How to Pursue Backlinks That Create Real Business Value
Not every backlink drives meaningful business results. The strongest links reach the right buyers, support useful content, and lead to measurable action. Insights from SEO and marketing experts reveal how to find, assess, and track links that contribute to real growth.
§ How this was made
The perspectives in this article were written by the named contributors and submitted through Connectively. Connectively's automated system selected them and used AI to write the headline, introduction, section headings, summary sections, and illustration, and to make minor edits for clarity. Articles are assembled and published automatically. Report errors to corrections@connectively.us. How we make our content
Illustration generated with AI
What the experts said
25 nodes- Prioritize Buyer-Research ContextRon Evan del RosarioRE

- Verify Page Traffic Before PlacementMd Mazidul HossainMM

- Seek New Buyer-Focused DomainsNirmal GyanwaliNG

- Favor Earned Editorial LinksChris RaulfCR

- Score Relevance Before AuthorityFaizan KhanFK

- Audit Traffic by Page TypeJack GenesinJG

- Track Buyer Behavior After ClicksWill MitchellWM

- Match Destinations to Campaign GoalsJM LittmanJL

- Trace Links to Customer LeadsMatt BowmanMB

- Direct Links Toward Revenue PagesTristan HarrisTH

- Favor Useful Contextual MentionsBryan VasquezBV

- Measure Links by Lead ValueBrandon GeorgeBG

- Read Host Pages Like EditorsTimothy ClarkeTC

- Inspect Durable, Trusted SourcesRohan KaliaRK

- Confirm Links Before PublicationRazvan LucaRL

- Vet Publisher Credibility IndividuallyMd Mizanur RahmanMM

- Choose Sources AI Already CitesDaniel GohDG

- Test Article-Level Audience RelevanceJacob PeistrupJP

- Require Indexing and Organic TrafficRic RubensteinRR

- Earn Links Buyers Would ValueSnket DesaiSD

- Reject Decorative Spam BacklinksNabin ThakurNT

- Build Links Through Client PartnershipsSilas BaunSB

- Align Placements With Campaign GoalsAnja Moskon SolnAM

- Serve Readers’ Next TasksAnatolii UlitovskyiAU

- Avoid Isolated Contributor SectionsVaishik KansaraVK

Prioritize Buyer-Research Context

Ron Evan del RosarioDemand Generation - SEO Link Building Manager / Thrive Digital Marketing Agency
Prioritizing Backlinks From Websites Our Potential Clients Actually Visit
The first question I ask when a backlink campaign results in many placement opportunities, but little business results, is; "Will this be the type of site that my targeted client base will likely use as part of their research process before purchasing a product or service?" At our agency, we have found that just because a site has great traffic and authority, it does NOT mean that the backlink is valuable.
I consider the publisher's audience, the subjects they write about, and whether the article's content is relevant to what our potential clients may be searching for. We've discovered that listicles, comparisons of services/products, and article recommendations from industry experts are very beneficial since these pieces of content appeal to people who are already researching their choices and are close to deciding which one to purchase.
I began paying more attention to the CONTEXT of each placement. This included looking for placements where our brand is mentioned alongside similar brands/competitors, or where the content helps the reader compare different options. It would likely be more valuable for our business to have a backlink in an article about the best marketing agencies for small businesses than to be mentioned on an unrelated website with high traffic. My knowledge of this helped me assess backlink quality based on how likely it is to reach the right audience, rather than solely on SEO metrics.
Verify Page Traffic Before Placement

