Is the Link You Bought Crawlable? href, JS and Image Traps

A bought link can look perfect on the page and still fail the one test that matters to a crawler: is it an a element with an href attribute? Google's link documentation says it can generally crawl a link only if it is an a element with an href. Everything else, from clickable cards to buttons to logos, depends on markup you cannot judge by eye. The short rule: confirm a and href, confirm that a JavaScript-inserted link exists in the rendered page, and confirm that an image link carries alt text that works as its anchor.

Crawlability is a precondition, not a promise. No Google page says that a crawlable link will move rankings. What the documents do give you is a checkable technical standard, and a checkable standard is exactly what a purchase order needs.

Start from the markup, not the screenshot

What Google's page actually says

The "Link best practices" (Opens in a new tab) page states that Google can generally crawl a link only if it is an a element with an href attribute, and that links without one may not be extracted. The page then lists examples. An a element with a valid href and an onclick handler sits among the recommended ones. Three forms are listed as not recommended, though Google may still try to parse them: an a element with routerLink instead of href, a span carrying an href, and an a element that navigates only through onclick.

Read the hedge carefully. "May still try to parse" is not "will count". If you accept a delivery in one of those forms, you are accepting uncertainty that the documentation itself declines to remove.

Crawlable is not the same as valuable

A crawlable link is only a discoverable link. How much a link contributes depends on other things: the quality of the page it sits on, its rel attribute and its surroundings. The documentation gives no figure for what a crawled link passes on. What you control is the technical checklist: element, attribute, resolvable address, anchor or alt text, rel value.

Three questions that people mix up

Is there a link? Is it crawlable? Is it the link I ordered? A seller who says the link is on the page is answering the first question. A buyer who says it does not work is asking the second. A dispute over a different anchor is the third. Separate them in your notes, because each has its own evidence and its own fix.

The two-minute test in browser developer tools

Reading the element

Practical suggestion: right-click the link and choose Inspect. Check that the tag is a, that the href equals the target address in your order, and that the visible text matches the anchor you requested. Menu names vary by browser and language. Viewing the page source and inspecting the live page are different views, which matters in the JavaScript section below.

Which forms pass

The table translates the Link best practices examples into a buyer's verdict. Forms the page does not show are marked "no example in the documentation" rather than guessed.

The address must resolve

The same page requires that the URL in the a element resolves into an actual web address Google can send requests to. Addresses such as /products or /products.php?id=123 are recommended; javascript:goTo('products') and javascript:window.location.href='/products' are not, although Google may still attempt them. A simple buyer's check: paste the href into a fresh browser tab. If it lands on your target page, the first condition holds. Parameters, fragments and letter case in the target are a separate review, written into the order on their own.

Empty anchors

The page defines anchor text as the visible text between the a tags and labels an empty link as bad. If the a element is empty, Google can fall back on the title attribute. For a buyer, an empty box where a readable anchor should be is two defects at once: the link may be fine, but the anchor is missing. Report them separately so a fix for one does not hide the other.

JavaScript: where the link lives before and after rendering

Raw HTML versus rendered page

Some publishers insert links after load with JavaScript. The documentation is clear that dynamically inserted links are crawlable as long as they use the markup shown for normal links. For anchor text inserted that way, it suggests verifying with the "URL Inspection Tool" that the text is present in the rendered HTML. If your interface is not in English, the menu names can differ.

Two consequences follow. JavaScript insertion is not itself a defect; a missing a and href is. And a link that is invisible in view-source but present in Inspect is not necessarily broken, because the first shows what the server sent and the second shows what the browser built.

The last row has a boundary: you can inspect only properties you own. On a publisher's domain you will rely on Inspect and a screen recording.

A link that disappears later

A script-inserted link can vanish when the publisher updates a theme or plugin. The documentation says nothing about this; it is an operational risk. Practical suggestion: store the a element copied from the rendered page together with your screenshot, so you can show what was delivered.

Image links: the alt attribute as anchor

When the link wraps an image there is no visible anchor text. According to the Link best practices page, Google uses the alt attribute of the img element as the anchor text. Empty link text combined with an empty alt is the bad example. The image guidelines say the same from the other side: alt text is useful as anchor text if you decide to use an image as a link, and stuffing keywords into alt attributes can make a site look like spam.

So a bought logo or banner link with a missing alt carries no anchor information. Whether you should accept image links at all is a separate decision. The minimum rule is that if you accept one, the alt wording is written in the order. For general alt text practice, see article image alt text and stock image licensing.

