There is no universally right answer. An agency buys breadth and production capacity, a consultant buys senior attention without headcount, and an in-house hire buys dedication and compounding context. The right choice depends on your spend, your stage, and how much of your problem is judgment versus throughput. I am a consultant, so read this knowing that.
I will try to earn your trust the only way that works in a piece like this: by being genuinely fair to the two options I do not sell. I have worked inside and alongside all three models for over a decade, and each one is the correct answer for a real set of companies. The failure is not picking the "wrong" model. It is picking a model whose failure mode you were not prepared for.
What is the real difference between the three?
You are not choosing a vendor. You are choosing whose judgment your budget answers to.
Strip away the packaging and that is the decision: an agency gives you a team's process, a consultant gives you one senior operator's judgment, and an in-house hire gives you an employee whose entire context is your app.
| Agency | Consultant | In-house hire | |
|---|---|---|---|
| Who does the work | A team, usually mixed seniority | The senior person you spoke to | Your employee |
| Cost shape | Monthly retainer, sometimes tied to spend | Fee for a defined scope or ongoing fractional time | Salary plus benefits, tools and ramp time |
| Breadth | Widest: channels, design, tooling | Deep in their specialism, narrow outside it | As broad as the person you hired |
| Speed to start | Weeks: onboarding, kickoff, handovers | Days: one person, no internal handoffs | Months: hiring plus ramp |
| Context depth | Split across a client roster | Split across a small client base | Fully yours, and it compounds |
| Biggest structural risk | Junior delivery behind a senior pitch | Bus factor: one person's capacity | Hiring risk plus a single perspective |
What does each option actually cost?
Think in relative terms, because absolute numbers vary wildly by market. A mid-sized agency retainer and a senior consultant's fee often land in a similar monthly range; what differs is what the money buys. The retainer buys a slice of a team and its process. The consultant fee buys a larger slice of one senior person's attention.
A senior in-house hire is usually the largest total commitment once you count salary, benefits, tools, recruitment and months of ramp, though at sustained scale it becomes the cheapest per hour of work delivered. A junior in-house hire is the cheapest on paper and routinely the most expensive in practice, because the tuition for their learning curve is paid out of your ad budget.
Some agency pricing is a percentage of ad spend. Not a scandal, but the commercial incentive points toward your spend growing. Ask how it is aligned with your cost per outcome, not just your volume.
When does an agency genuinely win?
When your bottleneck is throughput. If you are running meaningful budgets across four or more channels, need a steady flow of creative production, want coverage that survives one person's holiday, and value established process over bespoke judgment, an agency is the right tool. The good ones industrialise the repeatable parts of UA better than any individual can.
Agencies also win on redundancy and tooling. You get media buyers, designers and analysts under one invoice, plus enterprise tool access most single companies would not buy alone. For a team with strategy handled internally, that bundle is hard to beat.
When does a consultant genuinely win?
When your bottleneck is judgment. If the question is why costs doubled, whether your attribution can be trusted, which channels deserve the budget, or how to structure a testing programme, you need senior pattern recognition more than production volume. A consultant is the only model where the person diagnosing your account is guaranteed to be the senior person, because there is nobody else.
Consultants also win on speed and on independence. There is no onboarding machinery, and no incentive tied to your spend going up. And a consultant is often the right auditor of the other two models: a senior outside eye on an agency's work, or a mentor multiplying a young in-house team, is frequently the highest-leverage version of the role.
When does in-house genuinely win?
At sustained scale, and over time. Once UA is a permanent, daily, high-stakes function, nothing beats an owner who lives inside your data, sits next to product, and accumulates context that compounds month over month. External partners rent you expertise; an in-house team builds it as an asset you keep.
In-house also wins on integration. UA decisions increasingly depend on product decisions: onboarding flow, pricing, retention. If your growth problems are as much product as media, an employee who can walk over to the product team is worth a great deal more than an external partner waiting on a reply.
What are the honest failure modes of each?
