The uncomfortable fact about link building in 2026 is that most links a business pays for do nothing. They are not penalised, which is what everyone fears. They are simply ignored, filtered out by systems that have become good at recognising a link nobody would have given away for free. The budget still leaves the account, the report still shows a rising referring domain count, and the rankings do not move.
That is the context in which link building services have had to change shape. The work that still moves search visibility looks less like a procurement exercise and more like earned media with a technical brief attached. Understanding the difference is what stops a Malaysian business from spending twelve months building an asset that has already been discounted to zero.
What Google Actually Says About Links Now
Google’s spam policies treat link schemes as any arrangement intended to manipulate rankings, and they name the obvious cases: buying or selling links that pass ranking signals, excessive link exchanges, large-scale guest posting with keyword-rich anchors, and links embedded in widgets or templates across many sites. The policy also covers the reverse direction, where a site sells placements that pass PageRank.
Two mechanisms enforce this, and they have very different consequences.
- Algorithmic devaluation. Since Penguin became part of core ranking, unnatural links are largely neutralised rather than punished. Google’s spam detection systems identify patterns and discount the links. Nothing appears in Search Console. The site simply does not benefit, which is why so many campaigns produce impressive reports and flat results.
- Manual action. A human reviewer at Google can issue an unnatural links action, which does appear in Search Console and does suppress rankings until it is resolved through cleanup and a reconsideration request. This is rarer, and it is usually the result of a pattern too blatant to ignore.
The practical implication is that the question to ask about a proposed link is not is this safe. It is would this link exist if search engines did not. If the answer is clearly no, the realistic expectation is that it will be discounted, and the money is wasted whether or not anything bad happens.
The Tactics, Ranked by What They Actually Return
Not every method is equally durable. The table below is a working assessment of the main approaches by the effort they require, how long the value tends to last, and how exposed they are to the next spam update.
| Approach | Effort | Durability | Risk profile |
|---|---|---|---|
| Digital PR and data-led stories | High | High | Low |
| Original research or proprietary data | Very high | Very high | Low |
| Expert commentary to journalists | Moderate | High | Low |
| Genuine guest contributions to relevant publications | Moderate | Moderate | Low if editorial, high at scale |
| Broken link and unlinked mention reclamation | Low to moderate | Moderate | Low |
| Industry associations, chambers, and event listings | Low | Moderate | Low |
| Paid placements on general blogs | Low | Low | High |
| Private blog networks | Low | Very low | Very high |
| Automated directory and profile submissions | Very low | None | Moderate to high |
The pattern is consistent. Everything durable requires something the business already has and can share: data, expertise, a customer base, a physical presence, or a genuine story. Everything cheap is cheap because it is replicable at scale, and anything replicable at scale is exactly what detection systems are built to find.
Digital PR Is Where the Durable Links Come From
Data you already own
The most reliable link magnet for a Malaysian company is usually a dataset nobody else has. A property firm has transaction patterns by neighbourhood. A logistics operator has delivery time distributions across states. A recruitment agency has salary and vacancy movement by sector. Turning that into a short, clearly presented piece with a headline finding gives journalists at outlets like The Star, The Edge Malaysia, or New Straits Times a reason to cite you by name, and citations from national publications are the links that survive every update.
Expert commentary
Reporters need sources on deadline. A named executive who responds quickly, in usable language, on a defined set of topics will accumulate mentions steadily. This is slower than buying placements and it compounds, because a journalist who has used you once will come back. It also produces the entity signals that matter beyond links alone.
Reclamation work
Before commissioning anything new, find the mentions that already exist without a link and the pages that link to a competitor URL that now returns a 404. Both are cheap conversations rather than campaigns. Reclamation is consistently the highest return per hour in link building and is consistently skipped because it does not look impressive on a proposal.
Local Acquisition in the Malaysian Market
For businesses serving Kuala Lumpur, Penang, or Johor rather than a global audience, relevance beats authority metrics. A link from a Malaysian industry association, a state chamber of commerce, a university partnership page on a .edu.my domain, or a well-run local news site does more for local visibility than a higher-scoring but unrelated foreign blog.
- Sponsorships and community involvement that produce a genuine listing page rather than a logo image with no link.
- Membership directories of associations the business actually belongs to.
- Bilingual outreach. English-language pitching leaves the Bahasa Malaysia media landscape untouched, and competition for those placements is far lower.
- Supplier and partner pages, where an existing commercial relationship justifies a link naturally.
- Local event participation, including speaker bios and agenda pages that persist long after the event.
- Google Business Profile and structured local citations, which do not pass ranking signals the way editorial links do but underpin local pack visibility.