Md Mazidul HossainSEO Manager / eExpertize Digital Agency
When a campaign delivers plenty of placements but little business impact, I stop counting links and start judging where each link lives. Placement volume is a vanity metric. What matters is whether a link sits on a page that real people and search engines already trust.
A disappointing placement that changed my approach
Early on, I secured a guest post on a fashion site with a very high DR. On paper it looked perfect. In practice, the site had almost no organic traffic, and the DR didn't improve at all after the placement. When I checked in Ahrefs, I found no referral traffic, no ranking movement and no visible value from that link. The metric looked strong, but the site itself was basically empty.
That was when I realized DR alone tells you very little. A high-DR site with no traffic is often built to sell links, and Google has little reason to pass real value from it.
The rule I follow now
If I have to build a link, I prefer niche edits or link insertions on existing pages, so I can verify the value before placing anything. Before approving a page, I check:
The page gets real organic traffic, at least 3 visits. The page ranks for at least 3 indexed keywords. The topic is relevant to the page I'm linking to. The page is already indexed and has a history of performing.
If a page fails these checks, I skip it, no matter how impressive the domain rating looks.
The result
After switching to this process, the difference was clear. Within 9 weeks, I saw a 33 increase in organic traffic coming through these links, and rankings improved by 49%. I built fewer links, but each one carried real traffic and relevance.
Lesson learned
I now treat every backlink like an investment. DR is only a starting signal. Traffic, rankings, relevance and indexing tell me whether a placement can move the business forward. Page quality beats placement quantity every time.
Seek New Buyer-Focused Domains

Nirmal GyanwaliFounder & CEO / WP Creative
I judge a backlink by one question. Is this a site my buyer would actually read?
Placements are easy to count. Revenue isn't. And I'm a CEO, so revenue is the only number I really care about. Leads are one step back from it. Traffic is two.
The lesson came from reviewing my own profile. I looked at the expert-quote sites linking to my personal site, nirmal.com.au. Seven of the eight also linked to my agency's site already.
Same small circle of sites, over and over.
A second link from a site that already knows you adds very little. It looks like progress in a report. It isn't.
So now I filter on: 1/ a new referring domain, not a repeat 2/ real organic traffic, not just a rating 3/ an audience that could actually buy
The last one is the one most campaigns skip.
One link from the right site beats ten from sites your buyer will never open.
Favor Earned Editorial Links

Chris RaulfInternational AI and SEO Expert | Founder & Chief Visionary Officer / Boulder SEO Marketing
I can't point to one specific placement that flopped and say it taught me everything, so I'll tell you what the full review showed. Over 18 months, 24 placements on high-authority sites produced 8 consultation requests that named one of them. Most of those links never produced a lead I could trace. Domain authority told me nothing about which ones would.
Before pursuing a link now, I ask three questions. Would this link exist if SEO didn't? Does a real editor vet the page, and do real readers see it? Does it fit what I'm actually known for? If any answer is no, I pass.
The lesson came from 2021. When we lost 80% of our traffic, the directory links and link exchanges we'd collected were the first suspects, even though their metrics looked fine. The links from speaking engagements and press coverage were untouched. So I stopped judging links by the score on the site and started judging them by whether someone chose to publish me.
Score Relevance Before Authority

Faizan KhanPR and Content Marketing Specialist / Ubuy Indonesia
With 8 years of experience in link building, I decide which links are worth pursuing by asking whether a placement can influence a buyer, not just a metric.
Last year, a campaign earned 60 placements, yet qualified referral traffic was minimal and demo requests stayed flat. Reviewing the weakest links, I found most sat on high-authority sites with little relevance to our audience.
One placement had a strong domain rating but sent only 9 visits in six months. We built a checklist and stopped pursuing generic roundups.
Over two quarters, we earned 24 placements instead of 60, but referral traffic grew 3.1x, organic demo requests rose 46%, and cost per link fell from $540 to $310. Cost per qualified lead also dropped 38%.
Lesson: Score every target on relevance, audience fit, and editorial context before authority, and judge links by audience, not authority alone.
Audit Traffic by Page Type

