That deal works because links are one of the last things in marketing you buy without seeing. You'd never approve a media buy without the placement list. You approve link buys blind every month.

What you're actually paying for

Strip a $4,000 monthly retainer down and there are five line items underneath it.

Someone builds a prospect list. Someone emails those prospects and follows up four times. Someone writes an article. Someone pays the publisher. And the agency keeps what's left.

The last two are the interesting ones. Publisher fees are a real cost and they vary enormously, from about $40 on a small niche blog to well past $1,000 on a site with genuine editorial standards and real traffic. Agencies rarely show you that number. When the invoice says "10 links, $4,000," you're being told $400 a link and nothing about the split between what the site charged and what the agency kept.

That split is the whole argument. On cheap inventory, an agency buying at $60 and billing at $400 is taking 85% of your budget as fee. On expensive inventory the same agency might take 20%. You have no way of knowing which one happened, because the input price is the one number never printed.

The three-week outreach myth

Agencies justify the markup with outreach labour, and for genuinely earned links that's fair. Digital PR, HARO-style sourcing, original data that journalists pick up: those take real work and real relationships, and the good agencies are worth every cent.

Most retainers aren't buying that. They're buying placements on sites that already have a rate card, already say yes to everyone, and already sit on three other marketplaces at a published price. The "outreach" is an order form. You're paying an hourly-rate story for a transaction.

There's an easy way to test which one you're getting. Ask your agency for the last twenty domains they placed on and check them in Ahrefs. If most of them have a "Write for us" page, a Sponsored Post rate card, or a footprint of guest posts from unrelated industries, you bought inventory. Inventory has a market price, and you can look it up.

The audit that takes an hour

Four things to pull before your next renewal conversation. None of them require the agency's cooperation.

Export every link they've built and check organic traffic against domain rating. Not DR on its own. DR measures a backlink profile and says nothing about readership, so a DR 60 domain pulling 200 monthly visits is an old profile attached to a dead site. If a third of your placements sit in that gap, the agency has been buying on the cheap end of a list and billing on the expensive end of a report.

Check how many are still live. Take the links built twelve to eighteen months ago and open them. Sites get sold. Sponsored posts get pruned. A new SEO consultant arrives and nofollows every contributed article on the domain in an afternoon. Nobody in this category tracks the survival rate, which is precisely why it's the number that separates a cheap supplier from an expensive one. A $60 link that dies in month four cost more than a $250 link that doesn't.

Check the anchor distribution. Export the anchors and count them. If exact-match has crept past 10–15% of your profile because each placement was decided one at a time, that's a risk your agency created and didn't flag.

Check the outbound link count on the pages you're on. Three do-follow links to three unrelated commercial sites means the publisher sold that page to you and two other people.

An agency worth its retainer will already have these numbers. Ask for them and see how long the answer takes.

What an agency is genuinely good at

Three things, and it's worth naming them properly because the pitch here isn't that agencies are useless.

Strategy is real. Deciding which pages need links, which anchors to use, and how aggressively to move is a judgement call, and getting it wrong costs more than the links do. A good strategist will stop you pointing forty exact-match anchors at a commercial page and tanking it.

Volume management is real. If you need 40 placements a month across six client sites in five niches, coordinating that is a job. Someone has to chase publishers who go quiet, replace the ones who bail, and keep the reporting straight.

Access is sometimes real. A few agencies genuinely have relationships that get you into places that don't sell. Very few. If yours does, they'll be able to name the outlet.

Everything else on the invoice is procurement, and procurement is the part you can do yourself in an afternoon once the prices are visible.

The version where you can see the list

NO-BS Marketplace is a link building marketplace with 110,000+ publishers in it. Every listing carries the Ahrefs domain rating, the organic traffic figure, the spam score, the niche, the languages the site publishes in, and the price you'd pay. Turnaround is stored per publisher and shown on the listing, in weeks, because that's how publishers actually quote it.

You filter. You shortlist. You buy the specific sites you picked, one at a time or forty at once. Nobody sends you a "recommended package."

The parts an agency does that are actually labour are still handled. Writers produce the article to your brief, at the length you choose, with as many revision rounds as you want. Outreach and the publisher relationship sit with us. If you'd rather supply your own copy, bring it, and you don't pay for content you didn't need.

What changes is that the site list arrives before the money does.

Start with the marketplace if you want to see inventory, or link building services if you'd rather someone ran it for you.

Where the money goes instead

Two numbers a month is the honest way to compare. What did the placements cost, and what did the coordination cost.

