You spent money on links last quarter. Did it work?
Most answers to that question stop at rankings. You moved from position 14 to position 6, so the campaign worked. That feels like proof. It usually isn’t.
Return on investment is a business number. It compares what you spent against what came back. Rankings are a step along the way, not the money. Plenty of pages rank well and sell nothing, and you can run projections using the TAM, SAM, SOM method and still miss it.
There is also a bigger problem hiding in the usual math. Almost nobody accounts for the fact that links die.
This guide walks through how to measure link building ROI properly in 2026. We will use Google Analytics 4, Search Console, and our own data on 865,000 monitored backlinks.
Why most link building ROI math is wrong
Here is the standard calculation. You bought ten links at $500 each. That is $5,000. The page they point at now earns $9,000 a year. So your return is 80%.
The flaw is in the first number. You treated $5,000 as a one time cost.
We monitor 865,088 backlinks for our customers. When we looked at how long they survive, just over half were gone within twelve months. The median backlink lives about a year. The average lives roughly twenty months before it disappears. You can read the full breakdown in our link building statistics study.
So those ten links are not ten permanent assets. In a year, about five of them will be gone.
That changes what you are actually buying. Editorial.link surveyed 518 link building professionals in 2026 and found the average acceptable price for one quality backlink is $508.95. At a median lifespan of twelve months, you are not paying $509 for a link. You are paying $509 a year to rent one.
Call it cost per link year. It is the same idea a landlord uses, and it is the honest unit for this kind of spend.
Redo the earlier sum with that in mind. Year one costs $5,000. Year two costs another $2,500 just to stand still. Your 80% return starts looking thinner by month fourteen.
Cheap links are worse value than the price suggests
Survival is not evenly spread. It tracks authority quite closely.
In our data, links from domains with a Moz DA between 0 and 9 are still live 33.3% of the time. Links from domains in the DA 40 to 49 band are live 70.3% of the time. That is more than double the survival rate.
Now compare two purchases. A $100 link on a weak site and a $500 link on a mid authority one. On the invoice they are five times apart.
Adjust for survival and the gap narrows sharply. The cheap link dies about twice as fast, so you buy it again roughly twice as often. Over three years the real spread is closer to two or three times, not five. Add the effort of replacing it and the cheap option can end up costing more.
This is the single most useful thing you can do to your ROI model. Stop comparing sticker prices. Compare what a link costs you per year it stays up.
Does a better ranking prove your links paid off?
Rankings are the first thing everyone checks. They are easy to read and they move fast, so they feel like progress.
Marketers have quoted first page dominance for well over a decade. You will still find it in old roundups like HubSpot’s collection of marketing stats. The general shape holds up. Very few searchers go past page one, and the top few results take most of the clicks.
What has changed is how much of that page is a plain blue link at all. AI overviews, shopping panels and video carousels all sit above you now. The way a result is presented affects clicks as much as its position does.
So position six today is not worth what position six was worth in 2018. Treat rank as a signal that something moved. Do not treat it as revenue.
How Is Ranking Misleading?
There are sites that, even though they rank on some of the best positions in SERPs, the level of conversion is low or medium at best. So, put plainly, the best spots in Google don’t guarantee an increase in sales; and here’s why:
- Relevance of content – If the users who land on your site don’t find a solution to their problems (relevant information, service, or products), they will leave and try the sites beneath. So, being first doesn’t guarantee the right type of people will visit the site.
- Low-quality meta description – On the other hand, you may have the right product or service for the people looking for your keywords, but the meta description isn’t convincing enough. If a user doesn’t think they’ll find useful information on the site based on your snippet, they won’t click on it regardless of the fact that’s in top position.
So, even though a link building campaign translates into higher rankings, this may not mean more sales, increased traffic, or increased profitability. On the contrary, you have more chances to get organic backlinks and rank higher with high-quality traffic!
How do you measure link building ROI in GA4?
Google Analytics 4 is where most of this gets measured. Four steps cover what matters for link building.
Step 1. Define what counts as a conversion
GA4 calls these key events. Some menus still say conversions.
Go to Admin, then Data display, then Events. The Key events tab shows what already counts. To add one, open Recent events, find the event you care about, and switch on Mark as key event.