Jack GenesinSEO Consultant / Jack Genesin Consulting
The domain that made me change my screening process had monthly organic traffic of ~42k and a DA of 70+. But when I dug deeper, I noticed that ~80% of organic traffic went to the home page. The site ranked for only two dozen organic keywords. And of those, only one article had any organic traffic. In fact, that was the only blog post on the entire site. The link would have been on a blog that didn't show up in any site-wide scrape.
This is what I learned: You need to look at the page the link will be on, not the domain. In most cases, your link will be in a blog post. It's unlikely that your link will be on the home page, or the 'calculator' that gets most of the domain's traffic, or the viral post that gets the rest of the domain's traffic.
I just export the top 50 URLs for the site and classify them by page type. I want to see that the kind of page I'm buying a link on has a significant percentage of organic traffic. If I'm buying a link in an article about email marketing, I only care about the organic traffic to other articles about email marketing. If only one article gets SEO traffic, then I don't count the other 12 months of traffic in the neighborhood.
I'd rather have a domain with a smaller DA where half the blog posts get some SEO traffic, to one of those high-DA SaaS domains where 90% of traffic is to the home page, product pages, and product pricing pages. In short, the DA is about the building, but you only care about the foot traffic to the floors.
Track Buyer Behavior After Clicks

Will MitchellFounder / StartupBros
I pulled a placement I'd been excited about into my analytics and lined it up against my sales for the weeks after it went live. It was a high-authority general site, the kind that looks great in a monthly report. The referral column showed a trickle of visits, and most of them bounced off my product page within seconds. Nothing in my orders traced back to it, and the page it pointed at didn't move for any search term I cared about.
So I reopened the article and read it the way a buyer would. My link sat in a roundup with a dozen unrelated brands, under a headline nobody shopping my category would ever type. That site's readers had no reason to want what I sold.
Now every prospect goes through the same check before anyone spends an hour on outreach. I look at who reads the page, whether the article ranks for something my customers search, and whether my link sits where a reader would click it on purpose. Then I check which page it points to. A link into a product or category page that already converts gets priority over a homepage link every time.
Once a placement is live, I track referral visits, time on site, and assisted orders for it. Links that do nothing after a fair window come off my target list, and that budget goes to sites whose readers look like my buyers.
Match Destinations to Campaign Goals

JM LittmanCEO / Webheads
I open the article and click the link. That sounds rather basic, but it's easy to see a well-known publication on a report and assume the right people must be finding your website through it. I want to see where the link goes, what the reader is likely to expect and whether the page gives them a reason to stay.
I was quoted recently in Grit Daily about keeping in touch with agency clients. The link on my name took readers to LinkedIn. I'm pleased to have the coverage, and that's a perfectly sensible place for a reader to learn about me. It did make me think about how easy it is to count a mention as something it hasn't actually delivered. If the aim of a campaign is to introduce people to Webheads, I'd want to see whether the placement sends them to a useful page on our website.
When I review a placement now, I look at the subject, the publication's readership and the destination before giving much weight to a domain score. I'll check referrals and, where we can, what those readers went on to do. Some good coverage earns you recognition over time. Some sends people looking for a service today. Knowing what you wanted from the placement makes it a lot easier to judge whether to pursue another one like it.
Trace Links to Customer Leads

Matt BowmanFounder / Thrive Local
A backlink is a purchase, and I only buy what I can trace back to a customer calling or filling out a form. A placement count is an activity report. I run my company on month-to-month contracts, so clients can leave in 30 days, and that has made me allergic to any report that lists work done without saying what came back.
I start with the page, not the entire site. Which page generates revenue for the client? Which page is already on page one for a term someone would type? That's where we link to first. I'd take one site relevant to local buyers' interests over a ton of sites that have a high authority score.
The insight came from the owner I talked to last month. His last vendor sent him a report full of dozens of placements. He paid for every one of them. When I asked how many calls came from them, the owner had no idea. When we dug in, we found that most every link pointed to the home page or an old blog, on sites that weren't relevant to the service area. No page he paid to have linked on had actually done anything to generate revenue for him. He paid to learn about everything, but none had worked positively.
I now ask two questions before approving a placement. Would a customer even land on that page? And is the link pointing to something that can turn the visit into a lead?
If you can't tell me what it did for the phone, it was an invoice, not an investment.
Direct Links Toward Revenue Pages

