Every few months someone declares links dead. Every few months a site with better content and worse links loses to a site with worse content and better links, and the declaration quietly expires. Links remain one of the ways search engines decide which of several relevant pages deserves to be trusted. The problem is not whether links matter. The problem is that the activity most companies call "link building" produces links that do not.
I want to be precise about what works, because the gap between the link-building industry and link-building reality is wider in India than in most markets. This guide covers how links are evaluated, what the paid-placement economy actually buys you, how to earn links with assets, and the authority lever I use before any outreach: the links you already control.
How a link is actually evaluated
A link is a vote, but votes are weighted, and the weighting is what the industry gets wrong. A link's value depends on the page it sits on, the context around it, the site that page belongs to, and whether the pattern it belongs to looks earned or manufactured.
| Signal | What raises the value | What lowers it |
|---|---|---|
| Relevance of the linking page | The page is about a closely related topic and the link makes sense in the sentence | A generic "resources" page, an off-topic blog, a sidebar on an unrelated site |
| Position and context | In the body copy, surrounded by text that explains why the link is there | Footer, sidebar, author bio, a list of forty links |
| Quality of the linking site | Real audience, real editorial standards, its own organic visibility | Site exists to sell links; thin content; traffic that is mostly bots |
| Anchor text | Natural and varied: brand names, URLs, descriptive phrases | Exact-match commercial keywords repeated across many sites |
| Pattern | Links arrive over time from diverse sources, matching the content that earned them | Bursts of links from the same networks with the same anchors |
| Link attributes | Standard followed link in editorial context | rel="sponsored" or "nofollow" where appropriate; hidden or scripted links |
Two things follow from this. First, a single link from a page that genuinely ranks for a related topic can be worth more than a hundred links from directories and guest-post farms. Second, search engines have had well over a decade to learn what manufactured link patterns look like. They do not need to penalise most paid links; they simply learn to ignore them. You pay, and nothing happens, and the vendor's report shows a hundred new links. Both of you are satisfied and the rankings have not moved.
Why most Indian "link building" is link buying
I will describe the common model plainly. An agency offers a monthly package of "high-DA guest posts". Behind that package is a marketplace of site owners who accept payment to publish an article with your link. The articles are written cheaply, often by the same handful of writers, and the sites exist primarily to host them. The agency marks up the placement, delivers a spreadsheet of URLs with a domain metric next to each, and the client sees activity.
This is link buying with extra steps. It is not outreach, because nobody was persuaded of anything; it is a transaction. Google's guidelines are explicit that links bought for ranking purposes should carry a sponsored attribute, and links that do not are a scheme. The risk is rarely a dramatic manual action, though those happen. The common outcome is simpler and worse: the money is spent and the links are discounted, so you have purchased nothing.
I do not say this to be pious about it. There are grey areas, and there are markets where editorial placements are an accepted part of PR. My objection is practical: the paid-placement model consumes budget that would compound far better invested in something that earns links on its own. That is the model I actually use.
Earning links with assets
Links are earned by things worth referencing. Most companies have nothing worth referencing because everything on their site is a sales page or a blog post that paraphrases what already ranks. Before I think about outreach, I think about what asset would make a journalist, an industry blogger or a competitor's content team want to cite this company.
Four asset types that earn links in my experience
Original data
Aggregate something only you can see. A logistics company can publish delivery-time benchmarks by city. A SaaS product can publish anonymised usage patterns. A marketplace can publish price indices. Data is cited because the writer cannot get it anywhere else, and it continues earning links for years if you update it.
Tools and calculators
A small, useful tool is referenced far more often than an article about the same subject. My own SERP snippet preview and robots.txt tester exist partly because practitioners link to tools they use. The tool must actually solve a task; a thinly disguised lead form earns nothing.
Frameworks and methods
A named, documented way of doing something becomes the thing people cite when they explain it. This is why I publish my frameworks openly. A framework is harder to copy than a listicle and gets referenced by people teaching the topic.
Original research and experiments
Testing a claim and publishing the method and result, including when it fails, earns links from people who argue about it. My SEO Lab is built on this. Opinion is cheap; a documented test is scarce.
Notice what is not on the list: infographics summarising other people's data, "ultimate guides" that restate consensus, and expert round-ups. Those earned links a decade ago. Today they are what everyone produces, which is precisely why they no longer stand out.
Digital PR done honestly
Digital PR is the practice of getting assets like the ones above in front of journalists and editors who might cover them. It works, and it is also the area with the most inflated promises. The part that works is narrow.
- Lead with the story, not the brand. A journalist covering Indian e-commerce wants "delivery times in tier-2 cities are now faster than tier-1 cities for category X", not "Company Y launches a report".
- Target publications your buyers read. A link from a national business daily is good. A link from the trade publication your buyers actually read is often better for both authority relevance and referral value.
- Make the data easy to use. Provide the chart, the methodology, a quotable line and a named person to attribute it to. Reduce the writer's work to near zero.
- Accept that coverage without a link still counts. Brand mentions matter more than they used to, particularly for how AI systems describe an entity, which I cover in the AI SEO guide.
What I do not do is send templated "I loved your article, here is mine" emails at scale. Reply rates are close to zero and the links it produces are the low-value kind in the table above. If an asset is not interesting enough to pitch in two honest sentences, the fix is a better asset, not more emails.
Unlinked mentions: the easiest wins you are not claiming
If a company has existed for a while, people are already mentioning it without linking. Partners list it on their sites, customers mention it in case studies, journalists name it in passing. Each of those is a link request with a high success rate because the author already chose to mention you.
