AI doesn't need you full-time

Here's a thought that's been nagging at me.

You know the fractional CFO model — companies that can't justify (or don't need) a full-time finance chief hire one for 10–15 hours a week. It's been a niche arrangement for years, mostly limited to the C-suite and mostly associated with startups that haven't scaled enough for full-time leadership.

But I think that model is about to eat everything. And AI is the reason.

Part one: the case

The productivity math has changed

The Harvard Business School study with 758 BCG consultants found that AI-augmented workers were 25% faster and produced 40% higher quality work on tasks within AI's capability range. Below-average performers improved by 43%.

Read that again. If AI makes a knowledge worker 25–40% more productive on the right tasks, a single experienced professional can serve multiple clients at quality levels that previously required full-time engagement. The economic argument for hiring one person, five days a week, to do a job that now takes two days — that argument just collapsed.

This isn't theoretical. The fractional executive market doubled from 60,000 to 120,000 professionals between 2022 and 2024. It hit $9.4 billion in 2025. LinkedIn profiles mentioning "fractional" went from 2,000 to 110,000 in two years. And it's no longer just CFOs and CMOs. It's data scientists, product managers, HR leaders, and — yes — designers.

AI replaces the junior, complements the senior

The Dallas Federal Reserve published research in February that sharpens this picture. They found AI substitutes for entry-level workers handling codified knowledge but complements experienced workers with tacit knowledge. Their observation is worth sitting with: the traditional model of hiring juniors to do codifiable tasks while they slowly absorb institutional wisdom is becoming cost-ineffective. AI does the codifiable part faster and cheaper.

So what happens? Companies skip the junior hire and bring in the senior professional — but they don't need that senior person full-time, because AI is handling the execution volume. They need them fractionally. For the judgment. The strategy. The decisions AI can't make.

This is the shift. AI doesn't replace the experienced human. It makes them so productive that no single organisation needs them for 40 hours a week.

The design function is already there

Fractional UX is no longer speculative. Koi Studios, Founders Who UX, Common Ideas — there are now agencies built entirely around the fractional design model. Indeed lists over 270 fractional UX jobs. A.Team raised $60 million to build fractional product teams combining designers, engineers, and product managers in a single engagement.

And the AI tooling makes it viable. Figma Make turns text prompts into editable wireframes in minutes. Google Stitch generates high-fidelity UI from descriptions. Synthetic Users simulates research interviews at a few dollars per participant. Maze and Dovetail handle AI-powered analysis and transcription.

A senior UX designer working 10–20 hours a week with these tools can now produce what used to require a full-time seat. Not because the work got simpler, but because the execution overhead got compressed. The strategic thinking, the research synthesis, the judgment calls about what to build and why — that still takes an experienced human. It just doesn't take one for five days a week.

If you've been following this newsletter, you'll recognise the pattern. AI is a velocity tool, not a navigation tool. It accelerates execution but doesn't replace the person deciding where to go. Fractional models are the organisational structure that follows naturally from that reality.

Part two: the cost

I want to be honest about what doesn't work, because the fractional discourse has a habit of skipping over the human parts.

Authority without power

The biggest risk is what practitioners call the authority-accountability gap. You get hired to lead design strategy, but you don't have the decision rights to actually implement it. Without explicit authority transfer, fractional engagement becomes advisory theatre — you're giving recommendations that no one is structurally obligated to follow.

Shallow roots

Fractional professionals develop breadth across clients but sacrifice the deep context that comes from daily immersion. Culture, politics, the unwritten rules of how decisions actually get made — you don't absorb those in two days a week.

The flexibility myth

Fractional professionals lose health insurance, retirement contributions, unemployment protections, and the stability of predictable income. The flexibility narrative sounds great until your largest client churns and you're scrambling. One widely cited case describes a fractional CFO working 80-hour weeks across too many clients — the exact opposite of the lifestyle promise.

These aren't edge cases. They're structural features of the model that don't have systemic solutions yet.

But there's a layer beneath those structural critiques that's even more practical, and even less discussed.

Your brain doesn't load context at runtime

There's an irony here: AI loads context at runtime. You open a project, feed the model your files and constraints, and it's immediately up to speed. Humans don't work like that. When you sit down on Tuesday to pick up a project you last touched on Friday for a different company, you spend the first hour clearing brain fog. Remembering where you left off. Re-reading your own notes. Getting back into the headspace of that team's problems, their politics, their language.

Multiply that across four or five fractional clients and you're burning a meaningful chunk of your "productive" hours just re-establishing context. AI doesn't have that cost. You do.

Consulting rates won't survive the expansion

Right now, fractional work commands consulting-level rates — fractional CFOs charge $175–$450 an hour, fractional design directors bill $8,000+ a month. That works when fractional is a premium offering for senior professionals. But if the model expands to most roles, as I think it will, businesses aren't going to pay consulting rates for a fractional mid-level designer or researcher. The whole point, from the employer's side, is cost reduction.

So what does a one-day-per-week job actually pay for the average person? And is that enough to live on when you need four or five of those jobs to make up a salary? The maths gets shaky fast.

Five Christmas parties, one person

Every company you work with fractionally has a Christmas party. Team offsites. All-hands meetings. The casual hallway conversations where real decisions get made. If you're fractionally employed by five companies, you're invited to five Christmas parties, five planning days, five sets of social obligations — all drawing from the same finite pool of your time and energy.