Tristan HarrisSr. VP of Marketing / Next Net Media
I work backward from the pages that make money. For one client, I listed the five pages that produced the most enquiries and asked which topics and publications would naturally point to them. Many earlier placements had linked to blog posts that offered visitors no next step, so even the referral visits led nowhere. Reviewing one disappointing placement showed this clearly: a respected publication had linked to a general article with no service link, and visitors read it and left. After that, we targeted links to pages with a clear path to contact or purchase, and aimed for publications whose readers match the buyer. Before pursuing a link, I ask where a visitor would go after clicking and whether that destination can earn their business.
Favor Useful Contextual Mentions

Bryan VasquezHead of Sales / LinkBuilder.io
Before any outreach, a prospect has to pass three questions: would our customer read this site, would this page stay live for years, and does the link sit inside useful content? A "best resources" roundup on a quiet site once looked good by its metrics, yet visitors never appeared because the page was buried and the link was one of sixty. That placement taught me to treat link context as a quality signal as strong as the domain. I'd take a single link inside a helpful paragraph over a mention in a long list, even when the list sits on a bigger site. Short, strict filters save more time than long reports, and they keep a campaign focused on links that can influence results.
Measure Links by Lead Value

Brandon GeorgeDirector of Demand Generation & Content / Thrive Internet Marketing Agency
Each dollar invested in link development should be allocated to purchasing a lead. Link development should be treated with the same guidelines as any other business expenditure; develop a cost-per-lead model versus collecting a list of placements.
My single rule for determining whether to pursue potential targets is whether the link will have a positive effect on converting an existing page. A link directed to a page that is already converting on the second page for a buyer-intent search term will outperform a link to a blog post that receives little or no traffic. As a result, the Wilcox Investment Bankers targeted buyer-intent search term pages and saw 98% organic traffic growth and 74% conversion growth regardless of the volume of traffic generated.
This knowledge comes from one of my previous link campaigns. One of my links was posted on a well-known website, and my client was happy to see the logo on the report. However, the Analytics reported zero referral traffic, and our position did not change. Therefore, the link sat in a section of the site called a "dead contributor" and was nothing more than an attractive logo but a failed link.
With that in mind, I review three statistics following the publication of a link:
-Referral traffic. -Changes to my rankings for money-generating keywords. -New leads from the page. If these numbers do not change, I will no longer spend on that placement.
Links that do not lead to actual leads, are simply vanity links with a pretty logo.
Read Host Pages Like Editors

Timothy ClarkeSenior Reputation Manager / Thrive Local
The way I look at host pages is with an editor's eye. If I can't finish the article, I skip the link. Domain scores give a snapshot view of a site's performance from afar. The actual page is where I find out if anyone will see my client's content.
My starting point is always our own website. Which of our guide or service pages is closest to ranking for buyer-related keyword terms? This page receives my priority for links. Next, I locate websites that have readers who require that specific page. Finally, I will determine if the host page experiences sufficient levels of traffic. For an example of a successful finance client link-building campaign, we used our match-to-target buyer keyword search terms to link to the finance client's targets. As a result, this approach produced an increase in organic rankings of 1,118%. The successful placements we secured with the few relevant sites surpassed the number of general sites we linked to in other ways.
This is something I learned through experience! For example, I once received a quality link on a high authority website with a clean dofollow tag, so I put that link into the reporting process; then I read the content on that specific page. It was a very thin roundup page with many outbound links, and our link was located within an inconsequential paragraph on that page; therefore we received little, if any referral traffic from that source. The finance client's target URL stayed in the same position after we placed our link.
Now I check whether or not the surrounding paragraph would read correctly if our link was not present ON the host site as well as whether or not the host site has real levels of traffic. If either of these two conditions do not exist, I WILL pass on making the connection.
Inspect Durable, Trusted Sources

