Where things actually stand.

Joachim asked a fair question after our call: if the links are working, why does search traffic look flat year on year, and why is so much of the growth branded or from the new car seat towels range? Rather than defend our position we went and checked properly.

Everything below comes from your Google Search Console and your Shopify, not from estimates. Some of it is uncomfortable reading for us. Some of it points at a real opportunity that has been sitting in front of us the whole time.

They are not doing much, and we would rather say so.

Since November we have built 273 links across 211 domains. Most of them sit on free publishing and bookmarking platforms, places like Medium, Notion, Diigo, Tumblr and Blogger, where anyone can create a page. The platform itself has a high domain rating, so on a report the link looks strong. The individual page carries very little weight.

We first checked those platforms against Ahrefs and none of them registered. Ahrefs is not Google though, so on its own that proves little. So we went to the source that does count: the linking domains report inside your own Search Console, which is Google telling you directly which sites it credits as linking to you.

Google credits your site with 134 external links from 47 domains. We took our own link building log, every one of the 273 links across all eight months, and checked all 211 source domains against that list one by one. Nine of them appear, accounting for 19 of the 134 links.

So of 273 links built over eight months, Google credits 19. Roughly four in every hundred domains we used registered at all. The rest of your link profile was there already or arrived on its own.

What this actually tells us about link building in 2026

It is worth being precise about what the 134 links are. They are every link Google credits to your site from every source since it existed, not our work. Most of them nobody built on purpose. They split roughly like this:

Splitting our own log by the type of link makes the point sharper. The twenty business listings we set up in November, the foundational directory work, registered at roughly four times the rate of everything else we built:

The wider point stands though. Directory and profile listings still register with Google, so foundational links remain worth doing, and they are a one-off job rather than a monthly one. The free-platform links that made up the bulk of our monthly work have stopped counting for anything. Our own team's outreach backs this up: around 100 approaches to relevant blogs for editorial placements returned zero free placements, and almost every one asked for payment. So the realistic options now are foundational links once, genuine PR and relationships, or paid placements. There is no longer a meaningful free middle.

That is a change in the market rather than something unique to your site, but we should have caught it sooner and told you rather than kept invoicing the old approach.

To be fair to the data: Google's report is not a complete map of the web either, and we cannot prove with certainty these links did nothing at all. What we can say is that neither Google nor any tool we have shows them, and there is no pattern in your numbers that suggests they moved anything.

The site is in a much better position than it was.

Comparing April to July this year against the same months last year, with brand and car seat towels removed, here is where your search visibility actually sits:

Your total impressions are down about 40% over that period, and on its own that looks alarming. It is not. Almost all of the drop sits at positions 21 and beyond, which between them produced 37 clicks in the whole of last year. You have stopped being shown on page four for things that were never going to sell a towel, and started being shown on page one for things that will.

Your Shopify agrees at the money end: organic revenue is up 28.9% since kickoff, $30,075 to $38,755, and organic is now your single biggest channel.

Traffic is not down, but it has flattened.

On like-for-like months, total clicks are up 8%, from 2,157 to 2,330. So traffic is not down. But we checked Joachim's other point properly, and he is right about that too. Of the 173 extra clicks, roughly 104 are car seat towels and 40 are branded. About 83% of the growth is your new range and your brand. Everything else contributed around 29 clicks.

So on clicks alone, nine months of work has moved very little, and it is fair of him to say so.

What the click count hides is the table in section 02. Underneath that flat number the site has moved out of the positions where nothing happens and into the ones where it does. The work built the position. It has not been converted into clicks, and the next two sections explain exactly why.

Google cannot decide which of your pages to show.

This is the finding that explains most of the rest. For your biggest search terms, Google is not choosing one page from your site. It is showing several, all in the same place on the results page, and none of them is the obvious answer.

Take "quick dry beach towel" and "quick dry towels australia", two of your most valuable searches. Over the last twelve months Google served these five different pages for them:

A tea towel page is being shown to people searching for quick dry towels. So is a beach mat page. None of them gets clicked, because none of them is what the searcher asked for.

