Link Building Vendor Scorecard and Placement Approval Brief
Buying links is mostly a buying problem, not an SEO problem. Two vendors send near-identical emails, one placement lasts 3 years on a site your customers read, the other disappears in 6 weeks from a page nobody visits, and nothing in the pitch told you which was which. This scorecard makes both vendors answer the same 9 questions before price enters the conversation, and the approval brief puts each placement in writing before anyone spends money.
- Format
- Editable Word .docx
- Length
- 5 variants · ~6 pages
- Price
- 100% free
- Setup
- Copy or download
Get the editable Word doc in one click.
★★★★★4.9·Free · No signup · Instant download
Link Building Vendor Scorecard and Placement Approval Brief
Vendor scorecard
Score every link vendor on the same rows, before you see the price. Most bad link buys happen because the pitch arrived with a price and a promise, and nothing to compare it against.
| Row | What to record | Score 0 to 5 |
|---|---|---|
| Market | [countries and languages the placements actually reach] | [score] |
| Topical relevance | [how close the site is to what you sell, in one sentence] | [score] |
| Traffic-source evidence | [what proof they can show that real people read the site] | [score] |
| Editorial control | [who decides the angle, the anchor and whether the piece runs] | [score] |
| Placement type | [guest post, editorial mention, listing, sponsored, other] | [score] |
| Price | [amount, currency and what it covers] | [score] |
| Link attributes | [follow, nofollow or sponsored, and who sets it] | [score] |
| Exclusions | [what they will not do, in their words] | [score] |
| Replacement terms | [what happens if the link is removed or the page dies] | [score] |
Scoring
- 5: evidence supplied, in writing, specific to your market.
- 3: claimed and plausible, no evidence yet.
- 0: refused, dodged, or answered with a screenshot of a metric instead of an answer.
Total: [score] / 45 · Scored by: [name] · Date: [date]
Compare vendors on the total only when every row is filled. A high score on a half-filled card just means they answered the easy questions.
Why it works
Judge every link vendor on the same 9 rows before price enters the room, so you buy placements that survive and skip the ones that vanish.
- Score market, relevance, traffic evidence and editorial control first, so a cheap pitch cannot hide behind an authority metric.
- Approve each placement in writing, naming the page, anchor and attribute, so nothing gets published you did not agree to.
- Set replacement terms in days before you pay, so a removed link is a remedy rather than an argument.
5 ready-to-use variants
Copy allVendor Scorecard (fill-in)
When to use: Score every vendor on the same 9 rows before you look at price, and keep the evidence next to each score.
Vendor scorecard
Score every link vendor on the same rows, before you see the price. Most bad link buys happen because the pitch arrived with a price and a promise, and nothing to compare it against.
| Row | What to record | Score 0 to 5 |
|---|---|---|
| Market | [countries and languages the placements actually reach] | [score] |
| Topical relevance | [how close the site is to what you sell, in one sentence] | [score] |
| Traffic-source evidence | [what proof they can show that real people read the site] | [score] |
| Editorial control | [who decides the angle, the anchor and whether the piece runs] | [score] |
| Placement type | [guest post, editorial mention, listing, sponsored, other] | [score] |
| Price | [amount, currency and what it covers] | [score] |
| Link attributes | [follow, nofollow or sponsored, and who sets it] | [score] |
| Exclusions | [what they will not do, in their words] | [score] |
| Replacement terms | [what happens if the link is removed or the page dies] | [score] |
Scoring
- 5: evidence supplied, in writing, specific to your market.
- 3: claimed and plausible, no evidence yet.
- 0: refused, dodged, or answered with a screenshot of a metric instead of an answer.
Total: [score] / 45 · Scored by: [name] · Date: [date]
Compare vendors on the total only when every row is filled. A high score on a half-filled card just means they answered the easy questions.
Placement Approval Brief
When to use: Approve or decline each proposed placement in writing: site, anchor, attribute, cost, relevance and evidence.
Placement approval brief
One brief per proposed placement, sent before anything is written or paid for.
- Proposed site: [domain] · Country and language: [market]
- Page the link points to: [your URL]
- Anchor text proposed: [exact words]
- Link attribute: [follow / nofollow / sponsored]
- Placement type: [guest post / editorial mention / listing / other]
- Who writes it: [vendor / us / the publisher]
- Cost: [amount] · Included: [writing, images, revisions, anything else]
- Why this site is relevant: [one sentence a customer would agree with]
- Evidence of real readers: [link to the proof supplied]
- How long the link stays live: [period, in writing]
Buyer response
- Reviewed by: [name] · Date: [date]
- Response: [approved for drafting / changes requested / declined]
- Notes: [what must change before this proceeds]
Important: approval on this brief is editorial only. It confirms the placement fits, and it is not a purchase order, an instruction to pay, or a commitment to any invoice. Purchases go through your normal procurement route, signed by whoever is allowed to sign.
Relevance & Traffic Evidence
When to use: The questions that separate a site with readers from a site with metrics, and the answers to keep on file.
Relevance and traffic evidence
Ask for these in writing. A vendor who cannot produce them for a site they are selling you does not know that site either.
