This decision usually gets made on a spreadsheet comparing an agency retainer against a salary, and that spreadsheet is almost always wrong.
What the spreadsheet misses
On the in-house side people compare the retainer to base salary and stop. The real number includes payroll taxes and benefits, recruiting time or fees, tooling and subscriptions, the manager hours spent directing the work, and the ramp period before the person is productive.
It also includes the risk of a bad hire, which for a small company is not a line item, it is a bad six months.
On the agency side people forget that an agency still needs direction. Nobody is coming to run your marketing while you think about something else. If there is no one internally who can make decisions and give timely feedback, an agency will produce work that drifts, and you will conclude that agencies do not work.
The three questions that actually decide it
Is this a capability you need to own forever
Some things should live inside the company permanently. Knowing your customers. Owning the brand voice. Understanding why deals are won and lost. Those should never be fully outsourced, regardless of how good the partner is.
Other things are specialist crafts you need at intervals. Video production. Paid media at scale. Brand identity work. Hiring a permanent person for a craft you use twice a year is expensive and, worse, boring for the person you hired.
How many different skills does the work require
This is the one that surprises people. A single marketing hire is one person with maybe two strong skills. The work in front of most companies requires strategy, copy, design, web, paid media, analytics, and video.
Hire one person and either they do three things well and four things badly, or they spend their budget hiring freelancers, which is running an agency with extra steps and less experience.
Who is going to direct the work
Both models need an owner. If the answer is "the founder, in between everything else," that pushes toward an agency with strong process, because they will supply structure the business does not have.
If there is a capable marketing lead already, an agency becomes far more effective, because they get decisions quickly and stop guessing.
The failure mode is the same on both sides. Nobody is steering, so the work goes wherever the last conversation pointed it.
What we see work by size
Rough, and there are exceptions.
- Under 20 people. A partner for most execution and one internal person who owns the relationship and knows the customers. Full in-house teams at this size usually mean one overloaded generalist.
- 20 to 100 people. A hybrid. One or two internal people owning strategy, brand, and the parts that need daily attention. Specialist work outside.
- 100 plus. Enough volume to justify a real internal team, with agencies used for surges, specialist craft, and outside perspective.
The signals it is time to bring something in-house
- You are paying an outside partner for something that happens every single day.
- Turnaround time is now the bottleneck, not quality.
- The work requires context that takes longer to explain than to do.
- You have enough volume that a full-time person would be genuinely busy.
The signals it is time to bring something out
- Your one marketing person is doing four jobs and none of them well.
- You need a skill for a project, not a year.
- Quality has plateaued and nobody internally can see why.
- You are about to hire for a craft you do not know how to evaluate.
That last one deserves attention. Hiring a specialist you cannot assess is a coin flip, and the failure takes a year to become obvious.
The honest version
We are an agency, so read the above with that in mind. The clients we do our best work with all have someone internal who owns the relationship and can make decisions. When that person does not exist, we say so, and sometimes the right advice is to hire that person first and call us afterward.
That is not a sales position. It is just what makes the work good.