Paid placements: rel sits on top of crawlability

Crawlability does not settle the rel attribute. Google's "Qualify outbound links" (Opens in a new tab) page tells site owners to mark advertisements and paid placements with the sponsored value, says nofollow is still acceptable for them though sponsored is preferred, and notes that links marked with these values are generally not followed while the linked pages may still be crawled through other sources such as sitemaps or other sites' links.

A link can therefore be perfectly crawlable and still carry sponsored, which is consistent with a disclosed paid placement. Whether that fits your goal depends on what you bought and what the contract says. The full comparison is in dofollow, nofollow, sponsored and ugc compared, and anchor choice is covered in anchor text strategy for link profiles.

Turning this into order terms

Decision rule

Run these checks in order and stop at the first failure, then ask for a written correction:

1. Is the tag an a element with an href attribute?
2. Does the href equal the target address in the order, and does it resolve to a real web address?
3. If a script inserts the link, does the rendered page show an a element with href?
4. If it is an image link, is the alt attribute filled and does it match the order?
5. Does the rel value match the contract?
6. Does the visible anchor match the order?

Delivery record checklist

• Exact URL of the publishing page and the date of the check
• The a element copied from the rendered page
• How you checked: inspect, view source, or another tool
• Screenshot with date
• rel value and the value the order specified
• For an image link, the alt value
• Status: accepted, awaiting correction, rejected

The record format in backlink package delivery tracking fits this list, and the service request form guide shows how to write the whole scope down.

Clauses you can adapt

Practical suggestion; these are not contract templates, and each seller needs to agree to them separately:

• The link is delivered as an a element with an href attribute.
• The href equals the target URL written in the order.
• If the link is inserted by script, it exists as a and href in the rendered page.
• On an image link, the alt attribute carries the wording written in the order.
• The rel value matches the order.
• If these properties change after delivery, the seller corrects them after a written request.

Example scenario: an agency buying for a software client

Example scenario: An agency orders a sponsored article for a project management software client on a niche business blog. The delivery email shows a URL and a screenshot with an underlined phrase. The account manager is about to mark it done, but a colleague inspects the element and finds a span with a click handler. There is no href.

The agency sends a written request: deliver the phrase as an a element with the href from the order, within an agreed number of days. The publisher replaces the span, the colleague re-inspects, saves the copied element and the date, and only then closes the order. The agency adds the six questions to its delivery template, so the next vendor receives them before work starts.

Five questions to ask before ordering

The cheapest fix is the one made before delivery. Practical suggestions:

1. Will the link be in the article text, a shared area, or an image?
2. Is it written into static HTML or added by a component or script?
3. Will it be an a element with an href, confirmed in writing?
4. What will the rel value be, and why?
5. If the template changes after delivery, will the markup be preserved, and for how long?

The documentation has no answer to the fifth question; any duration exists only if the seller agrees to it in the order. A seller who cannot answer the first four clearly is leaving uncertainty to be settled after delivery. That is an observation, not a rule from Google.

Placement matters for where defects appear

In-text links

For a link inside an article paragraph, the expected markup is the simplest: an a element in the paragraph, the target in the href, the requested anchor as the visible text. Defects to look for are an editor replacing the address while pasting, or splitting the anchor across two elements.

Sidebars, footers and shared areas

Shared areas can be generated by a template component. If the component runs on script, view-source may not show the link. In that case the rendered copy matters more, and a template change could alter the link on every page at once.

Banners and logo strips

Practical suggestion: if the banner is an image wrapped in an a element, you need the a, the img and a filled alt. A scrolling logo strip may be built by a library that does not create a real a element for each logo. Inspect the element that wraps your logo. A shared click handler on the strip does not make your link crawlable.

Misreadings that cost buyers

It is clickable, so it works

Clickable is browser behavior. Crawlable is a markup question. A card can open the right page and still contain no a element, which you see only in Inspect.

The address is shown, so it is a link

A URL typed as plain text may not be a link at all. Click it, then inspect it.

It was right once, so it stays right

Templates, plugins and themes change. The documentation gives no duration commitment; a time frame exists only if the contract states one. Practical suggestion: recheck monthly with the same six questions.

It is crawlable, so it is indexed

Crawling does not promise indexing. For checking whether the publishing page is indexed, see how to check that a backlink page is indexed.

