Sixty-three per cent. That's the proportion of mid-sized B2B service firms we've spoken to in the last two years that described their digital resourcing model as "not quite right." Not broken, exactly. Just... not quite right. Too expensive for what they're getting, or too thin for what they need, or structured in a way that made sense three years ago but doesn't anymore.
Almost all of them framed the problem the same way: Should we outsource this or bring it in-house? As if those are the only two doors.
I want to make the case for a third option - but I want to earn it, not just assert it. Because I'm well aware that someone who runs a consultancy telling you to use a consultancy isn't exactly a plot twist. So let's go through the honest version of both sides first.
The argument for building internally is genuinely compelling. I hear it most often from managing partners and CTOs who've been burned by external partners, and it goes something like: We should bring everything in-house. It's cheaper in the long run and we keep the knowledge.
The logic holds, up to a point. When you hire well, an in-house team gives you things that are hard to replicate from outside. Someone who's been with your firm for two years understands your clients, your internal politics, the weird reason the CRM does that thing with duplicate contacts, and the unspoken hierarchy of who actually makes decisions versus who thinks they do. That context is worth a fortune. It takes an external team weeks to piece together, and even then they might not get the subtleties right.
There's also the simple fact that an in-house person wakes up thinking about your business. They're in the Monday morning meeting. They hear the throwaway comment from a partner about a client complaint and think, "Actually, that's related to the thing we're building." External teams can be brilliant, but they're splitting their attention across multiple clients. Your priority is their Tuesday afternoon.
And yes, long-term cost can genuinely be lower. If you have enough work to keep a senior developer or UX designer busy forty hours a week, fifty weeks a year, the maths favours employment over day rates. A good mid-level developer in-house might cost you £65-80k fully loaded. That same person from a consultancy, at £600-900 a day, would cost you £150k+ for the equivalent time.
So why doesn't everyone just hire?
Recruitment is the first wall you'll hit. And it's a proper wall, not a speed bump. The market for good digital talent - developers, UX designers, product managers, data engineers - is brutal. You're competing with tech companies, agencies, startups, and every other firm that's had the same "let's bring it in-house" epiphany. A mid-market law firm or consultancy isn't most people's dream employer if they want to do interesting digital work. I don't say that to be unkind. I say it because I've watched firms take six to nine months to fill a single senior digital role, and by the time they've hired, the brief has changed.
Then there's the capability gap. You might need a brilliant UX researcher for six weeks, a front-end developer for four months, and a data migration specialist for three weeks. Hiring all three full-time is absurd. Hiring one generalist who can do all three is a fantasy. So you end up with someone whose skills cover 60-70% of what you need, and you either stretch them into areas they're not strong in, or you outsource the gaps anyway.
The other headache is scaling. Digital work isn't linear. You'll have months where you need eight people sprinting on a platform rebuild, and months where you need two people doing maintenance and optimisation. In-house teams don't flex like that. I was talking to a COO at a 200-person consulting firm last year who'd built an internal digital team of five. He'd done it properly - good hires, decent salaries, proper onboarding. "Three of them are brilliant," he told me, "but for about four months of the year, I genuinely can't justify what two of them are doing." He wasn't complaining about the people. He was describing a structural mismatch between how digital work arrives and how employment contracts work. There's no clean fix for that.
And there's a subtler risk that nobody talks about until it happens. In-house doesn't mean the knowledge stays. It means the knowledge stays as long as the person stays. And people leave.
Right. So if in-house is harder than it looks, surely the answer is to outsource everything?
The advantages are real. A good external partner can mobilise a team in days, not months. No recruitment, no onboarding, no three-month notice period from their previous employer. When you've got a platform migration that needs to happen before your CMS vendor ends support, that speed matters. And a consultancy will have specialists across strategy, design, development, data, content, and testing - you get the UX researcher for six weeks and the migration specialist for three, without employing either of them.
But fully outsourced has its own problems.
Cost, for one. Day rates for senior digital talent from a credible consultancy are not cheap. They shouldn't be - you're paying for depth of experience, the ability to scale, and the reduced risk of getting it wrong. But if you're commissioning every piece of work externally, the annual bill adds up fast. I've seen mid-market firms spending £400-600k a year across multiple external partners on work that a well-structured internal team could handle for half that.
Knowledge walking out the door is the mirror image of the in-house retention argument. When an engagement ends, the external team moves on. They'll leave documentation - the good ones will, anyway - but the deep, contextual understanding of your business and your clients goes with them. The next time you need something, you're either paying for that context to be rebuilt, or you're accepting a slightly less informed version of the work.
And then there's the management overhead that nobody budgets for. External partners need managing. They need briefs, feedback, decisions. Someone internal has to be the bridge between the business and the delivery team. If that person doesn't exist - or if it's the marketing director doing it as 15% of their role alongside everything else - the quality of the output suffers. Not because the external team is bad, but because they're flying partially blind.
I've watched this play out more times than I can count. A firm outsources everything, the work is good but expensive, someone in the leadership team says "we should bring this in-house," they hire one or two people, the external partner is scaled back, the internal team discovers they can't do everything, the external partner is brought back in - and now you're paying for both, with neither model working optimally.