Relevance
- What does this site publish, in one sentence? [answer]
- Name 3 recent articles on it that our buyer would read: [3 URLs]
- Would the link make sense to a reader who never heard of us? [yes / no, and why]
Where its readers come from
- Which countries? [answer] · Does that match our market? [yes / no]
- Is traffic mostly search, social, newsletter or referral? [answer]
- Can you show the analytics or a screenshot with dates visible? [yes / no, evidence link]
- How many other paid links are on the same page? [number]
Signals that should lower the score
- Every page carries outbound links to unrelated industries, in several languages.
- The site's own traffic comes from one branded keyword and nothing else.
- A price list covering hundreds of sites, with no editorial contact for any of them.
- Authority metrics quoted with no traffic evidence at all. Metrics can be bought; readers cannot.
Check the site yourself before you decide. Our free Backlink Analyzer shows what a domain's own link profile looks like, which is usually enough to spot a network.
Terms, Exclusions & Replacement
When to use: Agree link attributes, exclusions and what happens when a link is removed, before the order is placed.
Terms, exclusions and replacement
Write these into the order. The time to agree what happens when a link disappears is before you pay for it.
Link terms
- Attribute agreed: [follow / nofollow / sponsored] · who controls it: [name]
- Anchor text: [exact words], changed only with our written approval
- Minimum live period: [period]
- Disclosure: [how the placement is labelled on the page]
Exclusions we require
- No placements on [sites, sectors or networks you will not appear on]
- No links from pages that exist only to host links: [yes / no]
- No changes to our landing page content: [yes / no]
- No resale of our placement to another client: [yes / no]
If it goes wrong
- Link removed inside the agreed period: [replaced within X days, at no cost, on a site of equal or better relevance]
- Page deleted or 404: [same remedy]
- Attribute changed without approval: [remedy]
- Site later found in a link network: [remedy, including refund terms]
- Who checks, and how often: [name, cadence]
Re-check live links on a schedule. A link you never check is a link you are paying for on trust alone.
Worked Example (two vendors)
When to use: Two invented pitches scored side by side, where the cheaper one loses on the rows that decide whether the link survives.
Worked example: two vendors, one placement
Example only. The vendors, sites and numbers below are invented to show how the scorecard separates two pitches that look similar in an email.
| Row | Vendor A | Vendor B |
|---|---|---|
| Market | UK and Ireland, matches ours (5) | Global mix, mostly outside our market (2) |
| Topical relevance | Logistics trade title, our buyers read it (5) | General business blog, any topic (2) |
| Traffic-source evidence | Analytics screenshot with dates, mostly search and newsletter (5) | Authority metric only, no traffic shown (1) |
| Editorial control | Publisher's editor approves the angle (4) | Vendor writes and publishes, no editor named (2) |
| Placement type | Guest article, labelled contributor (4) | Link inserted into an existing post (2) |
| Price | Quoted per placement, writing included (4) | Cheaper, writing extra, revisions extra (3) |
| Link attributes | Follow, set by the publisher, in writing (5) | Follow promised verbally, no written term (1) |
| Exclusions | Will not place on sites they do not read themselves (4) | No exclusions offered (1) |
| Replacement terms | Replaced within 14 days if removed in 12 months (5) | None (0) |
Totals: Vendor A 41 of 45, Vendor B 14 of 45. Vendor B was 40% cheaper, and the gap is entirely in the rows that decide whether the link survives and whether anyone reads it.
Decision recorded: approve Vendor A for a 3-placement trial, re-score after the third placement goes live. Approval is editorial only, and the purchase order goes through procurement.
Next step
Score your own shortlist, then bring it to us: we will tell you which rows we would fail, and quote against your criteria rather than ours. Book a free consultation, or see how we run the work on the link building services page.
How to use this template
- Score every vendor on the same 9 rows, and fill the evidence column before you look at the price.
- Ask for relevance and traffic evidence in writing, and treat authority metrics without traffic as unanswered.
- Send a placement approval brief for each proposed link, naming the page, the anchor and the attribute.
- Agree exclusions and replacement terms in the order, not after a link disappears.
- Re-check live links on a schedule, and re-score the vendor after the first 3 placements.
Pro tips
- Score before price. Once a number is on the table, every other row gets argued backwards from it.
- A vendor who cannot name 3 recent articles on a site they are selling you has not read that site either.
- Keep the approval brief editorial. Say plainly that approval is not a purchase order, so nobody treats an email as a signature.
- Write the replacement remedy in days and in writing. "We'll sort it out" is not a remedy.
Frequently asked questions
Does a high authority score mean a link is worth buying?
No. Authority scores are third-party estimates and can be inflated deliberately. Score the site on readers, relevance and editorial control first, and treat any metric quoted without traffic evidence as an unanswered question.
What should the placement approval brief not do?
It should not authorize a purchase. Keep it editorial: it confirms the site, page, anchor, attribute and cost are acceptable. The purchase goes through your normal procurement route, signed by whoever is allowed to sign.
Which link attribute should we ask for?
Ask what the publisher's policy actually is and get it in writing. A sponsored or nofollow link from a site your buyers read can be worth more than a follow link from a page built to sell links. What matters is that the attribute is agreed and does not change without your approval.
How do we handle a link that gets removed?
Agree the remedy before you pay: a replacement on a site of equal or better relevance, within a stated number of days, for any link removed inside the agreed live period. Then check your live links on a schedule, because nobody will tell you when one goes.
How many vendors should we score?
Three is usually enough to see the spread. Score them on identical rows, and only compare totals once every row is filled, since a high score on a half-filled card just means they answered the easy questions.