Anchor Text, Velocity, and the Patterns That Get Profiles Discounted
A naturally acquired profile is messy. Most links point at the homepage, most anchors are the brand name, a bare URL, or something functional like read more, and only a minority contain commercial keywords. Profiles built by a vendor tend to look the opposite: a suspicious concentration of exact-match commercial anchors pointing at money pages, acquired at a steady monthly rate that matches a retainer cycle rather than anything happening in the real world.
Three specifics worth enforcing in any brief. Keep exact-match commercial anchors to a small minority of the total. Vary the destination, because a profile where every link points at one service page is self-evidently constructed. And accept irregular timing, since real coverage arrives in bursts around launches, research, and news rather than in equal monthly instalments.
One more point on metrics. Domain Rating and Domain Authority are third-party estimates produced by tool vendors, not signals Google publishes or uses. They are useful for filtering out obvious rubbish and useless as a target in themselves, which is why vendors who sell by metric threshold are optimising for the number on your report rather than for your rankings.
How to Tell Whether Any of It Is Working
- Track rankings for the specific pages being linked to, not site-wide averages. Link impact is page-level first.
- Watch referring domains that are genuinely new, and discount duplicates from the same network or IP footprint.
- Measure referral traffic and what it does after arriving. A link that sends engaged visitors has value regardless of ranking effect.
- Compare against a control set of pages receiving no link work, so seasonal and algorithmic movement can be separated from campaign effect.
- Allow a realistic lag. Discovery, crawling, and the settling of ranking changes commonly take weeks to a few months, so judging a campaign at four weeks is judging noise.
- Review the actual placements, not the spreadsheet. Open ten links at random and read the pages they sit on.
Buying Without Getting Burned
If you are commissioning professional services rather than building in-house, the diligence is straightforward and most buyers skip it.
- Ask to see the last ten placements delivered for a client in a comparable sector, with live URLs.
- Ask whether placements are paid, and if so how they are disclosed. A fee arrangement without a sponsored or nofollow attribute is a policy breach the client site carries.
- Ask who writes the content and whether it is reviewed by anyone with subject knowledge.
- Ask what happens if a link is removed. Replacement policy and monitoring should be in the agreement.
- Ask for the reporting definition of a link, since some vendors count a mention, a redirect, or a nofollowed profile as a delivered placement.
- Be sceptical of guaranteed volumes at a fixed monthly number. Editorial outcomes cannot be guaranteed, and a vendor who guarantees them is buying, not earning.
Frequently Asked Questions
Do backlinks still matter for rankings?
Yes, but the returns have narrowed. Links remain one of the ways Google assesses whether other sites vouch for a page, and competitive commercial queries are still difficult to win without them. What has changed is that low-quality links are increasingly filtered rather than counted, so volume alone no longer produces movement.
Is buying backlinks against Google’s policies?
Exchanging money for links that pass ranking signals falls squarely within Google’s link spam policies. Paid placements are permitted if the link is marked as sponsored or nofollowed, which preserves the marketing value while removing the ranking signal. The risk sits with undisclosed paid links, which are the ones detection systems and manual reviewers look for.
How many backlinks does a site need?
There is no threshold. The useful comparison is against the pages currently ranking for your target queries: how many distinct, relevant domains link to those specific pages. That number varies enormously between a local service query in Selangor and a national commercial term, which is why any vendor quoting a universal figure is guessing.
How long before link building shows results?
Typically weeks for discovery and crawling, and often two to six months before ranking effects are clear, longer in competitive niches. Anything faster usually indicates the page was already close to the threshold. Campaigns should be evaluated over quarters, with leading indicators such as placements secured and referral traffic reviewed monthly.
Should I disavow bad links pointing at my site?
Usually not. Google states that it ignores the vast majority of unnatural links automatically, and careless disavowing can remove links that were helping. The tool is appropriate when there is a manual action to resolve, or a clear history of paid or manipulative links the site built itself and cannot remove.
Are guest posts still worth doing?
Genuine contributions to publications your audience actually reads are fine and always have been. Guest posting at scale, with keyword-rich anchors, on sites that accept anything, is named in the spam policies and is treated accordingly. The test is whether you would still want the placement if the link were nofollowed.
What to Do Next
Before approving another retainer, run one audit: export every link acquired in the last six months, open them, and mark each as one you would still want if it passed no ranking signal at all. If fewer than half survive that test, the problem is the brief, not the budget, and redirecting the same spend into one piece of original data about your market will almost certainly outperform another year of placements.
Related reading: Link Building for Lawyers, and more in Business Law.
