Decision tree
Run a backlink check, then follow this tree to your target page
Seven questions, each with a yes branch and a no branch. You finish with one URL and a reason for choosing it.

Do the check first. Open Google Search Console, find the Links report, and note the number of linking sites for each page you might promote. Ahrefs Webmaster Tools and Semrush show similar counts from their own indexes. Pick pages, not hunches.
Write the numbers down. They are your baseline, and without one every later claim is a feeling. If you want a worked example of the bookkeeping, the before and after sketch has one.
The tree, question by question
- 1
Q1. Does the page say who you write for and what you charge, or at least how to ask?
Yes: go on to Q2. No: stop buying anything. Fix the page first, because a link only delivers visitors to whatever is there.
- 2
Q2. Does it show at least five clips in your niche?
Yes: go on. No: add clips, even small ones. A lone sample page persuades nobody who arrives from a stranger's article.
- 3
Q3. Is the page on its own URL, not a tab inside a platform profile?
Yes: go on. No: a profile you cannot control is a poor target; move the portfolio to a page you own.
- 4
Q4. Do you already have more than ten sites linking to it?
Yes: your next link should go to a different page, such as the rate card. No: the clips page is a fair first target.
- 5
Q5. Do you write about a single topic or three?
Single topic: pick donors in that niche and keep the anchor plain. Three: choose the strongest one and run a separate small plan for each later.
- 6
Q6. Can you say, in one sentence, why a reader of the donor site would click?
Yes: order. No: the donor is the wrong fit; go back to the board and filter by niche and language again.
- 7
Q7. Is your spending ceiling written down?
Yes: fill the basket up to that line. No: write it now. The basket is free to fill and free to empty until you pay.
Where the branches usually end
| Your site today | Likely target | Leave alone for now |
|---|---|---|
| Clips page plus a blog | Clips page | Individual blog posts |
| One-page site with everything | The home URL | Anchors with exact service phrases |
| Hire-me page, no clips yet | Neither; add clips first | The whole link plan |
| Newsletter with archive | About page | Single issues |
| Portfolio on a Wix or Squarespace template | The portfolio page itself | Template pages with a footer-only contact link |

Two traps the tree cannot catch
The first is the vanity target. Writers love sending links to the home page because it feels like the front door. Yet a home page that mixes services, a blog feed and a newsletter pop-up gives the visitor too many exits.
The second is the anchor. If every placement says the same commercial phrase, the pattern looks planted. Vary it, and let some links use your name or the plain URL.
Neither trap needs a tool. Both need ten minutes and a willingness to be bored. Ahrefs, Semrush and Search Console can list your current anchors, and the order walkthrough shows where you set the keyword for a new placement.
Copy, leads or channel: which one to fix first
Before a link, a writer usually asks a different question: is the problem the page, the leads or the way people hear about you? A bought link helps only the first of those, so sort the problem before you spend.
The order below is practical. Start with the cheapest fix and work outward, because a link should not be the answer to a page that still needs its first paragraph rewritten.
Start with the page when visitors leave quickly
If people arrive and leave within seconds, the page fails before any link can matter. Read the first screen as a stranger would, and ask whether the offer is clear within a few lines.
Say your clips page opens with a long biography and buries the samples. Moving the samples up costs nothing, and it changes more than a new placement would. Fix that first, then measure again.
Look at the leads when the page holds up
When visitors stay but few of them get in touch, the leads are the weak point. Check whether the contact route is easy to find and whether the page says plainly who it is for.
A page can attract the right people and still ask for too much. A short form and one clear next step often beat a long one that makes the reader work to reply.
Change the channel last
Outreach channels are the most tempting place to spend and the least useful until the other two are working. A new channel multiplies whatever the page already does, good or bad.
Treat a link as one more channel, and add it only after the page and the leads have been checked. The tree then tells you which page should take it.
Reading buyer intent without cold outreach
Many writers ask how to find high-intent leads without reaching out first. The honest answer is that intent shows in what people search for and where they land, and a page can be built to catch it.
A link does not create intent. It can bring a reader to a page that already matches the question they were asking, and that match is the only place where intent turns into an enquiry.
Notice the question a visitor came with
A reader who lands from a donor article is answering a question about that article's subject. Your page should pick up that thread, not start a new one about yourself.
Picture a reader of a food blog who meets a link to a copywriter's menu samples. The match is close, and an enquiry from that reader is far more likely than one from a random feed.
Check what the visitor does after arriving
A visit that ends in one second tells you the landing screen failed. A visit that scrolls through the samples tells you the page is doing its job, even before any enquiry arrives.
Note which section a reader reaches before leaving. That single observation says more about the next target than a total visit count, and it costs nothing to record.
Pick the page that answers that question
Once the question is clear, choose the page that answers it most directly. For many writers that is the clips page, and for others it is a rate card that buyers hunt for.
Keep the answer to a few lines before the samples begin. A reader who gets an answer in the first screen will read on, and a reader who does not will leave.
Before you choose the target page
- A backlink check on each candidate page, saved with the date you ran it
- The one query you want that page to win, written down in full
- The first screen of the page, read as a stranger would read it
- The number of clips or samples visible without scrolling
- A note on whether the page will change in the next three months
- The spending ceiling, set before you open the board

When every branch says no
Sometimes the tree ends with no clear winner. Every answer points to a fix on the site. None points to a link. That is a useful outcome, even if it feels like a dead end.
Rivals in your niche can make the picture harder to read. The agency myths page explains why a bigger name is the wrong comparison for a solo writer.
Fix the page before the link
If the tree stops at Q1 or Q2, the page needs work. Add the clips or the answer first, then come back to the tree once the page can hold a reader.
Work in small steps and save each version with a date. Later, you will be able to tell which change mattered, and which one was only a mood.
Keep the first basket small
When you are unsure, start with a basket small enough that a mistake costs little. A few cheap, relevant rows teach you more than one large order aimed at the wrong target.
Set a date for the next review before the first placement goes live. Then the decision waits for facts rather than for a mood.
Four notes to make after the first placement
The tree ends at a link, but the work does not. These four notes decide what the next run of the tree will look like.
- 1
Write down the donor and the date
Record the donor domain, the day it went live and the exact URL it points at. Without that line, the next decision rests on memory.
- 2
Read the query over matching windows
Compare the same query over the same length of time before and after. Do not set a holiday week against an ordinary one.
- 3
Check the words around the link
Read the sentence that holds the link. If it sounds like a paid insertion, a human reader may notice before any search engine does.
- 4
Run the tree again with the new facts
The answer may still be the same page, or the next link may need to go somewhere new. Let the notes decide, not the last result you saw.
Placements are sold as permanent dofollow, so plan as if the link stays. A thirty-minute audit beforehand is cheap insurance against pointing it at a page you will redesign next month.