Backlink Report Checklist: What Every Delivery Report Needs

A trustworthy backlink report lists every link as its own row: live URL, target URL, exact anchor text, link type, rel attribute, date placed, indexing status and the source of any metrics. You can check one yourself in about 15 minutes by sampling 10 to 20 links with nothing but a browser.

Use one rule when you read it: a stranger with a browser should be able to confirm each row without asking the vendor anything. If a report relies on screenshots, has no anchor data, or lists pages Google can’t index, treat it as unverified. This checklist gives you the field spec, the timed routine and the red flags.

The fields a backlink report must contain, row by row

A delivery report is proof of placement. It shows what was built, where, and when, so you can confirm it independently. It doesn’t prove the links helped your rankings. Each field below closes a specific gap in what you can check.

The nine fields in one table

Sample report: three illustrative rows

These rows use placeholder domains and dates. They aren’t real data.

Four terms you need before you check anything

• Rel attribute: an HTML attribute on a link that tells search engines how to treat it. The common values are nofollow, sponsored and ugc (user-generated content). Google explains them in its guidance on qualifying outbound links.
• Noindex: a directive that tells Google not to keep a page in its index. It can sit in a robots meta tag in the page’s head or in an X-Robots-Tag HTTP header. See Google’s robots meta tag documentation.
• Indexed vs. discovered: “discovered” means Google knows the URL exists. “Indexed” means it has stored the page and can show it in results.
• Contextual link: a link placed inside body content, as opposed to a profile field, footer, sidebar or comment.

Anchor text is the clickable words of the link. Target URL is the page on your site the link points to.

How to spot-check a backlink report in 15 minutes

1. Minutes 0-2: choose 10 to 20 rows.
2. Minutes 2-9: open each page, find the link, read the source.
3. Minutes 9-12: check indexing.
4. Minutes 12-15: score the sample and decide.

Minutes 0-2: pick a sample that the vendor didn’t pick for you

Don’t check the first rows. They’re the ones a vendor is most likely to have polished. Take a few from the start, a few from the middle and a few from the end, and make sure every link type in the report appears at least once.

Minutes 2-9: open the page, find the link, read the source

Open each live URL. Confirm it loads normally (no 404, no redirect to somewhere else), find your link, and compare the anchor text and destination with the report.

Then press Ctrl+U (or right-click, View page source) and search for your domain. Read the rel value on the link. In the head, look for <meta name="robots" content="noindex"> and for a canonical tag pointing to a different URL. A canonical is a hint, not a command, but a page that canonicalizes elsewhere may not be indexed under its own address (Google’s canonicalization documentation).

Source view has limits. It can miss links injected by JavaScript, so check the rendered DOM with Inspect Element. It also can’t show X-Robots-Tag headers. Find those in the Network tab of your browser’s dev tools, or record them as an unchecked limit.

Minutes 9-12: check indexing, and know what the check can’t tell you

URL Inspection in Search Console only works for properties you’ve verified, so it’s no use on third-party pages. Search site: followed by the exact URL, or search a distinctive sentence from the page in quotes.

Both are imperfect. site: isn’t an exact index count, and a missing result isn’t proof the page is excluded, especially if it’s new. Mark it “not found, re-check” and try again in a week or two.

Minutes 12-15: score the sample and apply a decision rule

Copy the report rows into a sheet and mark each pass or fail on live, anchor, target, rel and indexability. Then count.

Our own rule of thumb, not an industry standard: if more than a small share of the sample fails on live, anchor or target, ask for fixes before you pay a balance or accept the order. If failures turn up, widen the sample. A 10-20 link check can’t prove the whole report is accurate. It only tells you how likely the rest is to be sound.

Worked example: a live link on a page Google is told not to index

Say row 1 points to a Web 2.0 post at example-blog.wordpress.com. You open it and your link is there. Press Ctrl+U and search for yourshop.com. You find:

<a href="https://yourshop.com/trail-shoes" rel="nofollow">lightweight trail running shoes</a>

The anchor and target match. Now scroll up to the head:

<meta name="robots" content="noindex, follow">

The report row reads “live, anchor matches” and still fails. The host page is excluded from Google’s index, and the report didn’t flag it. In the Notes column, write something like: “Noindex meta tag found 2026-03-20; page not eligible for indexing.”

Ask the vendor to replace the link with one on an indexable page, and to add a noindex check to every row of the revised report.

Red flags in a backlink report, and what you can verify in each format

The red-flag list

• Screenshots only. They’re easy to crop or fake, and you can’t open them.
• A link count with no URLs, or a single total for a drip-fed campaign.
• No anchor data, or the same anchor on every row.
• Links on pages Google can’t index: noindex, blocked by robots.txt, canonicalized elsewhere, or orphan pages nothing links to.
• Most links concentrated on a handful of domains.
• Metrics with no tool name or date.
• No rel column while the vendor sells “dofollow” links.
• Links pointing to a redirect or the wrong target URL.
• Indexing described as “submitted” or “guaranteed” rather than checked.

