Agency, freelancer, or in-house: how to actually decide
The comparison is usually monthly cost: a salary against a retainer against an hourly rate. That is the easiest thing to compare and close to the least useful.
Three questions get you further.
How many different skills does the work actually need?
Count them honestly.
A paid media program needs media buying, creative, landing pages, tracking, and someone who can read the CRM to tell whether the leads converted. That is four or five skills, and one hire will be strong in one or two.
One or two skills, sustained is the in-house case. Hire it, keep it, build the depth.
Four or more, needed at once is the agency case. Not because agencies are better, but because you are buying a set of people who already work together, without hiring five of them.
One skill, temporary is the freelancer case, and it is a genuinely good option that gets skipped because it feels less serious.
How long will you need this?
Under six months: freelance or agency. Hiring for a project is how you end up managing someone out of a role that no longer exists.
Six to eighteen months, and it will change: agency. This is the range where you do not yet know which skills you will need in month twelve, and an agency can shift the mix without a hiring cycle.
Longer, and it will be central to how the business runs: in-house, eventually. The knowledge compounds and you want it staying in the building.
Most companies get this backwards. They hire for the temporary thing because a headcount was approved, and outsource the permanent thing because a retainer is easier to cancel.
How much of the knowledge do you want to keep?
The real difference, and the one nobody puts in the comparison.
An in-house person accumulates knowledge about your business that stays when the project ends. An agency accumulates knowledge about your business that leaves.
That is not an argument against agencies. It is an argument for being deliberate: insist on documentation, keep accounts in your own name, and treat the handover as part of the scope rather than a courtesy. A good agency will already work this way. Asking costs nothing and tells you a lot.
The hybrid that usually wins
For most mid-sized companies the answer is not one of the three.
One strong in-house owner, plus external specialists. The internal person owns the strategy, the relationships, and the institutional memory. Specialists come in for the work that needs depth you cannot justify hiring.
This fails in one specific way: hiring the internal owner too junior. Someone who cannot evaluate the specialist's work becomes a message-passer, and you get the cost of both models with the benefit of neither.
The question underneath all of this
Is this a capability we need to own, or a problem we need solved?
Capabilities you own get hired. Problems you need solved get bought. Most of the confusion comes from treating a problem as a capability because a headcount happened to be available.