I've been writing about SEO since I was 15 years old, and a decade later I'm fortunate to have consulted for companies I love like Ahrefs, Kinsta, Buffer, ConvertKit and multi-billion dollar brands.
The basics can be incredibly effective, but hundreds of sites cover them well and I want to focus on unique, creative ways to achieve better rankings.
Instead, here's my promise: I will put my absolute all into guides like this one to give original insights that help you get an edge over your competition.
That's it. That's my pitch for you to stick around (or perhaps let you know this isn't the site for you).
This is already too much text for a 'click-to-read-more-fade-thing' but there's more if you like.
Thank you for being here!
In March of 2026, Ahrefs opened up API access to all paid plans, Lite and above.
As an Ahrefs customer for 10+ years, API access has always been something I wanted but never thought would be available to anyone but Enterprise customers.
The reaction to the announcement was incredible, but there was also a common request, “Please can you increase the API limits.”
A month later, Ahrefs did exactly that at no additional cost.
Here’s the graphic I made to highlight the increases:
This also means I can now share practical API use cases that anyone can follow, and there are a lot of them.
My Ahrefs API-powered dashboard highlights content & PR opportunities on autopilot
Ahrefs has always been well-known in the search engine optimization space, with tools for keyword research, competitor analysis, backlink research, site audits, and more.
As so many conversations and purchasing decisions are happening across ChatGPT, AI Mode, Claude and beyond, Ahrefs evolved to give brands visibility into how they’re performing across those platforms.
(We track 400M+ AI prompts monthly, which I believe is the biggest index in the entire space.)
Later, with the launch of Agent A, we also moved into helping people use AI to automate marketing workflows and processes.
As one way to keep track of what’s happening across the industry, I created a dashboard (powered by the Ahrefs API) to, among other things, monitor new mentions companies are receiving.
It’s not the only reason I monitor so many brands, and the aim is never to copy any content ideas — more on that in a future report — but it’s fascinating to stay on top of what’s being reported on in our space.
The API Costs to Replicate My Approach
While the idea of checking links to 180+ sites probably sounds expensive, I might be about to surprise you.
Checking recent backlinks to 180 sites required 67,538 API units, which is less than the 100,000 units even the cheapest Ahrefs plan offers each month.
(Higher tiers offer 400K, 1M, and 2M units, respectively.)
Keep in mind that I only run this check quarterly — there’s enough to act on and keep me busy — and you don’t need to check as many sites as I do.
Especially not to start with.
To be fully transparent, there were a number of sites which are too new to have 50 rows of data.
If every site had 50 rows of link data, I would have been able to check 95 for the same unit costs (or 142 if I used all Lite plan availability).
Using Claude Opus 4.8 to analyze hundreds of newly-acquired links and provide an actionable summary costs ~$0.19. That’s the most expensive model Anthropic has available, and is honestly far more capable than is required.
#1
Document notable competitors in your niche (and slightly beyond)
Depending on the size of your niche, there may be just a dozen brands worth monitoring, or hundreds.
It’s totally up to you how many brands you want to monitor for new links, but I’ll add the perhaps obvious note that the more sites you track, the more trends and opportunities you’re likely to see.
If I’m trying to promote a new sunglasses brand, I might look at big brands in the US, as well as local success stories in Europe or Africa.
If you’re doing marketing for a local plumber, I would want to look at links to top plumbers around the country, or even the continent.
I might even go a step further and monitor new links acquired by “big-box” home improvement stores and DIY blogs.
If one idea is to find content angles you could improve upon with your own unique data and insights, you don’t have to narrow your focus too tightly.
For Ahrefs, it’s easy to expand beyond SEO into AI visibility and marketing as a whole.
#2
Use these filters to find the best* new links they acquire
Depending on your industry, you may want to tweak the filters I’m about to share, but they should be a great fit for most scenarios.
Below you can see an example Ahrefs backlink report for ahrefs.com, with these filters automatically applied (you’ll need to be logged in for that link to work).
To explain each of them:
- Backlink type: In content — Links that are found in content, rather than anywhere (e.g. navigation links)
- DR: From 50 — Sites that are generally authorities in their space sending links less likely to be spammy
- Ref. page language: English — The language of the page the link is on (tweak where necessary)
- New — Links that were first discovered in our specified timeframe
- Last 3 months — How far back to check for newly acquired backlinks
- Status: Newly published — The link was created with the page being published, not added later
- Live links only — Excludes links which were initially in place but then later removed
- One link per domain — only shows a single link from each site linking out
Feel free to add more filters for your own use case, but note that they will incur additional API unit costs.
