Governance and responsibility
You can measure a site without tracking a person
A tracker tells you what people did on your site. The search engine tells you what the internet thinks you are.
We needed numbers. We did not want to track anybody. Those two sentences are supposed to be in conflict, and they are not. Here is where our numbers come from, what they told us, and the one thing they told us that we did not expect at all.
The thing everybody assumes
Measuring a website is assumed to mean the same thing everywhere: you put a script on your pages, the script writes something to the visitor's browser, and a dashboard somewhere fills up. Because that script stores something on a device that is not yours, you then owe the visitor a banner, a choice, and a record of that choice.
The banner exists because of what the script does to the visitor. It does not exist because measuring is forbidden. That distinction is the entire post.
If the measurement never touches the visitor's device, there is nothing to ask permission for, because nothing is being done to them.
Where the numbers actually live
Every time a search engine showed one of our pages to somebody, it wrote that down. Not us. It did. It knows the query, the position it put us in, whether the person clicked, and which page they were offered.
That record already exists, and the search engine will hand us the aggregate for free. Google calls it Search Console. Bing calls it Webmaster Tools and will import the setup from the other one rather than make you do it twice.
The part that matters for this post is how you prove the site is yours. You do not install anything on it. You put a text record in the domain's own settings, which is a thing only the owner of the domain can do. Nothing is added to a page. No script is served. Nothing is written to anybody's browser.
So there is no cookie, there is no consent to collect, and there is no banner, because there is nothing a visitor could consent to.
What it tells you, and it is not what a tracker tells you
This is the part people underestimate, because they assume the free thing must be the weaker thing.
- <b>The actual words people typed.</b> Not categories, not interests. The query.
- <b>Impressions.</b> How many times a page was put in front of somebody at all.
- <b>Clicks, and the position you were shown at.</b> Which together tell you whether the problem is that nobody sees you, or that everybody sees you and keeps scrolling.
- <b>Who links to you.</b>
- <b>Which of your pages are indexed, and the reason the rest are not.</b>
Why our numbers start three weeks ago on a domain we have had for a year
This needs saying before any number in this post, because without it every figure below reads as the record of a brand new company, and it is not.
We have had this domain for about a year. What we did not have for most of that year was a site a search engine could read.
The site was built inside a website builder. To a person it looked like a site. To a crawler it was close to empty, because the things a crawler is actually given, the title and the description and the rest of it, were written into the page by JavaScript after the page had already been handed over. The crawler was handed the version without them.
So we took the pieces that were worth keeping, threw the rest away, and rebuilt the site on infrastructure we own and can see into. Only then did anything begin to be recorded.
Which means the measurement in this post does not begin when the company began. It begins the day the site became readable, and between those two dates sits roughly a year of a domain existing and saying nothing.
What it told us, which we did not expect
Over three months this site surfaced for five queries and got zero clicks on every one of them.
Three of those five queries were, translated, Airtable consulting Vienna, Airtable specialist Vienna and Airtable adviser Vienna.
We do not sell Airtable consulting.
The other two were people typing the company name wrong, one of them badly enough to be unrecognisable to anybody but us.
Zero clicks on five queries in three months is not a failure. It is what a site looks like before it has anything ranking, and the site had been readable for three weeks. The useful signal is not the clicks. The useful signal is which words it surfaces for at all, and the answer was a product we do not consult on, in the city we happen to be in.
No analytics script would ever have told us that, because an analytics script only starts paying attention once somebody has already arrived. This told us about the people who did not.
A tracker tells you what people did on your site. The search engine tells you what the internet thinks you are.
The other half is about the crawler, not the person
The same place tells you something no visitor could ever tell you, which is what the search engine made of the site itself.
On the ninth of October ours read: nine pages indexed, twenty eight not, and the indexed number had gone backwards from thirteen. Twenty of the twenty eight sat under one heading, crawled and currently not indexed, which on a site this young is not a technical fault. It is a trust problem, and the honest translation is: we looked, and we did not think it was worth keeping.
Underneath that were two genuine broken links and one page that returns a normal response while being empty of anything worth having, which is worse than a broken link because nothing reports it.
None of that is about a visitor. All of it is about whether there will be any.
The delay, and why it does not matter
When we looked, the data was sixty four and a half hours old. Three days behind.
Real time numbers are a product feature. They are almost never a need. A decision made on three days of settled, aggregated data is a better decision than one made while watching a counter move, because a counter moving teaches you to react to noise.
And the history runs sixteen months, which is long enough to see a season, which is the only timescale on which any of this is actually legible.
What we do count on our own pages, and what we do not
It would be easy to end here and claim we count nothing. We are not going to, because it would not be true, and a company that claims to count nothing while counting something has destroyed the only thing this post is worth.
This blog counts one thing. How many people pressed the heart on an article. A number going up. No cookie, no identifier, nothing about who.
Our patient blog counts one more. How many times a page was opened, by day. No cookie, no profile, nothing recording what somebody read or how long they stayed. When we publish a link ourselves we can add a label to it, so we can tell which of our own placements sent people, and that label describes our link, not the reader.
Neither of those identifies anybody, neither writes to a visitor's device, and both are written down on the privacy page in those words rather than behind a promise that we track nothing.
The rule we work to is simple. Count things, not people.
Why this is commercial and not only ethical
A consent banner is not free. It is a legal surface you have to maintain, a record you have to keep, and a thing standing between a reader and the first sentence of your article. A proportion of people leave at it, and the ones who stay have been told, before they read a word, that something is about to be done to them.
Not needing one is not a sacrifice made for principle. It is one fewer thing to maintain, one fewer thing to get wrong in an audit, and one fewer obstacle in front of the writing.
And the measurement you get instead is about intent, which is the thing you can actually act on. Knowing that somebody scrolled sixty per cent of a page tells you very little. Knowing the exact sentence they typed before they found you tells you what to write next.
Not needing a consent banner is not a sacrifice you make for your principles. It is a capability, and it is cheaper than the alternative.
MK!
Maria Catalina Kovacs
There is no tracking on this site. No analytics, no cookies, nothing recording where you went or how long you stayed.
But I would like to know if something here was worth your time. If it was, press the heart. It tells me nothing about you. It is a number going up, and it is what tells me to keep writing.
If you want more of this, I am on X. @Queendemetriak