The same thing happens on your hooded towel range, where a single product page keeps beating the collection that was built for the job:

The collection ranks four to ten places better whenever Google picks it. It just does not pick it often enough.

You are being shown far more than you are being clicked.

Every non-branded search on your site over the last twelve months, grouped by where you rank:

A 2.24% click rate at positions four to ten is roughly a third of what those positions normally return. The reason is section 04, and you can see it directly by comparing a range where one page clearly owns the search against one where five pages compete:

Your quick dry family alone is shown over 20,000 times a year at positions 8 to 10 and returns under 200 clicks. At the car seat towels click rate that same visibility would return around 1,100.

The page two pool, priced on your own numbers.

1,433 of your non-branded searches sit at positions 11 to 20. They were shown 115,384 times last year and returned 658 clicks. Your organic revenue works out at $4.09 per organic click, so:

Where it is concentrated, and none of it needs a new link:

The way we set this up was too rigid.

We want to be straight about the root of this, because it is more useful than a list of fixes.

At the start we agreed a scope of thirty keywords and the thirty landing pages that matched them. Those pages were optimised well. Then Google did what Google does and started ranking different pages instead: the homepage, the beach mats collection, a tea towel page, individual products.

The deeper issue is that the engagement was built as a fixed list to work through rather than something we re-read every month. A search programme should look at what Google actually did in the last thirty days and respond to it. If Google decides your tea towel page answers "quick dry towels australia", the job that month is to fix that, not to carry on optimising a page Google has stopped showing.

So the change we would make, on your account and generally, is this:

• Monthly, we read what actually happened. Which pages Google chose, which searches moved, where clicks are being lost, and what changed on the results page itself.
• The work for that month follows from that, rather than from a list agreed nine months earlier. Some months that is on-page, some months content, some months technical.
• The keyword list is a direction, not a contract. If Google keeps ranking a page we did not target, we either make that page good or we make the intended page clearly better. We do not ignore it.
• Reporting shows revenue and the pages driving it, with the reasoning, rather than a list of links built.

Worth tidying up. Not what is holding you back.

On the disavow

A disavow file is in place and was refreshed on 30 July. It currently covers 43 domains. Our audit turned up 321 spam domains in total, so we have handed the fuller list across and it can be merged into the existing file whenever you want it done.

Here is what we would like to do about it.

We are not going to pitch you a bigger retainer off the back of an audit that found our own link work was not landing and our own process was too rigid. So here is what we think is fair.

• The on-page work in sections 04, 05 and 06, at no charge. Deciding the owner page for each search, the internal linking, rewriting the titles and headings on the pages that are actually ranking, and building out the travel towels and hooded towels collections properly. That is the work with the clearest return in this document and we would like to do it as our own cost.
• Link building stops. Not reduced, stopped. We are not replacing it with a paid link programme unless and until you want to have that conversation separately, with the costs on the table.
• Mobile speed goes to Gil directly. Your site takes 14 to 17 seconds to load on mobile against a 2.5 second target, and that suppresses everything else in this document. Gil has quoted it as a fixed-price job. We will introduce you and step out of the middle, so you are paying a specialist directly rather than paying us to manage him.
• Three months free on our new platform. We are building an AI search platform for Shopify stores and it goes live shortly. Your category is exactly what it is built for. We would like to give you the first three months at no cost, and in return we would value your honest feedback on it.
• You keep everything either way. The backlink audit, the disavow file, the keyword research, the AI prompt library, the cluster strategy and the page-by-page fix list. All documented and yours, whichever way you decide to go.

The short version.

Our proposal in one line: we do the on-page work for free, introduce you to Gil for site speed, give you three months on the new platform, and stop the retainer.

If that sounds fair, the on-page work can start next week and you would expect to see movement on the quick dry and hooded towel searches within four to eight weeks. We would report on it once, properly, with revenue and the pages driving it, and you can decide from there whether there is anything else worth doing together.

Happy to talk through any of it. Every figure here traces back to your own Search Console or your own Shopify, and we will show you the working behind any number you want to check.