Link building for ecommerce stores

By Bill Hartzer · Hartzer Consulting · 25+ years in search, domains and expert witness work

Nobody links to a product page

Start with the uncomfortable fact, because every mistake in ecommerce link building follows from refusing to accept it. Individual product pages almost never earn editorial links, and no amount of outreach changes the underlying reason.

People link to things for a small set of reasons: the page is a source of information they need to cite, it is a resource worth sharing, it makes an argument worth arguing with, or it is the definitive location of something. A product page is a transaction interface. It exists to sell one item to one person. A blogger writing about the topic has no reason to send their reader to a checkout, and a journalist has an active reason not to.

There are exceptions and they are narrow: a product that is genuinely unique or newsworthy, a product page that has become the de facto reference for a specification, and affiliate or gift-guide contexts where linking to a purchasable item is the entire point of the article. Those are worth pursuing when they apply. They do not constitute a program.

The consequence is that ecommerce link building is nearly always an internal linking and authority distribution problem in disguise. You earn links to pages that can earn links, and you build the site so that authority flows from those pages to the categories and products that convert. If your linkable content sits on a separate blog subdomain or is orphaned three clicks from anything commercial, you have earned links for nothing.

Category and buying guide pages sometimes do

One level up from the product, things improve considerably. Category pages and buying guides can earn links, because they can be genuinely useful to somebody who is not ready to buy.

A bare category page — a grid of products with a paragraph of keyword text above it — earns nothing, for the same reason a product page does not. A category page built as a reference does much better. That means the page answers the questions somebody choosing within the category actually has: how the variants differ, what sizes or specifications mean, what to look for, what the common mistakes are, and which choice suits which situation.

The buying guide is the fully developed version of this, and it is the single most productive page type in ecommerce link building. The ones that earn links share some traits:

• They are useful to somebody who buys elsewhere. A guide that only makes sense if you buy from this store is an advertisement. A guide that would help a reader at any retailer is a resource, and resources get cited.
• They contain something specific and verifiable — a comparison table of real specifications, measurements, a sizing chart that is actually accurate, a compatibility matrix.
• They say what not to buy. A guide willing to state that a whole product category is a waste of money is far more citable than one that recommends everything.
• They are maintained. Guides that go stale lose their links quietly. Roughly half of all links are eventually lost, and nearly all of that loss happens while the page is still live — the link is simply removed or the page around it is rewritten.

The other category-level asset worth building is the reference page nobody in your industry has bothered to make well: the sizing standard, the materials glossary, the care instructions, the compatibility chart, the historical model list. These pages are dull, they take a day, and they accumulate links for a decade.

Suppliers, manufacturers and stockist listings

This is the most reliable and most neglected link source in ecommerce, and it requires no content at all.

Almost every brand you stock has a "where to buy", "stockists", "authorized retailers" or "find a dealer" page. Being on it is normal commercial practice, the manufacturer generally wants it, and the link comes from a site with genuine standing in your exact category. Most retailers have never asked.

The full inventory of relationships worth working:

• Brands you stock. Ask your account manager, not a generic address. Frame it as helping their customers find the product.
• Brands you have stopped stocking where the listing may still exist and simply needs updating.
• Trade bodies and industry associations you belong to, most of which have member directories nobody maintains.
• Certification and standards bodies — safety, organic, fair trade, professional accreditation. Certified-supplier lists are strong, relevant and stable.
• Wholesalers and distributors with retailer directories.
• Payment, shipping and platform partners with customer showcases.
• Local business bodies if you have a physical presence.

Two practical notes. Ask specifically — a request to "add a link" gets ignored, a request to "add us to your stockist page for the UK north-west, here is the address and the URL" gets actioned. And audit these annually, because listings get dropped during site rebuilds without anyone telling you.

Unboxing, review and creator outreach

Sending products to people who publish about them is the tactic most associated with ecommerce, and it is more complicated than it looks.

The mechanism is straightforward: a reviewer receives a product, writes or films an honest assessment, and links to the product or the store. When it works it produces a relevant link, genuine referral traffic, and content you can use elsewhere. When it fails it produces nothing, or worse, it produces a disclosed paid placement that is neither a good link nor a good advertisement.

The things that determine which outcome you get:

• Send to people who cover the category, not to people with large followings. A niche reviewer with a small, specific audience produces a better link and a better customer than a generalist with a big one.
• Accept that you cannot control the review. If you cannot survive an honest assessment, the problem is the product.
• Understand the disclosure position. Gifted products generally require disclosure, and in many jurisdictions the resulting links are expected to be marked as sponsored. That is the correct treatment and you should not ask anyone to avoid it. A nofollowed link from a review a real audience reads still does work.
• Do not pay for a followed link. The moment money is exchanged for a link specifically, you are buying links, whatever the invoice says.
• Budget for the failure rate. A meaningful share of products sent will never be covered. Price that in before you start.

The version of this tactic that reliably disappoints is mass creator outreach at low value — sending cheap items to hundreds of accounts. It generates a footprint of near-identical posts and very little else.

Ecommerce link building is content marketing wearing a different hat

Having worked through the tactics, the honest summary: with the exception of supplier listings, almost everything that earns links for an ecommerce site is content marketing. The store is not the linkable thing. The material around the store is.

That has three consequences worth confronting directly.

First, the skills required are editorial rather than commercial. Writing a guide that a journalist will cite, producing original data about your category, photographing something better than anybody else has — these are publishing capabilities. Retailers frequently try to buy links without building any of them, which is why so much ecommerce link building ends up as paid placements.

Second, your merchandising data is a genuine and underused asset. You know what sold, what got returned and why, how demand moved by season and region, and how prices moved across a category. Aggregated and anonymized, that answers questions journalists ask constantly. A retailer sitting on years of transaction data has better material than most agencies could commission, and almost none of them publish it.

Third, the internal linking question decides whether any of it pays. Links earned by a guide help commercial pages only if the site's structure passes that authority along. Put the linkable content on the main domain, link it directly to the relevant categories from within the body copy, and make sure it is reachable from navigation rather than stranded in a blog archive.

There is one more thing ecommerce sites should do before any campaign: check what you already have. Retailers change platforms, restructure categories, and retire product lines constantly, and every migration strands links. Nearly all link loss happens while the linking page is still live and the target has moved. Mapping old URLs that still receive external links to their current equivalents is usually the highest-return afternoon available.

What to avoid, and what to measure

Three tactics are pushed hard at ecommerce and deserve a plain warning.

Coupon and deal sites. The links are usually paid, usually placed at scale by everybody in your category, and often nofollowed. There is a legitimate commercial reason to be on some of them — customers use them — but treat that as an acquisition channel with a cost, not as link building.

Product review swaps and mass gifting programs run through intermediaries. These produce templated posts with identical structure across hundreds of sites, which is a pattern, and patterns are recognizable.

Sitewide footer links from partner stores. A single sitewide link can produce an enormous number of links from one domain and add almost nothing. On my own profile a single domain accounts for 16.8% of all links through one sitewide placement, and the top 100 domains account for 84.3% of the total. Link counts flatter; referring domain counts inform.

Measure accordingly: referring domains by quality band, the number of category and guide pages that hold at least one external link, supplier listings live versus lapsed, and referral traffic from reviews. Rankings for commercial terms will move if the rest of the site deserves it, but they are a lagging and noisy signal for judging whether the link work itself is working.