When those are separate, the decision gets easy. A $900 site is worth $900 if the traffic and the editorial standards are there, and you can check both before you spend. A $60 site is worth $60 and no more, and no amount of relationship talk turns it into a $400 link. Once you can see which is which, you stop arguing about retainers and start arguing about sites, which is the argument worth having.

The practical effect for most buyers we talk to is that budget shifts upward, not downward. People who were buying twelve mediocre placements at a blended rate move to five properly good ones and one flyer, because for the first time they can see what "good" costs.

Work an example. A $4,000 month at a blended $400 a link gets you ten placements and no visibility. Split the same budget into $2,800 of media and $1,200 of management and the media buys you four placements on sites with real readership at $500, plus five relevant supporting links at $160. That's nine links instead of ten, and you know which four are doing the work.

Pricing detail is on the link building pricing page, including what drives the spread between a $50 site and a $900 one.

The three clauses to fix at renewal

Whatever you decide about the agency, three things belong in the contract and usually aren't.

A reporting spec, not a report. Live URL, anchor text, and the publisher's DR and organic traffic at the date of placement. Every month, per link. A report that says "10 links built, DR 45 average" is a summary of a summary, and the averaging is doing work.

Quality parameters with a traffic floor in them. Most agency contracts specify a minimum DR and nothing else, which is an instruction to buy old domains. Specify minimum organic traffic, maximum spam score, niche relevance, and the anchor split. A supplier optimising against your schedule will find whatever gap you leave.

What happens to the links when you leave. Nobody writes this down and it produces an unbillable month for somebody. Placements already live stay live, you're not entitled to have them removed or transferred, and any replacement obligation either survives termination or it doesn't. Decide which, in writing, before you need to know.

Every placement is covered for 12 months

A link that disappears in week six was never a link, and this is the failure mode nobody in the category discusses honestly. Publishers get sold. Pages get pruned. Someone's new SEO consultant runs a nofollow sweep.

Every marketplace placement carries a 12-month live-link guarantee. If a link drops inside that window, we replace it on a publisher of equal or better DR band. If we can't place a replacement within 30 days, that placement is refunded. You can decline a replacement and take the refund instead.

The DR band is measured at the time you order, which matters and gets argued about, so it's worth saying plainly: if a publisher's DR slides after your order, that isn't a covered event. The guarantee is about our delivery, not about Google's opinion of a site six months later.

What it doesn't cover: anything about your own rankings, your traffic, your domain rating, or whether a page gets indexed. Nobody can promise those, and the ones who do are quoting you a number they have no mechanism to hit.

For agencies buying on behalf of clients

If you're the agency, the same inventory is available white label. Your brand on the reporting, your pricing to the client, our fulfilment underneath. Most of the agencies using it were already reselling somebody's inventory. The difference is knowing which sites are in the order before the client asks.

What changes on your side

Two things get harder when you stop paying someone to think about it, and it's worth knowing before you switch.

You own the selection. That's the point, and it's also an hour a month of actually looking at sites rather than approving a summary. People who won't spend that hour should keep the agency, and there's no judgement in that.

You own the anchor plan. Write the split down once, branded against partial against exact, and buy against it. Anchor decisions made one placement at a time drift toward exact-match, every time, because exact-match always looks like the most useful option in isolation. An agency was doing this for you, well or badly, and now you are.

Everything else stays where it was. Writing, submission, the editor round, chasing a publisher who's gone quiet: all of that is still handled.

When you should still hire an agency

Hire one when the strategy is the hard part. If you don't know which pages need links, what your anchor distribution should look like, or how to sequence a campaign against a competitor who's four years ahead, that's a consulting problem and a marketplace won't solve it.

Hire one for genuine digital PR. Getting quoted in a trade publication or landing a data study in a national outlet is relationship work with a real hit rate, and it isn't inventory.

Hire one if you'd rather pay for the absence of the job than do the job. That's a legitimate purchase and there's no shame in it.

Don't hire one to buy placements on sites with published rate cards and then not tell you which sites. That's the version worth walking away from.

What we don't do

Worth being explicit, because a page like this that only lists strengths is a page nobody believes.

We don't write your strategy. Which pages need links, in what order, at what pace, against which competitor: that's a judgement about your business and we'd be guessing. It's the main thing a good agency sells and it's a real reason to hire one.

We don't do earned digital PR at scale. Getting a data study into a national outlet is relationship work with an unpredictable hit rate, and it isn't inventory. We sell placements on named outlets at named prices, which is a different product.

And we don't promise anything about your rankings. Nobody can, and the specificity of a promise is usually inversely related to whether the person making it has ever had to honour one.