So what, James - you're saying neither option works?
No. I'm saying neither option works as a complete answer.
Own the strategic capability. Partner for specialist execution.
That means building internal strength in the areas where deep business context is essential and the work is continuous - digital strategy, product ownership, content governance, analytics, vendor management. The people who decide what gets built, why it matters, and whether it's working.
And it means partnering externally for the specialist, intensive, or intermittent work - platform builds, UX research sprints, complex integrations, migration programmes, AI implementation. The people who know how to build it, have done it dozens of times before, and can mobilise at the scale you need when you need it.
This isn't a cop-out middle ground. The internal team holds the institutional knowledge, sets the direction, manages the measurement, and owns the relationship with the rest of the business. The external partner brings depth of execution capability that would be impossible to maintain in-house at a firm of your size.
Think of it like this. You wouldn't hire a full-time architect to live in your house on the off chance you want to extend the kitchen. But you also wouldn't let a builder redesign your house without any input from someone who lives there. The architect visits when you need them. You live there the rest of the time. Both are essential. Neither can do the other's job well.
I'm going to resist turning this into a two-by-two matrix because the world has enough of those. But there are a handful of questions that cut through most of the noise.
How often do you need this capability? If the answer is daily or weekly, it probably belongs in-house - content publishing, analytics review, day-to-day platform management. If the answer is quarterly, annually, or project-based, it's almost certainly more efficient externally - platform migrations, major redesigns, technology selection.
How much business context does it require? Work that depends heavily on understanding your clients, your internal processes, and your commercial model benefits from someone who's embedded. Work that depends more on technical or specialist expertise than business context can sit outside.
What's the cost of getting it wrong? Strategic decisions about what to build and why need to be owned internally. The cost of an external partner misunderstanding your market positioning is much higher than the cost of them writing slightly imperfect code. Execution errors are fixable. Strategic errors compound.
Can you actually recruit for it? Be honest. If you're a 150-person accountancy firm in Birmingham, you are not going to attract a senior product designer away from a London tech company. Build your internal team around the roles you can genuinely fill, and partner for the ones you can't.
Most firms I work with don't start from a blank sheet. They're already somewhere on the spectrum - usually either fully outsourced and frustrated by the cost, or partially in-house and frustrated by the gaps.
If you're starting from fully outsourced, the first hire should be someone who can own the relationship between the business and your external partners. Call it a digital lead, head of digital, whatever fits your structure. Their job isn't to replace the external team - it's to make the external team dramatically more effective by providing the context, decisions, and direction that only an insider can provide. That single hire can reduce external spend by 20-30% within a year, not by cutting scope, but by eliminating the waste that comes from poorly briefed or misdirected work.
If you're starting from an overstretched internal team, the move is usually about identifying the specialist work that's either not getting done or getting done badly because you're asking generalists to be specialists. That's where an external partner earns their fee - not by replacing your team, but by augmenting them at the moments when depth of expertise matters most.
If you're somewhere in the messy middle, the first step is an honest audit of who's doing what, how much it costs, and how well it's working. We did this for a wealth management firm a couple of years ago. They were spending roughly £600k annually across three external vendors and a four-person internal team. No clear delineation of responsibilities, no shared roadmap. When we mapped it out properly - which took about two weeks and a lot of uncomfortable conversations - the picture was pretty grim. The external partners were duplicating each other's work in places. The internal team was doing admin that could have been automated. Nobody had a clear view of the total spend or what it was producing.
The fix wasn't dramatic. Clarify who owns what. Consolidate external partners. Give the internal team the tools and authority to do their actual job. Within six months, they'd redirected £180k of annual spend to higher-impact work and decommissioned a legacy platform that had been costing them money and delivering nothing. The CMO told me afterwards that the most useful thing we'd done was just draw the picture. Nobody had ever drawn the picture before.
Look, I run a consultancy. I have an obvious interest in you using external partners for at least some of your digital work. I'd be lying if I pretended otherwise.
But I'd also be lying if I told you that outsourcing everything is the right answer, because it isn't. I've actively talked clients out of over-outsourcing when it was clear they'd be better served by building internal capability - including one firm last year where I told them they didn't need us anymore and should hire the person we'd been providing. They did. It was the right call. I don't regret it, even if my finance director gave me a look.
The firms that get this right tend to be clear about what they own and what they partner on. They invest in the internal capability to manage external relationships well - because a poorly managed external partner is worse than no external partner at all. And they revisit the model regularly, because the right balance shifts as the business grows and internal skills develop.
The firms that get it wrong almost always make the same mistake. They treat it as a binary choice, pick one side, and spend years compensating for the gaps.
We've been meaning to sort our digital resourcing out for a while, but it's not a priority right now.
I'd gently push back on that. Every month you spend with the wrong model - too much cost, too little control, or both - is a month where your digital investment is working harder than it needs to or delivering less than it should.
If you're not sure where your current model sits, that audit I mentioned is a good place to start. It doesn't have to be us - but it does have to be honest. And sometimes honest is easier when it comes from someone who doesn't report to the same board you do.