Link Building in 2025: What I’ve Learned After Doing SEO for 8 Years

I spent three years and roughly $47,000 learning how to build links. Most of it was a waste. I wrote guest posts nobody read. I paid for “niche edits” that did nothing. I subscribed to three different SaaS tools and used maybe 20% of their features. This is the part that actually ended up working — and more importantly, what I would not do again if I had a redo button.

The Stuff I Got Wrong

Let me get the failures out of the way first, because honestly I see too many articles that only talk about the wins. It gives people unrealistic expectations.

Guest posting. $4,000 on 40 articles for sites I found by searching “write for us” plus whatever niche I was in at the time. Two of those links did anything measurable. The other 38 sites had Domain Ratings under 22. Google treats those the same as a comment section link. Basically invisible.

Directory submissions. I hit every one — Yext, Hotfrog, all those. The traffic from all of them combined: zero users. If someone tells you directories work in 2025, ask them for a screenshot of their analytics showing it. They won’t have one.

Buying links from Fiverr. Yeah I did this too. $500 for 50 links. Every single one was from a site that looked fine until you checked the traffic — zero monthly visitors. Google figured it out in about two weeks and the links stopped counting.

What Actually Worked (Three Things)

1. Fixing outdated content (34 links from one campaign)

There is this thing that happens with older content on the web — people write a guide, it ranks, it gets links, and then the data goes stale. Nobody updates it because the original author moved on. If you can find those pages and replace the old data with current numbers, the people who linked to the original are often happy to link to your version instead.

I found a guide on “email marketing ROI” from 2022. It cited a 2018 study saying $42 return per $1 spent. The real 2024 number from the same research firm (DMA) is $36. Not a huge difference, but enough that anyone citing that 2018 stat looks sloppy.

I wrote a new version with all 2024-2025 data. 12 sources, comparison table, methodology section. Then I checked who linked to the old version — about 35 sites. I emailed every one.

Email was something like: “Hey, saw you linked to that old email ROI guide. Just a heads up — the data in there is from 2018. I put together an updated version at [URL] with 2024 numbers. Might be worth swapping out. No pressure either way.”

34 out of 35 replied. 22 swapped the link. 12 kept both. That’s 34 new links from one afternoon of work.

2. HARO — but you need a system

HARO is free and the links come from real news sites. The problem is everyone knows about it now, so journalists get flooded. I set up Gmail filters for 7 keywords: SEO, digital marketing, content strategy, Google, search, conversion, analytics. When a matching query came in, I responded within 15 minutes. Every time.

My response formula: one specific data point or story, under 150 words. No fluff. No “as an SEO expert.” Just the useful part. Attached a link to supporting data if needed.

Over six months: 127 responses sent. 22 journalists replied asking for more. 12 published links. The outlets were Entrepreneur, Inc., HubSpot, Search Engine Journal. Traffic from those links? Roughly 400 visits a month. Not earth-shattering but the SEO value from those domain authorities is significant.

3. Broken links using stuff I already wrote

This is the one that people seem most surprised by because it requires zero new content. I keep a Google Sheet of about 50 pieces I’ve already published that are good enough to earn links. Each entry has the URL, a two-sentence pitch, and a key stat.

Every week I run broken link checks on 10-15 resource pages in my niche. When I find a 404, I check if I have anything in my sheet that covers the same topic. If yes — a 60-second email.

“Hey, was reading your resource page and noticed [URL] is broken. I have something similar at [my URL]. Figured I’d flag the broken link regardless. Cheers.”

In Q4 last year: 67 broken links found. 43 matched my sheet. 22 turned into links. Total weekly time: about 3 hours.

The Tool Thing

I use Ahrefs ($129/mo) for finding broken links and checking who links to competitors. Hunter.io ($34/mo) for finding emails. Streak ($15/mo) for tracking outreach in Gmail. Google Sheets (free). HARO (free).

That’s $178/month. If you’re starting out, skip Ahrefs and use the 7-day free trial once a quarter. Use Hunter’s free tier (25 verifications). Your first few months can legitimately cost zero.

Realistic Expectations

If you spend 5 hours a week on link building, here is roughly what happens:

• Month 1: 40 outreach emails, maybe 2-4 links. Feels pointless.
• Month 3: 60 emails, 5-8 links. Starting to feel real.
• Month 6: 75 emails, 10-15 links. Things start compounding.
• Month 12: 100 emails, 60-90 total links. You are now competitive.

I kept a spreadsheet. Month 2 was the hardest — I had 5 links and $178 in tool costs and seriously considered stopping. I didn’t because I had already written about it publicly and felt stupid quitting. Sometimes public accountability is the only thing that keeps you going.

One Last Thing

Link building is boring. It is not strategic or creative. It is sending emails, tracking responses, updating spreadsheets. The people who win are not the ones with the best understanding of Google’s algorithm. They are the ones who do not stop after month two when results are invisible.

I almost quit three times. The third time I had a client whose traffic was growing and I could not afford to let them down. That was the turning point.

Start with a link bank. Ten pieces of content you have already published. Write a two-sentence pitch for each. Find three resource pages with broken links this week. That is one hour. Do it again next week. And the week after.

Related Articles

Why Your SEO Strategy Isn’t Working (And What to Do Instead)

How to Drive Real Traffic to Your WordPress Site (Without Burning Cash)