Rohan KaliaSenior AI SEO Specialist / COM8 STUDIO
The test I use now is simple: will this link still exist in six months, and is it on a site that buyers and AI tools actually look at? Placements are easy to count. Links that survive, pass value and sit in the right place are much rarer.
The lesson came from our own campaign at COM8 STUDIO this year. We built a batch of free links on social bookmarking sites, and they looked fine in the report. About three days later an anti spam plugin suspended the accounts and every one of those links disappeared. On another free platform the link was there, but it was wrapped in a redirect, so it passed nothing. Both would have counted as placements in a monthly report.
So now I check three things before I count a link:
1. Is it real? I open the page and inspect the actual href and the rel attribute, not just the anchor text.
2. Will it last? Sites where anyone can post instantly with no review tend to get cleaned up. One slow editorial placement is worth more than ten instant ones.
3. Is it a source buyers and AI tools use? In October we pulled the sources that AI answers cite for 16 competing agencies and software companies in Dubai. Clutch was cited for 15 of them, DesignRush for 12, GoodFirms for 11, and OpenStreetMap and Reddit for 10 each. Our own site had only a handful of cited sources. Most of those platforms give nofollow links, and they still matter more for being recommended than a pile of dofollow directory links.
The numbers I care about now are how many links are still live after 90 days, how many of the linking pages Google has indexed, and how often AI answers mention us for our core searches.
Confirm Links Before Publication

Razvan LucaFounder / Talval Research
The disappointing one taught me more than the good ones. In August a real financial news site published a bylined article of mine. Proper publication, my name on it, an author box saying I founded Talval Research. I counted it as a win. Then I opened the page source: zero links to us. Not nofollow. None. The company name was plain text.
Two rules came out of that. First, the link and its exact URL get agreed in writing before the article goes anywhere. No link, no article, and we don't pay to have one added afterwards. Second, I check the anchor after publication, not the mention. A mention without an teaches a search engine nothing, and it's easy to miss because the article looks great.
For deciding which links are worth chasing at all, I ask three things. Does the site's audience overlap with the people who'd actually use the product? Is the link in the body, next to the claim it supports, or parked in a bio? And would a reader click it to check a number?
The placements that worked for us cited a specific figure from our data, so the link was there for the reader to verify it. In my experience that's also the easiest link to ask for. An editor will add a link that lets their readers check a number far more readily than one framed as a favour.
Vet Publisher Credibility Individually

Md Mizanur RahmanFounder and CEO / Invrse Marketing LLC
When a backlink campaign brings plenty of placements but little business impact, we stop trusting the domain rating and vet each placement on its own. Today, a link is worth pursuing only if the source is a genuine publication rather than a link-selling site, the page has an identifiable author and internal links from the rest of the site, and the topic is relevant to what our client sells.
We learned this early on, before we had an in-house outreach team. After finishing the on-page and technical work for a client, we hired a freelancer for $650 to deliver ten guest posts on DR 60+ sites. The samples looked good and the budget fit, so we agreed. A month or so after delivery, we saw no real impact, so we audited all ten.
Four were solid. They were relevant, ranking, and sending referral traffic, for example a luxury wedding limousine client linked from a page ranking for event car rental. The other six each failed differently: some sat on sites known for selling links, some were not indexed or ranking at all, two had no author profile, and some were orphaned. One put our luxury limousine client in a comparison against cheap taxi and ride-hailing apps, and others used anchors like "click here to get more offers."
Since then, we build links through content exchange: we write for publishers who need relevant content, agree on the author and a contextual link upfront, and ask their editors to link to the piece from other pages. Because we control the content, quality is no longer left to chance.
Choose Sources AI Already Cites

