How do we fix the capacity gap?
We're going to need a bigger bucket.
Ever feel like there just aren’t enough hours in the day? That everything you need to do simply cannot be done by one person? That it’s literally impossible to ever ‘complete’ your work?
If you are employed, I highly suspect your answer to these questions is ‘yes’. I’m fairly confident you’ve also thought or said some variation of the phrase ‘I’m at capacity’. Perhaps the more emotive twist: ‘I’m drowning here’. Or you’ve kept it simple: ‘I can’t do this’.
I’m assured in this presumption because this is the nature of most traditional full-time jobs in 2025. There’s too much work to be done; a neverending onslaught of KPIs and OKRs and targets that reset before you’ve had the chance to celebrate nearing one. Teams have shrunk and been stretched paper thin. The rising cost of everything has meant bosses are trying to get more done with less money (and thus fewer people). Whatever industry you’re in, you’ve likely come across a problem, immediately identified the solution as more money or more people, and then had to sit with the knowledge that neither can be offered up. So… what then?
There’s a term for all this: the capacity gap. That’s what it’s called when an organisation doesn’t have the resources or capabilities to fulfil the goals it needs to achieve. In Microsoft’s Work Trend Index for 2025, the company declares this as a need-to-know term for “a new world of work”, defining it as: “The deficit between business demands and the maximum capacity of humans alone to meet them”.
When a company has a capacity gap, they have a few different choices for how to fill it. The first option is to simply force the capacity up. In this model, you expect staff to go far above and beyond, to work long hours, to just improve their performance and be more productive (easy!). The end result of this is burnout, unhappy staff, and difficulty keeping anyone around. The second option is to pare back the goals to match current staff’s capacity. This is rarer. The end result of this is a dip in profits, slower growth, that kind of thing. So you can see why few organisations go for it. The third option is the one Microsoft’s definition of the capacity gap is subtly insinuating: to recognise that humans alone can’t fulfill business demands, and to turn towards AI instead.
In recent years, some organisations have gone for option three. But within option three are more options, ranging from good to bad. Guess which end of the scale the majority of businesses who’ve opted for option three have landed?
The bad side of option three is, of course, outsourcing almost everything to artificial intelligence, dramatically chopping down the workforce or reducing their hours because running computer programmes is much cheaper than employing human people. We all know this is bad, for many reasons! For one, it’s morally shitty to take away a load of people’s incomes. For two, it turns out that a lot of stuff isn’t best done by AI (and definitely not by poorly designed AI), but a real, creative, emotion-feeling person. For three, a lot of AI right now is not up to the job.
We all know this, and yet… we can also all see why businesses optimistically go down this route. Something has to be done when there’s a mismatch between capability and needs, and options one and two suck. If option three sucks, too… then what? Is the ubiquitous capacity gap just a problem that’s impossible to solve?
The good news is this: no! Remember the scale of goodness I mentioned about option three? That means there is a good version of looking towards AI; one that doesn’t produce the burnout of option one or the business failure of option two or the moral failure of option three.
What exactly would this good version of using AI look like? Let’s break it down.
A moral duty to give people meaningful work
The future of work requires those in power to have a big, grand realisation: that meaning matters and we, as a society, have a responsibility to enable all people to pursue it. And then another realisation: for a lot of people, that meaning will come through the form of work. There’s a whole load of scientific research on the importance of meaning to people’s wellbeing. In layman’s terms, the general vibe of humanity is that a meaningful life, a life with purpose, a life where you’re putting in some form of effort, is better than a life that’s meaningless idling.
The future of work requires those in power to realise that they can’t only pursue personal profit, but that they need to also create some net good in the world, and part of that can be done by the provision of meaningful work.
Outsourcing meaningless work to AI
With the above point in mind, the way business leaders approach AI would be radically different. The decision of when and where to use AI wouldn’t be down to productivity, but the provision of meaning. We’d scoop up all the pointless, annoying, unenjoyable bits of a person’s job and outsource them to AI, leaving the meaty, interesting stuff to a human. Not because the interesting stuff will be done better by a human (although it probably will), but because we have a moral responsibility to use work as a way to provide people with meaning.
