This Week, In Recruiting - Issue 171

Sponsored by our friends Odro

The recruiting community suffered a heavy blow with the loss of one of our most inspirational members last month. I want to use this space to pay special tribute to Ryan McCabe, a man who was much loved and respected throughout the industry. It was my honour and privilege to know you mate 👊

If you are looking for a video interview solution for your staffing agency business, check out Ryan's market leading platform, Odro

Open Kitchen: State of Tech Hiring Today, Tomorrow

The collapse of the software development job market from the post lockdown peak of 2021 was vividly captured in this Federal Reserve Economic Data chart of Indeed US job postings from 2020 to 2024.

Many commentators talked about it, trying to rationalise the phenomenon in order to better understand what might happen to tech hiring going forward. Whilst nobody thinks we'll be returning to the hyper-scaling era when the right size for an engineering team was 'as many as you can hire', there were many who thought that we are seeing the return to normal from the 'over hiring' peak of post-lockdown. Employers had charged forward with enthusiasm fuelled by two years plus of dry powder and the prospect of new economy of remote first / human first capitalism to build for and sell to, and now had to course correct back to a more business as usual rate of hiring. It only feels catastrophic because of the dizzying heights reached in the mania of 2021/2022.

The over hiring thesis isn't invalid but it discounts the very real factors which derailed the build-back-better moment in the first place. These have set us off onto an entirely different timeline - one that features slower growth, higher cost of doing business, leaner, globally dispersed organisations and - I think - progressively smaller software engineering teams.

There are 6 factors which persuade me of this case.

1. Collapse of VC Investment into B2C SaaS

All of the hyper scalers of the 2010-2019 period were generally B2C SaaS companies. These were the companies where the strategy was to scale as fast as possible to secure the dominant position in the category, locking out competitors by the simple fact of becoming the de facto market itself. Think about Uber in it's heyday, or WeWork in theirs. Users didn't search for alternatives because there was no point - what you wanted or needed was going to there in the single platform. These companies - and the hundreds which wanted to be like them - were the ones which secured most of the funding in the ZIRP era, mostly converting most of that investment into payroll for software engineers. A fair number of us worked as recruiters for these companies - and had a great time doing so!

The collapse of VC investment correlates with the collapse of this tech job market. In 2022, European VC funds raised a record $22 Billion, in 2023, they raised $10 Billion. That's $10-$12 Billion worth of tech salaries / recruiting fees / employer branding budgets etc less than the year before. In this context, the third order effect - the collapse of our tech recruiter job market - shouldn't be a surprise.

Atomico / Slush publish the State of European Tech in Q4 every year so watch out for it when its released. However, we know from our own interactions with the market that we're not going to see much more VC investment in 2024 than we saw in 2023. What's more, the investment that we are seeing is going not to headcount heavy SaaS but headcount light AI startups. Today's investors are buying chips, not engineers.

2. Global Dispersal of Tech Jobs

Some early Cassandra's predicted that remote working was going to be a short term pitstop to off shoring. Sean Blanda was one of those, penning the prescient 'Our Remote Work Future is Going to Suck' in 2020 at the height of remote. He struck a chord because he was obviously right at the time, but we were too giddy in the moment (we were building back better, remember?) to want to recognise the directional truth of remote work - if your boss doesn't need you in the office, maybe s/he doesn't need you in the country at all. Off shoring has been around since the early phase of globalisation so is nothing new to us but we were suspicious of it until the shift to remote taught trained everyone there was nothing much to worry about.

Regardless of department, I were a managing a P&L today I would be actively looking to offshore as many roles in my team as I could. Remember that we are in a high cost of business era, where every tier of the business has to reduce cost, increase efficiency, maintain or boost productivity. Getting resources at 1/10 to 1/4 of the cost is going to be a no brainer. The shift of tech jobs from Global North to Global South doesn't mean any less recruitment; it actually might mean more, given that you could hire more people for the money you spend. However it does mean that recruitment will change, and Global North tech recruiters will have become Global Global tech recruiters with fluency in talent intelligence, local jurisdictions, supplier / partner management and so on.

3. Peak FAANG

Facebook, Google (Alphabet), Amazon, Microsoft - these were biggest and most attractive of employers of software engineers over the past 20 years. They were also the most well compensated - if there was ever place where the War for Talent was real, it was in FAANG for tech talent during the 2000-2020 era. Whilst these companies still retain huge engineering teams (10,000 - 50,000?), there is a good argument to make that we may have already reached 'Peak FAANG'. Both Meta and Alphabet are fundamentally ad revenue businesses. They are dependent on consumers surfing the internet, in the same way retailers were dependent on citizens walking down the high street. It doesn't take a genius to notice that internet user behaviour has changed - all you need do is speak to anyone younger than 25. Facebook - Meta's flagship revenue maker - has long had the feel of abandonware, and has logged precipitous decline in usage amongst Gen Z and Millenials. As the internet population ages out, so the traffic will recede and so too with the ROI on adverting budget. Google faces a similar threat - search has been plagued with advertisement and SEO'ed articles, such that no one really believes that Google believes in the their maxim of 'organising the world's data'. Users are abandoning Google for small community intelligence, or short form video like TikTok or even LinkedIn. The fact that both of these companies remain massively profitable disguises the trend that people are using their services less and less every year.