Daniel GohChief Marketing Officer / RankAISearch
1. I use the free version of the test for one service in a given area to see the best sources cited by ChatGPT, Claude and Gemini (with web search enabled) for that "best" query. These are the sources that search engines already trust for that type of query, so these are the types of placements that I care about getting for my clients. In October 2024 I did a test with 10 local Singapore queries and got 28 usable answers with 89 sources cited total. Of those cited sources, one directory, threebestrated.sg, received 15 mentions of its links. So while this is a very small sample size and only one day's test, no DR score was needed to tell me that threebestrated.sg was getting great links for its directory. I then check one more thing for each cited link: does the page clearly state who they are, what type of content they produce and where they are located? If a link has my exact brand name in the URL and states accurately who I am as owner of said website in the link description, then that helps search engines like chatbots understand who I am too. A simple anchor text on an unrelated page does not.
2. When I did a "clean up" of the RankAISearch domain I was shocked to find hundreds of "directory" links of a spammy nature that I had never added to the site. None of these links brought any visitors to the site and none of the "AI powered search engines" cited the links on their results. A potential liability. My views on adding links have now changed to the view of whether or not I would be happy for visitors to find my site through a given page? If not then regardless of cost or "power" then I shall avoid adding said links. I have stopped purchasing links to try to improve my search engine rankings. What I have is good in terms of accurate directory listings, quality reviews and practical use on the internet to create more content that people can use. I have no idea if any of these links will improve my ranking. All I can do is set up a baseline for my site using Search Console and Bing Webmaster Tools, then change one thing at a time and track the results.
Test Article-Level Audience Relevance

Jacob PeistrupFounder / Formbreak
Several paid placements taught me to look beyond whether a website is broadly "home-related." I'd requested links on highly relevant real estate websites, with articles about the real estate market. What I received were placements in home-and-garden, interior-decor and architecture magazines. None of the articles addressed the market the way I'd intended.
Those subjects overlap with real estate, but that overlap wasn't enough for what I wanted. The lesson was to evaluate the actual article and its audience, not just the website's category or authority score.
I now ask: if everyone reading this page clicked through, would I want that traffic? For a local real estate website, I need a clear connection between those readers, the location I serve and the page receiving the link.
That's a relevance test. I don't expect every backlink to produce an inquiry I can directly attribute to it. My broader aim is search visibility, and isolating one link's contribution is difficult. But I should still be able to explain why that particular link belongs there.
Require Indexing and Organic Traffic

Ric RubensteinFounder / Kadima Digital
I judge a backlink campaign by whether the keyword moves, not by what one link sends. The goal is pushing a ranking toward the first few positions, and the traffic follows.
To pick links, I look for sites with a higher domain rating than ours, at a fair price, so the authority passes to us. They also have to be on topic and clean, nothing spammy. The one check I run before accepting any placement is organic traffic. The site needs at least 1,000 organic visitors a month before we'll even consider it.
If the keyword isn't moving, I put more placements behind it rather than stop.
My worst lesson was a paid placement Google never indexed. The provider had no re-indexing process, and some pages sat for six months while I couldn't reach the webmaster. Indexing is now part of how I judge a link.
Earn Links Buyers Would Value

Snket DesaiDirector of SEO & Chief Branding Officer / Spot Easy
I judge a link by what it can move, not by the logo it adds to a report. A link is worth pursuing when the page it sits on ranks for something my buyers search, sends referral visits that behave like customers, or strengthens the topical authority of a page that already earns revenue. The disappointing-placement lesson that changed my process: a batch of placements on respectable-looking sites produced almost no movement, because none of those pages ranked for anything and none of them were ever read by our audience. Since then, every target has to pass a simple test before outreach — would this page still be worth earning if the link were nofollow? If the audience and the topic are right, the answer is yes; if not, the placement was decoration. At Spot Easy, tying link work to pages that drive clicks — up 48% year over year in Search Console — keeps the campaign honest.
Reject Decorative Spam Backlinks