In practice, you'll skip most of them. And in skipping them, you miss the relationship-building that makes you effective in the first place. The fractional model assumes work is purely transactional. Work has never been purely transactional.

The holiday problem

If you want two weeks off, do you need leave approval from five separate companies? Do you coordinate schedules across five different teams who all have their own crunch periods? What happens when Company A's product launch falls in the same fortnight you promised Company B you'd be available for their usability testing?

Full-time employment has a single, clear answer to "can I take a holiday." Fractional employment has five competing answers, none of them simple.

Liberation or precarity?

And then the question I keep coming back to: does this future point towards under-employment or over-employment?

The optimistic version says fractional gives people portfolio careers — varied, autonomous, well-compensated. The pessimistic version says it gives people five part-time jobs, none of which provide benefits or stability, all of which demand more energy than they pay for.

Which one materialises probably depends on how senior you are, how much leverage you have, and whether labour protections evolve to match. For the experienced design director, fractional might be liberating. For the mid-career designer trying to make rent, it might just be precarity with better branding.

Part three: what comes next

Someone will try to solve the coordination problem

All those human coordination problems — the context-switching, the holiday logistics, the five Christmas parties — point towards an obvious market opportunity: unified employment agencies that simplify a person's engagement across multiple fractional clients. One entity handles your contracts, invoicing, scheduling, and benefits. You just do the work.

This isn't a new idea. Consulting firms like Accenture already do exactly this. You're technically employed by Accenture, but you work on client projects. The firm handles the administrative overhead, provides benefits, manages client relationships, and smooths out the gaps between engagements. It works. But consulting firms take a significant margin for that service — often 2–3x what the individual earns — because they're selling expertise at premium rates.

But the margin model won't hold

If fractional employment expands beyond the C-suite and consulting-rate roles, the margin that funds that model evaporates. Businesses won't pay Accenture-level markups for a fractional mid-level designer. So the intermediary has to get cheaper. Maybe it looks more like a recruitment firm that takes a smaller cut for ongoing placement and admin. But if salaries are already under pressure from the fractional model itself, why would workers pay even a modest fee just for convenience?

Every person becomes their own agency

This is where it gets interesting. If the economics don't support a human intermediary, the intermediary becomes AI. Every person becomes their own employment agency — a personal AI agent that manages your client relationships, handles scheduling conflicts, negotiates rates, tracks contracts, and coordinates your availability across engagements.

I wrote about something adjacent to this in Open hiring harness: the idea that AI could sit between individuals and the market, managing the complexity that currently requires either a full-time employer or an expensive agency.

The progression would be: consulting firms (high margin, high service) → fractional platforms like A.Team and Toptal (lower margin, marketplace model) → personal AI agents (near-zero margin, self-managed). Each step removes cost from the intermediary layer and pushes more coordination capability to the individual. Whether that's empowering or exhausting probably depends on how good the AI agents actually get at handling the messy, human parts of work relationships — the parts that aren't just scheduling.

So what does all of this mean if you're actually in the design space?

The early-adopter window is open right now

Here's the thing about adoption curves: the people who move first get paid disproportionately well.

Right now, fractional is still unfamiliar enough that companies hiring fractionally are competing for a small pool of people who've figured out the model. Supply is limited. Demand is growing. That imbalance creates a premium. A senior designer who positions themselves for fractional work today could realistically land two part-time engagements, each paying 70–75% of a full-time equivalent salary — totalling 140–150% of what they'd earn in a single full-time role. For less total hours. That's not a fantasy number. That's the maths of a supply-constrained market.

This won't last. As the model normalises, as more professionals adopt it, as AI agents start brokering fractional engagements at scale, the supply side catches up. Rates compress. The premium erodes. The structural problems I outlined in part two — the context-switching tax, the compensation pressure, the coordination overhead — those problems surface more acutely when the market is crowded and the leverage shifts back to employers.

So there's a window. Probably two to three years, maybe a bit longer in the design space where adoption is still early. If you're a senior designer or design leader thinking about this model, the time to experiment is now — while the economics favour the practitioner, not later when they favour the platform.

But go in with your eyes open about the context-switching tax, the social costs, and the compensation trajectory. The early-adopter premium buys you time to figure out the operational reality before everyone else arrives and the margins tighten.

The junior path is narrowing

If you're earlier in your career, the picture is more complicated. The Dallas Fed research suggests the entry-level path into design is narrowing as AI absorbs the codifiable tasks that juniors traditionally cut their teeth on. This doesn't mean junior roles disappear, but it does mean the path to seniority might get steeper and less linear. And it's the seniors — the people with the tacit knowledge and judgment that AI can't replicate — who are best positioned to capture the fractional premium.

The governance question nobody's answering

If you're building a team, the question is no longer "should we hire a full-time designer or a contractor?" It's becoming "what's the right blend of full-time core team and fractional specialists, and how do we give fractional people enough authority and context to actually deliver?"

That's where most organisations will stumble. Getting the talent is the easy part. Structuring the authority, the knowledge transfer, the holidays, the social integration, and the accountability? That's the design problem nobody's solved yet.

The fractional future is coming. I just think we should be more honest about what it actually costs the humans inside it — and more strategic about when to jump in.

AK