The extension of Mark Zuckerberg's 'Year of Efficiency' into a permanent setting for the Meta is an expected outcome for a company which is looking for its next business. There will still be tons of hiring of course, but at slower rate than we've seen before and likely never again at the levels of 2021. That moment was peak remote, peak recruiter and peak FAANG.

4. No One Knows How To Right Size Software Engineering Teams

It's probably true to say that it's still too early to know how AI will disrupt software engineering. By now, most software engineers will have used AI in some way (GitHub released Copilot one year before OpenAI launched ChatGPT) and most seem confident that it is a useful productivity / assurance tool but not a replacement for an experienced software engineer.

There are three things we know about this: Firstly, productivity uplift is irrefutable - the bottom line is, code can be generated faster than it can be typed. The combination of machine generation + human oversight is faster and better than human only coding. Secondly, there is increasing usage of GenAI, with code submissions on GitHub becoming increasingly dominated by generated versus typed code. Already the code on GitHub is majority generated (over 50%); the CEO of Github estimates that 80% generated vs human is only months away. Thirdly, the innovation rate of AI disconcertingly fast, with step change upgrades happening daily, weekly, monthly cadence.

In these circumstances, it is becoming impossible for employers to forecast how many engineers they may need to perform the work. How do you know how many engineers you need to perform task X or achieve objective Y? It was easier to figure out when the capacity of the individual engineer was stable, but in the era of AI-assisted software engineering, workforce planning is now like throwing darts at rotating board; we no longer know how to right size engineering teams because the circumstances are changing too fast. This doubt, of course, creates hesitation when it comes to hiring - reducing the number of released job reqs and lengthening time to hire as hiring managers prevaricate.

5. Democratisation of Software Production

Whilst GenAI has enabled experienced developers to increase their productivity, it has also opened the door for non-specialists to enter into the activity of software production. I call it 'activity' rather than 'work' because it is unclear to what degree these new entrants have been able to convert their effort into compensation but if they haven't already I suspect it will only be a matter of time before they do. GenAI is turning out to be a 'Gutenberg moment' where technology innovation (printing press) disseminated a previously tightly locked and controlled capability (literacy) giving everyone with access the potential to become productive in this new capability.

Some of the best examples actually come from within our own industry, with friends like Paul Karrmann or Johannes Sundlo producing functional solutions using GenAI to output code, configuration and deployment instructions. Now this activity does not directly replace any individual software developer but it does reduce the necessity of having a software developer around to transmute the business idea into technical reality. The work that Paul or Johannes is producing today would have required a software developer a year ago. So perhaps the demand for software developers on aggregate will also goes down as AI-literacy spreads, and DIY solutions become 'good enough' for most of us.

6. True Containerisation of Code

One of the unintended consequences of Generative AI is standardisation of output.

AI is trained on what has gone before, and its recommendations are most likely to be the most common previous responses to similar enquiries. We all understand that GenAI cannot really reason, but comes up with the 'right' answer based on 'frequency of previous acceptance'. This means that code will become increasingly standardised as known solutions become more consistently recommended. Over time, this means that the idiosyncracies of human written code will be stripped out as every refactor of the codebase reduce individuated solutions in favour of standardised, machine recommended ones.

Code quality will 'get better' - meaning that it will become more efficient, more understandable, more logical, more interoperable. This also means less maintenance, less troubleshooting, and less requirement for software developers to conduct this type of work. There is a precise parallel with the admin work that recruiters might get rid of with the advent of AI - potentially this is freeing for us because we can then focus on the higher value work. However, it might also mean that employers see this as a reduction of the overall work requirement and adjust (downward) the resourcing estimates.

I'm reminded of the containerisation revolution that transformed international trade. Prior to the simple invention of the standardised shipping container, goods were manually handled as cargo by legions of dockworkers and warehousemen worldwide.

Nobody does this work anymore, because containers can be transported by automated cranes, before being loaded onto driverless lorries, before being dispatched to automated warehouses. This would not have happened if the goods were not prepackaged into stackable standardised containers. Code is kind of going to go the same way.

What Does This Mean for Tech Recruiters?