For example, you may want to check if links are followed, rather than nofollow. I’ve found that large publications often link to brands via nofollow links, so I still want to be aware of them.
#3
Get the API URL directly from Ahrefs, then (optionally) optimize it
One of the coolest features of Ahrefs is that, where applicable, the page you’re on provides an API URL with your chosen filters already applied.
You can also see the expected unit cost per row of data.
If you’re a developer, it should be very easy to see from the official API docs how to optimize this to use fewer credits.
Removing traffic data, which I don’t require, reduced my units per row from 66 to just 13.
If you’re not a developer, you can ask AI to tweak the API URL for you. Claude did that for me first time, without issue.
I also set the maximum rows per request to 50, which is enough to get insights into what’s happening for each brand.
Note that an Ahrefs API call requires a minimum total cost of 50 units, so if you check a site and it has no relevant links for your filters, you’ll use those 50.
For context, you could check 2,000 sites without data before you hit the limits of the cheapest paid plan.
#4
Develop or vibe-code a simple platform to store and sort the data
If you’re already familiar with working with API keys and building apps, you can skip to section #5 without missing anything.
If you’re new to building apps of any form, I can’t get into the topic in too much detail (especially around security), but I’m happy to offer some recommendations.
For a few years now, I’ve been building most of my projects using Cursor and Google Firebase. Cursor handles the code via simple prompts, and Firebase keeps everything online for me with backups and authentication.
These days, I’m building most projects in Letaido, an Ahrefs-owned platform where building and hosting projects happens in one place.
Other popular options out there include Lovable, Base44, and building projects with Claude Code.
If this is your first time building any sort of application, I think you might be amazed by what can be created with simple prompts in natural language.
A request like,
I’m looking to use the Ahrefs API to extract recent backlinks from a list of sites. Here’s the API URL Ahrefs provided with the specific filters I want in place. I’ll need to be able to save and sort the data, and rerun the analysis on my own schedule. Please help me create a simple interface to manage the sites and results from an individual run.
Even if security and user access are built into the platform you’re using, be very cautious about making something public immediately, as you may accidentally share API keys or other sensitive data.
There are plenty of great tutorials on these topics, or professionals who would be happy to guide you through the process.
#5
Scan the results manually, then use AI for additional insights
One of the first tables I asked my AI-assistant to create for me was a list of all domains, sorted by how many of the brands I was tracking they linked to in the past 90 days.
Logically, the more brands a site links to, the greater the chance you have of them linking to you as well.
I also asked for the ability to mark as favourite the sites I value mentions on highly, like Fast Company, Fortune, and The New York Times.
Then I had a separate report that looked only at the links from those sites, which I could then analyze further on my own or with the help of AI.
From there, I then ask AI to tell me things like:
- What are common themes in the types of pages that have picked up links
- When journalists link out, what’s the context of the mention for a brand (a study, an individual review, etc.)
- Any interesting or suspicious patterns in anchor text used in links
- Articles or studies missing from the analysis that might make sense to cover
- Any other notable insights from the data I didn’t specifically ask for
Of course, I pair this all with a manual check as well (I love going through the data) but AI is great at this kind of work.
Frequently Asked Questions
This article only just went live, so I’m assuming these are some questions I might get.
If there’s something you want to ask, please leave a comment on LinkedIn or X.
“I want to take this idea to the next level. What else can I do?”
If there’s enough interest, I’ll share more competitor research angles you can use to help with your marketing efforts.
For this idea specifically, you could:
- Pull more data about each link (e.g. estimated traffic)
- Pull more data about the page receiving the link (e.g. its title)
- Check more rows for each brand you’re monitoring
- Increase the frequency of checking to potentially jump on opportunities quicker
- Set this up for adjacent niches to find ideas that might apply to your own
“I’m seeing some links which are a bit older than my suggested timeframe”
Unfortunately no tool is perfect here, but the vast majority of links should be from the timeframe you chose.
The discrepancy can come from how long it takes to crawl a page after it was published, whether links were added after publication, and so on.