No reduction in staff
Hey, bosses: did you know that AI can be additive? It doesn’t have to replace people! The future of an organisation with a capacity gap could look like this: they realise there’s more work than is humanly possible to accomplish, so they bring in AI tools to take on the overflow of work, and then – here’s the crazy bit – they don’t gut their workforce. That way, the capacity increases. All the work fits in the bucket. After a while, the capacity might even exceed the demands. Imagine if we responded to that by then increasing the demands without reducing or stretching staff? What if we upped capacity, which upped output, so then we could afford to up capacity even more, and so we did, and then output increased even more, and so on and so on?
Recognition of what’s better done by humans
AI is good at lots of things. It is shit at lots of other things. We all need to get to grips with this and stop pretending that AI art is good or that a decision is better made without emotion involved. Let’s appreciate our human skills please.
AI agents and agency
You might have read the above about meaningful work and thought something like: ‘but who decides what work is meaningful? What if I give a bunch of work to AI and then actually that’s the work that a human would like?’. If you did think this, bonus points for you. This is a good thought, because what’s meaningful and interesting to me might be radically different to what’s meaningful and interesting to you. So how do bosses decide? The answer is: they don’t! Instead, the ideal future of work would lie in providing human staff with their own AI agents, and then leaving the decision of how and when to use those agents up to the human staff.
Shifts in traditional working patterns
If we start using AI in this way, we’ll need to shift the way we define ‘good work’. Presenteeism needs to go away forever. Same goes for the idea that you can measure someone’s worth in only the hours they spend grafting. AI will speed everything up, and the temptation will be to fill the saved time with more work. No! Don’t! If the work will suddenly take three days instead of five, we shouldn’t just add on more busywork to keep everyone from twiddling their thumbs. Instead we should acknowledge that we’re all likely working too much rather than not enough, and give the time back to workers to use as we wish… without cutting pay, because the work we produce will only increase in volume and quality.
I know this vision of the future of work is optimistic. It asks for a lot from bosses; prioritising people over profit, radically overhauling processes, and – seemingly one of the most challenging asks – acknowledging that there’s a problem. In order to solve the capacity gap, leaders will need to accept that it exists, and that the solution doesn’t lie in ‘just working harder’ or ‘optimisation’ or relying on high performers pushing themselves to exhaustion. They may need to hit pause and spend time and money and effort on this transformation. And in the capacity vs goals equation, upping capacity may lead to a reduction in goals delivered at first. But it’s the sustainable way forward. The prevalence of the capacity gap means we’re going to need a much bigger bucket.
A reminder: Working On Purpose the book
I wrote about the future of work (and the importance of purpose and meaning) quite a bit in my upcoming book, Working On Purpose. But if that’s not your jam, you’re still in luck! The book also covers a lot of present-day important stuff about work, such as how to manage the different generations, exercises to figure out what work means to you, easy hacks to make your job better, and a no-bullshit guide to all the work trends flying about.
It’s out in August and I would greatly appreciate it if you would give the book a pre-order. If that’s out of budget at the moment, no worries! You can help me out instead by telling people about the book. Recommend it to your friends, family, bosses, colleagues, exes, whoever you like.
Work-related reading recs:
I was honoured to be invited on a panel about a new report from King’s, which looks at the impact of return-to-office orders. You’ll soon be able watch back the webinar, which features me sounding very sickly (I probably shouldn’t have worked this day, tbh. I was so tired I was slightly delirious. I have very little memory of what I said. Hopefully nothing bad), or you can read the full report now here.
Recommending not just because I love the picture (which represents *exactly* how I feel when challenged in a meeting), but because it’s a very handy guide to keeping your cool in a tricky situation at work
Interesting piece on the increasingly negative way bosses are talking about workers
An example of a bad use of AI to increase capacity. The point about having ‘less time to think’ is worth noting. How much time do you have in your job to actually think? I imagine not as much as you’d benefit from. More thinking time, please.
No news on impact on cats, but here’s a new study that says our work stress could be harming our dogs.
Bonus: here is a work-related headline. I won’t share my initial reaction to it.
Really interesting read Ellen thanks. One of the things I see a lot in my work is the capacity gap can sometimes be interpreted wrongly by senior execs and bosses as a "capability" gap, which then leads my clients to start questioning and working on the wrong things. The reframe you suggest in option 3 is really worth us exploring.