What a Content Marketing Agency Should Actually Do

If a content marketing agency’s proposal leads with a number of posts per month, you are looking at a production contract. That can be a legitimate purchase, but it is not a content marketing engagement, and the two get sold at the same price.

Volume is the easiest thing to sell because it is the easiest thing to measure. It is also the part with the least leverage.

What is the actual work?

The work	Where the value is	What you get without it
Choosing topics and defending the choice	Before a word is drafted	Your own list, written back to you
Interviewing your experts	Information gain	A summary of competitor articles
Structure for skimming	Subheads carry the argument	A wall of text nobody finishes
Distribution per piece	Where it goes after publish	A library with no doors
Maintenance of ranking pages	Compounding return	Slow decay nobody notices

Deciding what to write about, and defending the choice. Most of the value sits here. Which questions your buyers already ask, which of those you can plausibly rank for, and which are worth writing even though they will never rank because they close deals when sales sends them. An agency that takes your topic list and writes it has skipped the part you were paying for.

Getting the thinking out of your experts. The reason most B2B content is thin is that the writer had nothing to work from except competitor articles. Good process means recorded interviews with people who actually know things. It is slow, and it is the difference between an article that says something and one that summarises other articles.

If nobody has asked to interview your head of product, do not expect the output to contain anything proprietary.

Structure that fits how the page will be read. Nobody reads a blog post top to bottom. The subheads carry the argument, the first sentence of each section does real work, and the page has to be navigable at a skim.

Distribution. Publishing is not distribution. If the plan ends at “published to the blog,” you have bought a library with no doors. Where each piece goes afterwards should be in the plan, specific to the piece.

Maintenance. Ranking pages decay. Somebody has to go back into the top twenty pieces every couple of quarters and update them. This is unglamorous, returns better than most new production, and almost never appears in proposals because it does not feel like new work.

Volume is the easiest thing to sell because it is the easiest thing to measure. It is also the part with the least leverage.

How do you read a proposal?

Read the deliverables list and count how many items are quantities and how many are decisions.

A list that is entirely quantities, eight posts, four social variants, one newsletter, is a production agreement. Fine if that is what you need and you hold the strategy in-house. Expensive if you assumed judgment was included.

What to ask for instead

Ask for a topic map with reasoning attached. Not a calendar with dates, a map with a why against each cluster and an honest note on which ones will take a year to move.

Ask who they will interview inside your company and how often.

Ask what they will do with a piece in the ninety days after it goes live.

Ask what they will stop writing if something is not working, and how they will know. The same specificity test applies across evaluating any agency, and the brief format that makes good output possible is in the SEO brief that works.

The uncomfortable part

Good content work makes the client do things. Show up for interviews. Approve faster. Let a piece take a position a competitor might dispute.

Agencies that never ask anything of you are usually producing safe work, and safe work is why so much B2B content is indistinguishable from every other article on the same topic.