Pick something that maps to money. A newsletter signup is fine if you know what a subscriber is worth. If you do not know, measure something you can price.
Step 2. Find the traffic your links actually sent
Go to Reports, then Acquisition, then Traffic acquisition. Change the primary dimension to Session source or medium. Referral traffic is the part that came from your backlinks.
Add your key event as a column. Now you can see which linking domains send people who convert, and which send people who bounce straight back out.
Expect this list to be shorter than your backlink profile. Most links send very little direct traffic. That is normal and it is not the same as the link being worthless.
Step 3. Build a referral report you can reuse
Explorations is where you build your own reports. Go to Explore and start a blank free form exploration.
Drop Session source in as a row dimension. Add Sessions and Key events as metrics. If a metric is missing, click the plus next to Metrics in the Variables panel and import it. Save the exploration and you can come back to it every month.
Step 4. Tag the links you control
Some placements you own outright. Guest posts, partner pages, anything you submit yourself.
Tag those with UTM parameters before they go live. It costs nothing and it turns a vague referral line into a named campaign you can total up later.
A warning about attribution models
GA4 gives you fewer models to choose from than you might expect. Google removed first click, linear, time decay and position based attribution in 2023. What is left is data driven attribution and last click. You will find the setting under Admin, then Attribution settings.
This matters for link building specifically. Links tend to do their work early, when someone first hears of you. Last click attribution gives that credit to whatever they searched three weeks later. Use data driven attribution if you have the volume for it, and assume your links are worth somewhat more than the report says.
What GA4 cannot tell you
GA4 is good at what happens after someone arrives. It is blind to two things that decide your ROI.
The first is rankings. GA4 does not know your position for anything. Connect Search Console instead, or use a rank tracker.
The second is whether your links still exist.
This is the gap that quietly ruins ROI reporting. A removed link sends no referral traffic, so it simply vanishes from your analytics. Nothing turns red. The line just flattens, and it looks like the campaign stopped working rather than the link coming down.
Given that half of all links are gone within a year, this is not an edge case. It is the normal state of a backlink profile.
Monitoring closes it. Linkody checks your backlinks and tells you when one is removed, when the anchor changes, or when a dofollow link quietly turns nofollow. Whatever tool you use, the point is the same. You cannot calculate a return on an asset without knowing whether you still own it.
A worked example
Put it together with real numbers. Say you run a twelve month campaign.
- You place 20 links at an average of $500. Spend is $10,000.
- Agency and content time adds $4,000. Total spend is $14,000.
- GA4 shows referral and organic sessions to the target page produced 120 key events.
- Your average order value is $180, so that is $21,600 returned.
Naive return is $21,600 against $14,000. That is a 54% gain and everyone goes home happy.
Now apply survival. By month twelve, expect about half those links to be gone. To hold the position into year two you need to replace roughly ten of them. That is another $5,000 in link spend before you have grown anything.
Your true two year picture is $19,000 spent, not $14,000. The campaign still works. It just works about a third less well than the first sum implied.
That is the number to take to whoever signs off the budget. It survives contact with reality, and the naive one does not.
How long before any of this shows up?
Later than most people plan for. In the Editorial.link survey, 57.1% of professionals said they expect results within one to three months.
That expectation is optimistic for anything competitive. Six to twelve months is a fairer window for link building campaigns in a crowded niche. If you measure at week six and call it a failure, you will kill campaigns that were about to work.
So what should you actually track?
Four numbers cover most of it.
- Cost per link year. What you paid, divided by how long the link stays up.
- Live rate. The share of links from a given campaign or vendor that are still there. This is your quality check on a supplier.
- Key events from referral and organic sessions to the pages you built links to.
- Replacement cost. What it takes each year to stand still.
None of that is hard once the tracking is set up. The hard part is being honest about the second and fourth, because they make good campaigns look less impressive than the version in the slide deck.
Link building still pays. Done well, it is one of the most durable investments in search, and our own data shows the good links really do last, with high authority placements surviving at more than twice the rate of cheap ones.
You just have to price it like a lease and not like a purchase. So how many of the links you paid for last year are still live today?