A founder asking this has usually already done the salary maths. They work out that an in-house marketer plus an SDR is 25,000 a month and the agency wants 10,000. So the agency wins and the subject is closed.
That is not a good comparison, because the two do not deliver the same thing. Here is how they differ and when to pick which, including the situations where I say no.
An in-house team wins when you already know who your customer is and have a repeatable sales process to copy. An agency wins when you still have to work out who your customer is and do not want to pay for a year of learning.
At seed to Series A, the second situation is usually the true one, even when the founders are convinced it is the first.
An in-house marketer delivers presence. They are there every day, know the product better than any outsider within three months, sit next to sales and overhear the calls. That is not replaceable.
An agency delivers patterns from a dozen companies and a process that already exists. It will never know your product as well as your own person. In exchange it has seen what happened at companies that were where you are now.
The practical difference shows in month one. A new hire is learning the product. An agency is starting a process that has worked before. With hiring you pay for the learning; with an agency you pay for someone else having already paid for it.
Salary is usually less than half the bill, and the other lines surprise people.
Recruitment takes six to twelve weeks from advert to first day. Notice periods add another one to three months. Then comes onboarding, during which the new person delivers less than they cost for the first quarter.
On top of that come the tools you have to buy anyway, because a person without a database and a CRM cannot do ABM. That is the same monthly spend covered in the costs article.
And the most expensive line, which nobody puts in the spreadsheet: the risk of a bad hire. A B2B marketer with real ABM experience is a thin market. Get the first person wrong and you lose two quarters, not two months.
| Criterion | In-house team | Agency |
|---|---|---|
| Time to first activity | 3 to 6 months | 2 to 6 weeks |
| Product knowledge | Very high after a quarter | Limited, always |
| ABM experience | Depends on one person | Patterns from many companies |
| Cost in year one | Higher, recruitment and ramp | Lower |
| Cost in year two | Lower | Higher |
| Risk | A bad hire costs two quarters | Ending a contract costs a month |
| What stays in the company | A person and their knowledge | Only what you write down |
The last row matters most and is the one most often skipped when choosing an agency. With hiring, the knowledge stays by itself, because the person stays. With an agency, only what can be reproduced without them stays: the account list, the definitions, the playbooks, the dashboards, the CRM processes.
That is why I build so the team can run the programme alone afterwards. Not because it sounds good in a proposal, but because otherwise the client pays twice for the same thing.
Three situations where I tell a founder not to take an agency.
First: the product has no repeatable customer yet. With ten customers from ten industries there is nothing to build a list from. You need someone inside talking to customers daily, not an ABM programme.
Second: average contract value below 30,000 a year. At that level the maths does not close and it is better to build inbound with your own hands.
Third: the leadership team does not agree on what a good customer is. An agency cannot settle that, because it is an ownership decision. I have seen programmes stall at exactly this point with the invoice already paid.
You do not have to choose between the two in that order. You can do one after the other.
The agency builds the programme and writes it down so it can be handed over. Meanwhile you recruit the person who will take it over, and that person walks into a working process rather than a blank page. You also recruit more easily, because you have something to show a candidate.
That is the model I work in most often, which is why my engagement ends with a handover rather than a renewal. The first three months are a pilot, after which the client runs the programme alone or keeps lighter support.
Four questions that filter out most of them.
Ask to see a playbook from another engagement, with client names removed. If there is nothing to show, the agency runs campaigns rather than building systems.
Ask what remains after the contract ends. An answer of “a report and a presentation” means that in a year you are back where you started.
Ask how they are paid. Per lead or per meeting forces volume over fit and recreates exactly the sales-marketing conflict ABM is meant to remove.
Ask when they say no. If the answer is never, they take anyone who pays, and they will work out your case in month three.
The agency-side numbers are in what ABM costs, and you can test both options on your own figures in the ABM ROI calculator. The full model is in the guide to Account-Based Marketing for B2B SaaS.
Daniel Tupczyński runs MOAT Marketing, an Account-Based Marketing partner for B2B SaaS companies. MOAT builds and runs ABM systems alongside the client’s own sales and marketing team, from market research and prospecting into Tier 1 accounts through to insights for product and strategy, so the know-how stays in the company. He also runs marketing at a B2B SaaS company.