100 screenshots and “250 links” vs. a 250-row spreadsheet

Take a PDF with 100 screenshots and a total of “250 links”. You can look at 100 images, none of which you can reopen, and you can say nothing about the other 150.

A 250-row spreadsheet is different. Every row has a URL you can open, sort and sample. A spreadsheet can still contain false rows, which is why you sample across it rather than trusting it. But it gives you something to test, and a screenshot PDF doesn’t.

Follow, nofollow, sponsored: what the rel column should tell you

The rel column records how each link is qualified. Google’s guidance on qualifying outbound links says to use rel="sponsored" for paid or ad links, ugc for user-generated content such as comments and profiles, and nofollow when you don’t want to associate with the page. Since Google’s 2019 announcement, it treats all three as hints, not strict commands, for ranking purposes.

Don’t read a column of plain “follow” links as good news. Google’s spam policies cover links meant to manipulate rankings, and purchased placements with no qualifier are the clearest example. Unqualified links may carry more risk. Qualified links may pass less ranking credit, or none. Nobody outside Google knows exactly how much, and it varies by site and niche.

You can lower the risk, not remove it: pick relevant pages, keep volume moderate, vary anchors, use sponsored or nofollow where a link is paid, and favor quality over quantity. What the column must do is match the page source. A report that says “follow” while the HTML says nofollow is wrong.

Turn the report into a client-ready template with an anchor tally

Summary block to put above the raw table

Open with three small tallies: links by type, anchors by category (branded, exact-match, partial-match, generic, naked URL), and links per target page. Keep the raw per-link table underneath. Show nofollow, sponsored and ugc links there too. They belong in a natural profile, and hiding them costs you client trust.

Reading an anchor distribution tally

Here is an illustrative tally for 20 links:

Most anchors don’t repeat a money keyword, so no single exact phrase dominates. That’s the point. Many identical exact-match anchors pointing at one page is a pattern that looks engineered. There is no official ratio, so treat this as a way to spot concentration, not a target.

Spreadsheet columns you can copy

Row ID, Live URL, HTTP status, Target URL, Anchor text, Anchor category, Link type, Contextual (Y/N), Rel, Noindex/canonical check, Date placed, Date verified, Indexed (Y/N/Pending + method), Metric + tool + date, Notes.

On metrics: DR (Ahrefs) and DA (Moz) are proprietary third-party estimates, not Google signals, and Google has said it doesn’t use Domain Authority. Tools disagree and the numbers can be gamed, so never treat them as proof of quality.

What a clean report still can’t tell you

A clean report proves placement. It doesn’t prove ranking impact, link value or safety, and results vary by site and niche. Buying links to influence rankings can also violate Google’s spam policies, however tidy the spreadsheet looks.

It is also a snapshot. Indexing can lag behind placement, and links can be removed, edited or de-indexed after delivery. Date-stamp every report, ask the vendor for a re-check at a set interval (30 days is a reasonable choice), and re-sample rows yourself rather than relying on their second report.

Metrics in the report are estimates from one tool on one day. Use them to compare rows, not to judge whether a link was worth buying.

What to do with your findings, and your next step

Send the failed rows to the vendor with your notes and ask for replacements, not refunds first. Keep your marked-up copy as a record. If your terms allow it, hold the balance until the failures are fixed. Then set a 30-day reminder to re-sample the same rows plus a few new ones.

If you’d rather start from a campaign that mixes link types, BacklinkUpSEO’s Premium Backlink Packages (Starter, Growth, Dominance) combine several into one order. Every order comes with a report listing each live link, so you can run this same check on it. Campaigns usually start within 48-72 hours and are delivered gradually over the stated period. Apply the checklist to us as hard as to anyone else, and don’t expect designed to supportings, indexing or safety from any vendor.

Your next step: open your latest report, pick 10 rows from the middle and end, and start the timer.

Frequently asked questions

What should a backlink report include?

One row per link with the live URL, target URL, exact anchor text, link type, rel attribute, date placed, indexing status, and the tool and date behind any metric. A total count or a screenshot gallery alone isn’t enough to verify.

How can I check a backlink report without paid tools?

Sample 10 to 20 rows across the whole sheet. Open each page, view the source with Ctrl+U, find your domain, and read the rel attribute. Also check for a noindex robots tag and run a site: search to see whether the page is indexed.

Is a report with only screenshots acceptable?

Treat it as unverified. Screenshots are easy to crop or edit and you can’t reopen them. Ask for a spreadsheet with a clickable URL on every row.

Should all the links in my report be dofollow?

No. A mix of followed, nofollow, ugc and sponsored links is normal, and paid links are meant to be qualified under Google’s guidance. A report that is entirely plain follow links deserves a closer look, not applause.

What if a link isn’t indexed yet?

Mark it Pending and re-check later. New pages can take time to be indexed, and a site: search is an imperfect test. If the page is noindex or blocked, though, waiting won’t fix it. Ask for a replacement.

Does a clean report mean the links will improve my rankings?

No. It shows the links were placed as described. Whether they help depends on your site, your niche and how Google treats those links, and buying links to influence rankings can breach its spam policies.