We are an agency, so treat everything below with appropriate suspicion. We have tried to write the version we would have wanted when we were on the client side, which means it includes the cases where hiring us would be the wrong call.
The decision turns on a single question: which costs in link acquisition are fixed, and which scale with volume? Fixed costs are the argument for an agency, because an agency spreads them across a roster. Variable costs are the argument for doing it yourself, because you are paying someone else's margin on them.
What is actually fixed
Four things, and they are more expensive than most teams expect.
A prospect index
Knowing which publications in your category accept contributed content, who edits them, what they have published before and what they have refused. Building this for one company takes six to nine months of work that produces no links at all. An agency amortises it across every client in the vertical.
Sender reputation
Cold outreach from a domain with no sending history goes to spam. Establishing an identity editors recognise takes months of low-volume, carefully written correspondence — and one enthusiastic person with a sequencing tool destroys it permanently. This is the most underrated line in the whole comparison.
Scoring infrastructure
Deciding which of two hundred candidate domains are worth pitching requires traffic verification, topical overlap scoring, footprint clustering and outbound-link hygiene checks. You can do this by hand for ten domains a month. You cannot do it by hand for two hundred.
Editorial relationships
The reason a pitch gets read in week one rather than week nine. These accrue slowly and belong to whoever built them — which, if you use an agency, is the agency. That is a genuine downside of the agency model and worth naming.
What is variable
Writing the drafts. Sending the pitches. Chasing replies. Recording placements. These scale roughly linearly with volume and carry an agency margin on top. If you are running fifteen to twenty placements a month indefinitely, that margin becomes the dominant cost.
The cross-over
Somewhere around fifteen to twenty acquired links per month, sustained for a year, the fixed costs stop dominating and in-house economics begin to win. Below that, an agency is usually cheaper per acquired link and dramatically faster to start.
| Situation | Usually right | Why |
|---|---|---|
| Seed, $1,500–$3,000/mo | Agency, or one asset instead | 5–10 links a month cannot justify a hire; the fixed costs would eat the entire budget |
| Series A, one category | Agency | Speed matters more than unit cost at this stage |
| Series B, 15–25 links/mo | Genuinely close | Run the numbers properly; a hybrid often wins |
| Series C+, sustained volume | Hire, keep an agency for PR | Margin on variable work now exceeds the value of shared fixed costs |
| One-off cleanup or audit | Agency, project basis | No reason to build capability for a task you do once |
Where a freelancer genuinely wins
Freelancers get dismissed too quickly in these comparisons. A good independent operator with existing relationships in your specific niche can outperform a mid-tier agency comfortably, because you are buying their relationships directly rather than an agency's average.
The trade-offs are real and predictable: no redundancy when they are ill or busy, limited capacity above roughly ten placements a month, and no infrastructure — you will be the one maintaining the tracking sheet. For a company needing six to ten quality links a month in a narrow vertical, that can be exactly the right shape.
The failure mode is hiring a generalist freelancer with no relationships in your category. At that point you are paying for someone to learn your industry using your budget, which is the worst version of every option.
The question that decides it
Ask yourself: in twelve months, will we still be acquiring links at this volume?
If yes, and the volume is above fifteen a month, start planning a hire — and use an agency in the meantime so the work does not stop while you recruit.
If no, or the volume is lower, an agency or a specialist freelancer will be cheaper per link and considerably faster to start.
The hybrid, which usually beats both
Most of our best engagements sit next to an in-house team that owns technical SEO, on-page and content. We take acquisition because it needs infrastructure that is expensive to build for a single company, and they keep everything that benefits from deep product knowledge.
Nobody has to win this argument. The division that works is: in-house owns what requires knowing the product; outsourced owns what requires knowing the publishers.
Three things people get wrong
Counting the agency fee against a salary. A $7,000 monthly retainer looks expensive next to an $80,000 salary until you add employer costs, tooling, the six-month ramp before the first link lands, and the fact that the hire will spend a third of their time on things that are not acquisition.
Assuming in-house means higher quality. It means higher product knowledge, which is not the same thing. An in-house team with no publisher relationships often ends up buying from the same marketplaces the cheap agencies use, having concluded that outreach "does not work".
Ignoring the exit. If you use an agency, ask what you keep. Gap analyses, prospect lists and content you paid for should transfer to you. If they do not, you are renting your own strategy, and the build-or-buy calculation looks very different.
The honest summary: below fifteen links a month, hire an agency or a specialist freelancer. Above it, hire a person and keep an agency for the digital PR work that needs relationships you will never build alone.