There is still a huge demand for software engineers, so it is premature to think about jumping into veterinarian recruitment, though that too is super in-demand, especially in post Brexit UK. However, the nature of the market has transformed and so we need to transform with it. In summary:

1. Hyper scaling era is over, reset expectations accordingly
2. 2000 era digital transformation still not complete (!), so plenty of tech hiring still to be done in mature organisations
3. Off shoring tech will become a corporate necessity, capability of delivering to this will become a premium recruitment asset
4. Become AI-enabled recruiter, hiring AI-enabled software engineers - new relationships will form based on new two sets of jobs undergoing transformation
5. And evergreen stuff - lean into community & build personal profile - both will be key in accessing collective intelligence, the only antidote we humans have to the artificial kind of intelligence.

What do you think? I am out of my mind? Give me a heckle in comments, love to know your thoughts, especially tech recruiters and recruiters in tech!

Now out of the kitchen, onto the lounge 👇

What's in the News?

KarmaCheck raise $45 Billion Series B Round

Nice to see some fundraising go back into the HR tech - a sign that the hiring demand may have some recovery, especially in the US. It's also the case that AI-enablement is going to increase to fraud risk for employers everywhere, so background checking will be one of those phases in the hiring process which is guaranteed to require more investment. Blog post PR here

If you have any news which you think the wider market needs to know about, put into comments below on this thread.

What's Going On?

Big List of Recruiter Events to Attend in 2024

Big List of Recruiting and HR Events to Attend in 2024 - we're mid year and there seems to be no end of the events coming up! It's great that we can move forward into the summer with a sense of celebration and community. Bookmark this spreadsheet, add to it, share it in your community and make sure you attend some of these events!

How to Build An Effective Employeer Referral Programme, 2nd July 1pm BST

Excited to be partnering with our friends Real Links on this webinar series. Did you know that referrals remain the No1 channel for retention? That means that referred candidates are more likely to meet the Quality of Hire bar and be the sort of fit to company which makes them stay longer. We're going to hammer out what makes an effective ERP in this webinar - consider it 1 hour of free consultancy from the likes of Bill Boorman, Carolina Guillen, Michael Boud and Sarah Meijer, co-hosted by me and Sam Davies. Make sure you register here.

Founders Focus - Ep47 - Up close and personal with Dina Bay, CEO of PitchMe, Weds 3rd July, 12pm BST

Founders Focus on a weekly rhythm folks - next up is Dina Bay, CEO of PitchMe. I've known Dina for a few years and her business is one of the great supporters of Recruiting Brainfood, but we have not had the chance to take a deep dive into Dina's founder's journey - until now. What leadership lessons has Dina picked up from her path as founder and CEO of PitchMe? We're on Weds, 3rd July, 12pm - register here

Summer TA Leadership Event, WeAreKeen, Thurs 4th July

Impromptu return visit to Amsterdam for me this week as I go and visit our good friends WeAreKeen in order to take part in their TA Exec events series. I'm going to be there, alongside Rina Joosten-Rabou, CEO of Pera. I think you got to be a TA leader in Amstedam, but if that's you, register here.

Brainfood Live On Air - Ep263 - Public Speaking Tips for Newbies, Fri 5th July, 2pm BST

One week to go before we hit Recfest and it's fitting that we're going to do a Brainfood Live on how to give a great public speech. It's a nerve wracking time for even experienced speakers, but this there will be dozens of new voices on stage for the first time. How to give a great talk? We're going to be bring in trained speakers, experienced stage hosts and conference organisers to give their best tips. We're with Akbar Karenga, Founder (Maarusi), Sam Berthoud, VP of Talent Solutions (hackajob), Angela Cripps, MD (Recruitment Gym), Christina Robinson, Founder (Green Umbrella), Bill Boorman, Founder (TRU) & friends. Register here

Understand Quality of Hire with a Person’s Web Presence, Tuesday 18th July, 6pm BST

So this one is going to be cool - can you assess QoH based on person's digital footprint? I'm in conversation with a man who thinks you can. The Ben Mones story is fascinating example of how a challenging incident sparked an idea, which has subsequently gone on to become a unique assessment solution in our space. Huge amounts to talk about here - we're on Tuesday 18th July, 1pm EDT / 6pm BST. Register here

If you have an event, webinar or podcast going on next week and want it featured on next week's newsletter, comment below with the link and event details. Don't forget to at mention me so that I see it

End Notes

Massive week for a lot of people this week - national elections round 1 complete in France, General Election in UK, US celebrate Independence Day on the same day, the start of Wimbledon, continuation of Glastonbury, and my arrival to Amsterdam for the end of the week celebrations. What can I say but wish everyone best of luck!

Cheers

Hung

Hung Lee is the curator of Recruiting Brainfood, and now This Week In Recruiting. Subscribe to both if you are into recruiting or HR or just interested in world of work.