Every model fails in a characteristic way, and you should choose with eyes open:
- Agency: the senior pitch, junior delivery problem. The people who win the account are rarely the people who run it. Add the volume incentive where fees scale with spend, plus playbooks templated across a roster, and a mediocre agency can be busy every day while your cost per outcome quietly worsens.
- Consultant: the bus factor. One person means one calendar, one holiday, one bandwidth ceiling, and knowledge that leaves when the engagement ends unless documentation is contractual. A consultant who will not put handover in writing is planning to be indispensable, which serves them, not you.
- In-house: hiring risk and a single perspective. You are asking someone to interview for a skill you cannot yourself evaluate; a wrong senior hire costs a year. Even a right hire sees only your account, so their patterns stop updating, and the best ones eventually leave for a bigger account, taking the context with them.
What should you ask before hiring each?
- Agency: Who exactly will work on my account each week, and what is their seniority? Can I speak to them before signing?
- Agency: How are your fees structured, and what happens to your revenue if my spend goes down but my results improve?
- Agency: Tell me about an account you resigned or advised to spend less. What triggered it?
- Consultant: What happens when you are unavailable, and what does handover look like if we stop working together?
- Consultant: How many clients do you serve at once, and where would I sit in that queue?
- Consultant: What is explicitly outside your specialism that I would still need covered?
- In-house: Who will evaluate and mentor this person, given nobody internally has done this job?
- In-house: What is the plan for the first 90 days, and what tools and budget authority will they actually have?
- All three: Show me, concretely, how you would know within eight weeks whether this is working.
When should you NOT hire a consultant like me?
Honestly, in at least four situations. First, if your monthly ad budget is still small, advice is a luxury; put the money into spend and learn from your own data first. Second, if your problem is production throughput, dozens of creatives and daily optimisation across many channels, you need an agency's capacity, not one person's judgment. Third, if UA is already your core, permanent function at real scale, hire the senior operator in-house and let the asset compound. Fourth, if your growth problem is really a retention problem, no acquisition help of any kind should be bought until the bucket stops leaking.
A consultant is the right call in a narrower band than consultants like to admit: when the stakes of your spend have outgrown your internal judgment, but not yet justified a full senior team. That band is where I do my best work, and outside it, one of the other two models will serve you better.
FAQ
Is a consultant cheaper than an agency?
Often comparable in monthly cost, different in what you get. The agency fee buys a team's hours and process; the consultant fee buys concentrated senior attention. For judgment-heavy problems the consultant is usually better value; for production-heavy workloads the agency is. Price per outcome, not per invoice.
Can I combine a consultant with an agency?
Yes, and it is a genuinely strong pattern: the agency executes, the consultant sits on your side of the table auditing measurement, strategy and the agency's own reporting. It removes the awkwardness of asking an agency to grade its own homework. Just make sure roles and decision rights are explicit.
When is it time to move UA in-house?
When UA is permanent, daily and central to the business, and the budget justifies senior full-time attention, typically once external fees approach what a strong senior hire would cost. Many teams sequence it: consultant or agency first, then hire in-house and have the external partner help ramp them.
What is the biggest red flag when hiring an agency?
The seniority switch: senior people in the pitch, junior people on the account. Ask who touches your campaigns weekly and meet them before signing. A close second is any fee structure that rewards your spend growing regardless of what happens to your cost per outcome.
Should my first UA hire be junior or senior?
Senior, or supported. A junior first hire with no mentor learns on your ad budget, which is the most expensive classroom there is. If you cannot afford senior full-time, a junior hire plus fractional senior oversight, from a consultant or a strong advisor, beats either extreme.
How long should a consulting engagement last?
Long enough to fix the problem and hand it over, not indefinitely. Diagnostic work runs weeks; a rebuild of measurement and testing usually runs a few months; ongoing fractional oversight is legitimate but should be a choice, not a dependency. A good consultant makes themselves progressively less necessary.
If you are weighing these options right now, I run a free App Growth Review: a short, no-pitch look at your live campaigns, your measurement, and where the budget is leaking. If the honest answer is that you need an agency or a hire rather than me, the review will tell you that too.
Related reading: Why GCC apps waste paid media budget, App install vs event-optimised campaigns and How to run a UA teardown on your own app.