How to budget a monthly link campaign, and what link velocity really means

A typical brief says "we have £2,000 a month for links". Then someone works out how many links that buys, and the campaign becomes a race to fill the quota. It's backwards. The budget should come out of the plan, and the plan should come out of the gap.

Start from the gap, not the number

Pick the pages you want to move. For each one, look at the pages ranking in the top five for its main terms and count their referring domains. Then cut that count down to the ones that are relevant, editorial and getting traffic. Raw referring domain counts include a lot of scraper sites and directory junk, and you're not trying to match those.

Here is an illustrative example, not a benchmark. A comparison page has 12 relevant referring domains. The pages above it have between 35 and 55. Call the gap 30. Spread over nine months, that's three or four placements a month for that page. Do the same for your other priority pages and add them up.

Two adjustments. Your own content may be the real problem; links won't drag a thin page past a better one, so fix the page first. And if AI visibility is part of the goal, add a second list: the publications that engines cite for your prompts. Those are worth budgeting for separately, whatever their DR. We explain how to build that list in brand mentions in AI answers.

What links cost in 2026

Public price data is thin and skewed to the US. The most useful recent set we know of is BuzzStream's analysis of its vendor database of more than 500,000 guest post sites (BuzzStream, link building pricing). The figures are in US dollars.

Averages hide a lot. A relevant UK trade title with real readers will cost a multiple of a generic blog with the same DR, and it should. Which format to buy depends on the page and the goal more than on the average price.

Cost per link is the wrong denominator

The sticker price counts every order. It should count only the links still doing their job. We work with a true cost per link:

true cost per link = (placement fees + content + revisions + your team's time) ÷ links live, indexed and unchanged after six months

Some placements never go live. Some go live and drop out of the index. Some get edited, redirected or turned nofollow without notice. If one in five of a cheap vendor's links fails over six months, a £120 link really costs £150 before you count the time spent chasing. Run the numbers on your own history; the failure rate differs a lot between suppliers, and it's usually the cheap ones that fail. Our guide to measuring link building ROI takes this further, into rankings and revenue.

Putting a monthly number on it

Once you have the gap and a realistic cost per link, the monthly figure mostly writes itself. Say the gap analysis calls for ten placements a month across three priority pages. Price them at your true cost per link for the tiers you expect to buy (more on tiers below), add content if publishers don't write it, and you have the core budget.

Then split it. We'd usually put the large majority into placements that serve the ranking plan, a separate slice into publications cited by AI engines in your niche, and hold back 10 to 20% as a reserve. The reserve matters more than it sounds. Good opportunities don't arrive on a schedule: a strong trade title opens a slot, a journalist wants a comment, a competitor's citation source starts taking contributed pieces. Without a reserve you either miss them or blow next month's plan.

What if the number you get is far above what the business will spend? Then cut scope, not quality. Drop to one or two priority pages and close those gaps properly. Ten cheap links spread across eight pages will move none of them; four good links on one page might move it. The same logic applies to agencies quoting clients. A small budget is fine if the page list shrinks to match it.

Link velocity: myths and what's real

"Don't build links too fast or Google will penalise you" is one of the most repeated rules in SEO. There is very little behind it.

Where the idea came from

Google's 2003 patent application "Information retrieval based on historical data" says a "spiky rate of growth in the number of back links" may be a factor in scoring documents, and that it can signal an attempt to spam. That patent is the root of most velocity talk. A patent shows what Google's engineers considered in 2003. It doesn't show what runs today.

What Google has said since

Search Engine Journal's review of the claim found no evidence that Google uses a signal called link velocity. It quotes John Mueller saying "it's not so much a matter of how many links you get in which time period", and reports Gary Illyes calling link velocity a "made-up term". In April 2024 Illyes went further at a conference in Bulgaria: "We need very few links to rank pages… Over the years we've made links less important." He later said he shouldn't have said it, which is probably the most honest thing anyone at Google has said about links.

What's real

Speed itself isn't the problem. What a burst usually contains is. Twenty paid posts in a fortnight, on unrelated sites, with exact-match anchors, looks like exactly what it is, and Google's link systems are built to spot link schemes whatever pace they arrive at. A news story that earns 40 links in a week looks like news.

Practitioners still prefer steady growth. In Whitespark's 2026 Local Search Ranking Factors survey of 47 contributors, "sustained influx of links over time (rather than bursts)" ranked 34th for local organic results. Quality and authority of inbound links ranked 3rd (Whitespark). That is expert opinion rather than Google data. The order still looks right to us: what you get matters far more than how fast.

Pacing a monthly campaign

The real limit on pace is your ability to vet publishers, write decent briefs and review content. A campaign that buys 30 links a month with a team that can only review ten properly will end up with 20 links nobody checked.

A few rules we'd hold to:

• Keep a steady monthly spend rather than spending a quarter's budget in one week.
• Spread placements across publishers. One link per publisher per quarter is a good default, unless a site is unusually strong and relevant.
• Keep exact-match anchors rare. The anchor text guide has the ratios we use.
• Point most links at pages that deserve them: the hub and its best supporting pages, not whichever page is closest to converting.
• Review results monthly, but judge them over a quarter at least.

Max cost per link rules

A ceiling stops a campaign paying a premium for DR alone. Tier it by what you actually value, then set a cap per tier. The caps below are placeholders. Set yours from your own cost history.

See topical relevance vs Domain Rating for how we judge the criteria.

Auto vs semi-auto

Once the rules are written, you can let a system place within them or review each recommendation yourself. We'd start almost everyone on review.

Agencies running many clients tend to mix the two: auto for mature accounts with tight rules, semi-auto for new and regulated ones. The agencies page covers that set-up.

Doing this in Rankticker

Campaigns take the plan above as settings: a monthly budget, an objective (improve backlink profile or improve AI visibility), target URLs, anchors, keywords, geo, niche, a max cost per link and a brief. Choose Auto or Semi-auto per campaign. In semi-auto, each recommendation shows why the publisher was picked, such as "DR 69 · strong fintech topical overlap · low spam score", and you approve or decline it.

The project overview tracks spend, live links and pending orders, and the Live Link Tracker on each order keeps checking that the link is live, indexed and unchanged. That gives you the denominator for true cost per link without a spreadsheet. See pricing for how fees work.