Nabin ThakurFounder and SEO Recovery Specialist / SEORevive
I judge a link by one question first: would a real person click it and land on a page that helps them? If not, the link is decoration, whatever its metrics say.
The lesson came from our own profile at SEORevive. A backlink tool reported 372 referring domains to seorevive.in while our Domain Rating sat at 1.1. When I went through them one by one, only about 3 were real sites that had chosen to link to us. The rest were scrapers and spam domains. That 372 had looked like progress for months.
Since then I check three things before chasing any link:
1. Is it actually followed? I checked 21 agency directories by hand. Most were nofollow or showed no website link at all. Two passed a followed link. 2. Does the site get real traffic of its own, from people rather than from other link sites? 3. Is the page about the same problem my page solves?
Links that pass all three are rare, and they usually come from publishing something people need to cite, like original data. Two placements that pass are worth more than fifty that don't.
Build Links Through Client Partnerships

Silas BaunFounder & Web Developer / ForaWeb
When I analysed the backlink profiles of seven Danish web agencies, the strongest one was not built on placements at all. Almost all of its 76 real referring domains were 'made by' credits in the footers of client websites. Directory listings and spam domains made several other profiles look bigger in SEO tools, but that is not what the strongest profile was built on. It changed how I judge a link: I ask one question, would a real customer ever click it? Client sites, partners and local media pass that test.
Align Placements With Campaign Goals

Anja Moskon SolnMarketing Comms & Demand Manager
It starts with the goal. A link campaign can serve visibility and awareness, or it can drive conversions, and those call for very different placements. Most "disappointing" campaigns weren't failures; they were judged against the wrong goal.
For awareness, I look at who actually reads the site. Is it where our buyers go to solve problems, such as a trade publication, an industry association or a niche community, or just a high-authority domain with the wrong audience?
For conversions, I work backwards. I check which referral sources already bring visitors who convert, then put more effort into those sources and similar ones, rather than chasing new placements.
The lesson came from comparing two placements. Articles on larger sites brought zero conversions, but Search Console showed a rise in organic traffic and branded searches, so they were doing awareness work. Meanwhile, a simple listing on a community page brought specific inquiries. Since then, I don't write off a placement too early. I first ask which job it was doing, then invest in each kind for its own goal.
Serve Readers’ Next Tasks

Anatolii UlitovskyiFounder / SEOtools.pro
I stopped judging links by domain score when I looked at where our own traffic comes from. In Search Console for our SEO platform, every top page by clicks is a free tool page. One tool page brought 20,609 clicks in 16 months, 13 times more than the homepage. That changed my question. I no longer ask how strong a site is. I ask whether a real person on that page will want what my page offers.
Now I check three things before I pursue a link. Does the site rank for topics close to ours? Does the section where the article will live get search traffic? Does the link point to a page that solves the reader's next task? If one answer is no, I skip it at any price.
The placements that disappointed me had strong authority scores and no readers. They were live, indexed and invisible. A link nobody can click is a line in a report, not a business result.
Avoid Isolated Contributor Sections

Vaishik KansaraSEO Executive / eCare Infoway LLP
If your link-building activity is driving lots of traffic but no revenue or rankings, it's almost certainly because you're going for the third-party metrics. Over the years, I've moved away from looking purely at the DR/DA and decided to look at things from a customer's point of view to decide what links to go for.
I look for: Actual traffic - Do they already have articles that are ranking for competitive non-branded keywords? It's no good trying to get links from a site that doesn't have enough credibility in Google's eyes to rank for their content.
Real audience - I'd much rather get one link from a small niche publication that actually reaches decision makers than 10 links from general lifestyle sites.
Editorial strictness - Look at the latest articles on the site. If they are full of paid links with loads of exact match anchor text, then avoid. Good links will come from sites where the editors fight for their links, ask for real data and genuinely want to help their audience.
Lesson learnt: A few years ago we were celebrating a trophy link on a well-known business domain with a DR higher than 75. Three months down the line, our target page had not improved in rankings, and there was no referral traffic showing in our analytics. Looking at the live URL, I found a separate subfolder for guest contributors with no links to the main site. This meant that crawlers weren't visiting the site much and no real traffic was reaching it. This definitely changed how we approached link building. We now take a look at how their contributor pages are linked internally. Are the pages staying indexed? Is the site actually increasing?