Decision tree
Which of your 500 posts should get a link first?
A tree you can run on a spreadsheet export. Start at the top, answer yes or no, and you end on one of four outcomes: link now, link later, improve, or cut.
Pull pages, impressions and clicks from Search Console for the last 90 days. Add word count from your crawler. Everything below runs on those three columns.
The tree, one question at a time
Does the page have impressions in 90 days?
No impressions at all means Google has not shown it to anyone. Go to branch D. One or more means keep going.
Is it a near-copy of another page?
Two posts targeting the same phrase split every signal. Merge them with a redirect, then ask the question again for the survivor.
Does it sit between positions 8 and 30?
Yes: this is branch A and your best use of money. A page that already shows up is cheaper to nudge than one that never appears.
Is it ranking above position 8 with clicks?
Then leave it alone for now. Links there protect little and cost the same as links elsewhere.
Is it evergreen or tied to a date?
Evergreen pages go to branch B. Dated news and event posts go to branch C, which usually means no spend.
The four outcomes
| Branch | Page looks like | Action | Example spend |
|---|---|---|---|
| A: link now | Positions 8–30, steady impressions, answers the query well | Buy 2–3 dofollow placements aimed at it, one anchor each | $30–$90 across three rows |
| B: link later | Evergreen, few impressions yet | Improve internal links, wait four weeks, recheck | $0 for now |
| C: skip | News, event or seasonal post | Let it age; link only if it will return next year | $0 |
| D: improve or cut | No impressions in 90 days, or a near-copy | Rewrite, merge or noindex. No outside link. | $0 |
Reading the donor row once you know the branch
Branch A pages deserve a matching donor. Filter the marketplace by your niche and language first, then sort by organic traffic. A site with real visitors tells you more than any single score. Skip the rest.
Domain Rating is a quick filter, not a verdict. Ahrefs computes it, and a DR 50 site in a dead niche is still a poor neighbour for your page, so open a few of the donor's articles before you add the row. Trust your eyes.
When you decide to buy dofollow backlink placements, remember one thing. Prices scale with authority, traffic and niche, from $1 to about $500. For a branch A page at positions 12–20, three mid-priced rows usually teach you more than one expensive one. Placements are permanent and dofollow, and nobody can promise movement. Test small.
Why the tree starts with impressions
Impressions are the cheapest honest signal you own. They say Google has already tried the page on real searchers, and that is more than a word count or a plugin score can tell you. Start there.
A catalogue of 500 posts hides a skew. Look closer. Usually a few dozen pages produce most of the clicks, while the long tail sits at zero. The mistakes and fixes list shows what happens when a batch of links goes to that long tail by accident, and the myths page covers why volume itself is not a signal. Run the tree again every quarter, since positions drift and a branch B page may be ready for its first link by then.
Rule of thumb
Fix the cut branch before you spend on branch A
If a third of your archive lands in branch D, that is the finding. Prune or merge first, since Google's March 2024 update targeted scaled content that exists mainly to rank.
- Merge doubles with a 301
- Noindex pages with no demand
- Then buy links for branch A only