, well done. For the rest of us, you probably want to look at, well, what can we do that makes our candidates go, wow, and want to join your company? So experience, speed, whatever it happens to be, that's the thing that I'm constantly trying to help improve. Because that engagement is going to be critical for the next year, 2 years, however long this hot market runs. You've got to figure out your competitive differentiation. And the other piece is all of the people that interact with the candidates have to be on the top of their performance, from the interviewers to the profiles of your executives on LinkedIn. I was— no names— I was approached by a company last week for a job, you know, the jobs drop into your inbox. The first thing I did was look at the CEO's profile, and they'd lied about where they'd worked, and it was pretty obvious on their profile. And that integrity, I was like, nope, that's not a company I would ever want to join. I wasn't considering it anyway, but that's one of the examples. Those things can damage your company.

Oh yeah, big time.

You as the recruiter, you're the steward of the candidate's process and their experience. So go look at your executive's profiles, go, go check them out, see how you can improve that piece, the experience, the time through hiring and so on.

Because you're a candidate and like do your due diligence. How will someone view your company? That's good advice. I was actually going to ask you, for a lot of people, as you said, the market's really hot. We're hiring recruiters left and right, but there's also a lot of people entering the market that are pretty junior that don't know how to do this. So for a junior recruiter that's only got, you know, a few months of experience, How would you advise them to find the competitive differentiator? You just advise them how to find the, you know, the issues. But what about how do I identify my company's competitive differentiator?

Think about the values of the company. Go dive into that and research that. Think about what you are proud of for working at that company. Think about how you tell your parents, your grandparents about what your job is and who the company you work for and who they are. And distill that down into something that is meaningful that you can use in your first pitch, your email that you're sending out, but also turn it into something that can be a standard throughout the recruitment process that you can influence others to take on board and create something that will stand the test of time.

Perfect. That was a great closing statement. Great. I'm gonna turn this into a little bit of a post and follow up and tag you if you don't mind.

No problem at all. It's been a real pleasure. Thank you.

Thank you so much for being my first guinea pig, my test bunny.

I'll be in the audience for the next one. Appreciate it.

All right, thanks, Tim. Take care. Have a good weekend. Bye, everybody. Bye.