Manual outreach still works, but it's no longer where link programs fail.
Finding prospects, writing pitches and sending follow-ups has been commoditized. Marketplaces and vendors will sell you placements by the dozen, and AI will happily write the pitch. What hasn't been commoditized is judgment: deciding which placements are worth paying for, and checking afterward that you got what you paid for.
The old way
Classic outreach looks like this: 1. Search for "best [topic] tools" or "[topic] resources" 2. Find the pages that rank 3. Find a contact 4. Write a cold email: "Found your article on [topic]. We built something similar. Thought you might find it useful." 5. Probably get ignored.
This is link building as it's been taught for a decade. It works, but it's slow and needs a person to do it.
What replaced it, and the trap inside it
Most agency link programs now run through vendors and marketplaces. At the agency where I lead SEO, that's Loganix. You pick sites from an inventory, choose a service tier, supply a target URL and anchor, and the vendor handles placement.
That solves the volume problem and creates a new one: nobody is forced to look closely at what's being bought.
Two things I've seen firsthand:
- A recurring subscription shipping without review. On one recurring tier, the vendor emails an approval request for its site list, and after a few unanswered messages it proceeds with its own selection. A $1,900/month cycle went out with zero vetting. I cut that subscription.
- A campaign that didn't hold. Before approving a vendor reorder, I audited the earlier campaign: only 20% of the links were still in place. The reorder stopped there.
Neither problem shows up on the invoice or the delivery report. Both were obvious as soon as someone checked.
What vetting actually looks like
Every placement has to be defensible on metrics, on relevance, or both, with a loose check that the price is in line with what those metrics buy elsewhere. A cheap, on-topic, low-authority site (a golf blog for a golf club) is a keep. An expensive, off-topic, mid-authority site is a swap, even if it sounds premium.
The gates I run before anything goes in a cart:
- Open the site. Check the language and market. A Spanish-language site is a bad fit for a US English page, no matter what its authority score says.
- Read the publisher's guidelines. Some sites listed as taking guest posts have quietly stopped.
- Re-price after choosing the service. Listed prices go stale. One placement listed at $708 repriced to $1,879 once the guest post option was selected.
- Set a metrics floor. Authority or real traffic, plus dofollow, with relevance able to override a weak score.
- Ignore the vendor's "match" score. In my experience it doesn't discriminate between good and bad fits.
The pattern behind every failed placement I've reviewed: it was publisher fit or price, never a lack of authority.
Anchors are part of the vetting
A program that buys links without an anchor plan drifts toward exact-match commercial anchors, because that's what feels productive. The framework I use per cycle is roughly 40% branded, 30% natural, 20% partial-match and 10% exact commercial, with no more than one commercial anchor per money page per cycle. A running ledger of every anchor and target catches repeats before checkout.
What this means for freelancers
If you sell link building, "I'll find 50 prospects and send outreach" is the commodity part. The services that will survive provide judgment:
- Understanding which sites matter for a niche
- Showing the client why each placement is worth its price
- Auditing what's still live after the invoice is paid
- Saying no to a reorder when the last campaign didn't hold
The hybrid approach
I'm not saying abandon outreach. The best strategy is hybrid:
Use vendors and automation for:
- Sourcing inventory and candidates
- Placement logistics and tracking
- Follow-up sequences
Use your own judgment for:
- Approving every site before it's bought
- The anchor and target plan
- High-value prospects where a real relationship beats any marketplace
- Retention audits before renewing anything
The bigger shift
This is part of a larger trend in SEO. Every part of the job is being chunked into:
- Commodity work (automatable, cheap)
- Expert work (requires judgment, expensive)
In link building, sending emails is the commodity. Deciding what's worth buying, and proving afterward that it held, is the expert work. If you can show a table that defends every placement, you're the one making the automation safe to use.