Writing the correction request

When a check fails, the request should be short, measurable and tied to your record. Practical suggestion for the skeleton:

1. Order number and the full URL of the publishing page.
2. What you see: for example, "the link is a span element with no href", or "the a element exists but the href is empty".
3. What the order says: "delivered as an a element with an href".
4. The correction requested: replace the markup with an a element carrying the ordered href.
5. A deadline and a date for the recheck.
6. The screenshot and the copied element.

Add the seller's reply to the record. If the seller says the link already works, do not argue from the documentation alone. Google's page says unrecommended forms may be parsed, but it also lists the recommended form separately. If your order named the recommended form, the dispute is about your written clause, not about whether a lucky parse is possible.

If the seller refuses

Check whether the contract states acceptance and correction terms. If it does not, tie your request to the delivery condition in the order and keep the record. Beyond that point the outcome depends on the commercial agreement; Google's documents have nothing to say about it. Read the seller's correction and refund terms before you place the order, because they differ from one platform to another.

A short drill: three deliveries, three decisions

These cases are invented for practice.

• Inspect shows an a element, an href equal to the ordered address and the ordered anchor text: accept.
• Inspect shows a span, and the click behavior lives in an onclick handler: request a written correction and hold the delivery as pending.
• Inspect shows an a element whose href begins with javascript: request a correction, because the value is not a real web address.

A fourth case exists: the page has no link at all. There is no markup dispute then; the delivery has not happened, and the process returns to the missing-delivery flow in delivery tracking.

Checks in the first week after delivery

The delivery-day check is your first record. Schedule a second and a third. On the second, open the same URL again, inspect the a element and compare it to the first record. If anything differs, write down exactly what: a changed href, a changed anchor, an added rel value, a missing element. On the third, confirm that the page is still reachable. If it has moved or been removed, the question stops being about markup and becomes a question about the fate of the publishing page.

The point of the schedule is to catch a change early. In a conversation with the seller, "it was correct on delivery day" is only a claim without a record. With a dated record you can show what you saw and narrow down when the change happened, which helps both sides resolve it quickly.

Quick reference

Reporting for agencies

When several people check deliveries, one record format keeps vendors comparable. A one-line record is enough: publishing URL, check date, a and href result, script status, alt value for image links, rel value, decision. A client report that quotes these fields is stronger than the sentence "the link is live", because in a dispute it gives both sides something measurable. A single small business can keep the same fields in a spreadsheet. If the publishing page itself later moves or disappears, that is a separate process outside this checklist.

Google's documentation can change, so check the current pages before relying on any detail.

Frequently Asked Questions

If a link is clickable on the page, is it valid for Google too?

Not necessarily. The Link best practices page says Google can generally crawl a link only if it is an a element with an href attribute. An element that navigates only through onclick is listed as not recommended, and although Google may try to parse it, the page gives no guarantee.

Is it a problem if an a element has both href and onclick?

No. An a element with a valid href and an onclick handler is among the recommended examples. The problem is a missing href, not the presence of onclick.

Does a JavaScript-inserted link count?

According to Google's documentation, links added dynamically are crawlable as long as they use the recommended a and href markup. The page suggests using the URL Inspection Tool to confirm that anchor text inserted this way is present in the rendered HTML.

Does a banner link have anchor text?

For image links, Google uses the alt attribute of the img element as the anchor text. If both the link text and the alt are empty, the documentation marks that as a bad example, so the alt wording belongs in the order.

Can a crawlable paid link still carry sponsored?

Yes. Google's Qualify outbound links page asks site owners to mark advertisements and paid placements with sponsored, and says nofollow is still acceptable although sponsored is preferred. Crawlability and rel are separate checks.

Does a crawlable link raise rankings?

No Google page promises that. Crawlability is a precondition for the link to be discovered, not a guarantee of any outcome.

What should I keep as a delivery record?

Keep the publishing URL, the date, the a element copied from the rendered page, the rel value, the alt value for image links and a screenshot. That record is your evidence in a correction request.

Conclusion

The technical quality of a bought link lives in its markup, not its appearance. Google's Link best practices page says a link is generally crawled only when it is an a element with an href, that script-inserted links are crawlable when they use the same markup, and that an image link's alt attribute works as its anchor. For paid placements, the Qualify outbound links page prefers sponsored. None of this promises rankings, but all of it can be checked and written into an order.

Move the six-step decision rule into your order terms, save the a element from the rendered page at delivery and ask for a written correction at the first failed step. If you want a package with its scope and conditions stated up front, see the backlink packages on Akua Adforce; the scope and terms are described on the product page.