The process is mechanical: search for the brand name, product names and the names of senior people, exclude your own domain and social profiles, filter to pages where a link would be natural, and ask politely. The ask is one sentence. Most of these pages are maintained by someone who will add a link in thirty seconds if it is easy. This is the first outreach activity I run on any site, before any new asset is built, because it costs nothing but time.
Internal linking: the authority lever you already control
Here is the part most link-building conversations skip entirely. The authority a site has already earned is distributed through its internal links, and on most sites that distribution is accidental. The home page and a few old posts hold most of the authority; the commercial pages that need it are three clicks deep with two links pointing at them.
Before spending a rupee on external links, I rebuild the internal structure so that the authority that exists flows to the pages that matter. On sites with a reasonable existing backlink profile, this is the single highest-return activity in the authority category, and it needs no vendor and no outreach.
- Map authority by URL
- Identify pages that need it
- Add contextual links from strong pages
- Fix orphans and depth
- Measure the shift
The method is documented in my Internal Linking Framework and the detailed mechanics are in the internal linking guide. I tested the effect at scale in the internal linking at scale experiment, which is a better argument than anything I can write here. The short version: the pages that gained links from existing strong pages moved; the ones that did not, did not.
What to ignore
The link-building industry runs on metrics it invented, and the most damaging is the obsession with Domain Authority and its equivalents. These are third-party estimates built from the vendor's own crawl. They are not used by Google, they can be manipulated by the same link farms that inflate them, and they tell you nothing about whether a specific page is relevant or read.
A short list of things I have stopped paying attention to:
- Domain-level authority scores as a target. Useful as a rough filter for obviously junk sites; useless as a goal or as a measure of a link's value.
- Link counts. Fifty links from one network equal one link. Linking domains, and the quality of each, are what matter.
- Link velocity rules. "Build ten links a month consistently" is a vendor's billing schedule, not a ranking signal.
- Disavowing by default. Unless there is a manual action or a clear history of paid links, disavow files are busywork. Google ignores most bad links on its own.
- Reciprocal and three-way link swaps. Small-scale exchanges between genuinely related sites are fine. Organised swap networks are the same pattern as paid links and are treated the same way.
I wrote about several of these in the note on SEO advice I stopped following, because the cost of these habits is not just wasted budget. They make teams feel busy while the real levers stay untouched.
Measuring authority honestly
If domain scores are out, what replaces them? I look at outcomes and at a small number of inputs I can verify.
| Measure | Source | What it tells me |
|---|---|---|
| Referring domains with organic visibility of their own | Backlink tool, filtered to domains with meaningful organic traffic | Whether links come from sites that search engines already trust |
| Links to commercial and hub pages, not just the home page | Backlink export grouped by landing URL | Whether authority reaches the pages that need to rank |
| Brand search volume over time | Search Console queries containing the brand | Whether people are hearing about the company outside search; the best proxy for real authority |
| Rankings on competitive non-brand queries after content is fixed | Rank tracker, segmented by page type | Whether authority, rather than content, is the constraint |
| Unlinked mentions converted | Simple spreadsheet log | Whether the cheapest wins are being claimed |
The diagnostic question I ask is: on queries where our page is clearly as good as or better than what ranks, do we still lose? If yes, authority is the constraint and link work is justified. If our content is thinner than the competition, more links will not fix it, and the budget belongs in content. I see the opposite diagnosis made constantly, which is why I treat links as a component of an SEO strategy rather than a standalone programme.
Links are earned by things worth referencing. If nothing on the site is worth referencing, outreach is a tax on that fact.
Frequently asked questions
Do links still matter for SEO?
Yes. Links remain one of the signals search engines use to decide which relevant page to trust, especially on competitive queries. What has changed is that manufactured links are increasingly ignored rather than penalised, so the question is not whether to build links but whether the links you are building are the kind that count.
Is buying guest posts risky, or just ineffective?
Both, in different proportions. Manual actions for unnatural links do still happen, but the far more common outcome is that paid placements on sites built to sell links are simply discounted. You pay for activity, not impact. The budget is better spent on an asset that earns links on its own.
What is a good Domain Authority to aim for?
None. Domain Authority and similar scores are third-party estimates that Google does not use. They are a rough filter for obviously low-quality sites and nothing more. Aim instead for links from pages that rank for related topics and have real readers, and measure outcomes like brand search growth and rankings on competitive queries.
How many links do I need to rank?
There is no number. It depends on the query, the competition and whether your content is already as good as what ranks. The useful test is whether you lose on queries where your page is clearly comparable; if so, authority is the constraint. If your content is thinner, links will not compensate.
Should I disavow bad links?
Only if there is a manual action, or a documented history of large-scale paid links that you want to clean up. For most sites Google already ignores low-quality links, and disavowing by default risks removing links that were helping. It is rarely a good use of time.
Can internal links replace external links?
Not replace, but redistribute. Internal links move the authority you already have to the pages that need it. On sites with an existing backlink profile this is often the higher-return activity, and it should be done before buying or earning new external links, so that any new authority also flows to the right pages.
Where this connects
The internal side of authority is documented in my Internal Linking Framework and tested in the internal linking at scale experiment. The technical SEO and content architecture case study shows what happens when structure and authority flow are rebuilt together, and my SEO strategy service is where I decide, for a specific business, whether links are the constraint at all. If you are working through the whole discipline in order, this guide sits at Level 06 of the SEO Roadmap.