A study of cookie consent in Spanish online stores

Who is behind it

Cristhian Velásquez, measurement and MarTech analyst based in Madrid. Contact: hola@cristhianvelasquez.com.

What it does

I am preparing a study on how Spanish online stores respect cookie consent. The robot loads the home page, checks whether there is a cookie banner, clicks that banner's accept and reject buttons, and looks at whether tracking data is still sent after rejecting.

What it does not do

• It does not fill in forms, log in or buy anything.
• It does not visit any page other than the home page.
• It does not try to bypass anti-bot protection: if your server blocks it, that is noted and it does not try again.
• It does not pretend to be a browser: it identifies itself by name and with this page.
• It reads your robots.txt before visiting anything.

How often it visits

Once per site, with up to three loads of the home page (the normal visit plus a clean session to test the reject button). Sites are visited spaced apart, at a very low rate. Because the page is loaded with a browser, the visit may show up in your analytics as one more visit; the robot name in the user-agent lets you tell it apart.

Updated on 3 November 2026: during the method-testing phase, the first stores examined (about 50) may receive a second visit to check an improvement in banner detection. In the final study, one visit per site.

Updated on 5 November 2026: before studying a site, the robot checks whether it is an online store with one request to your robots.txt and one to the home page, without a browser. That check is made on quite a few more sites than are finally studied. For sites that are not stores, only the domain is kept, in a private file, so they are not visited again; they are not studied or published.

What is stored and what is published

Only technical results are stored (whether there is a banner, which cookie manager it is, whether rejecting works). No personal data is stored. Results are published in aggregate, as percentages of the total number of sites. No companies or sites are named.

Updated on 6 November 2026: to check the robot's results, some of the sites (about 80) were also reviewed by hand, by one person, with a regular browser and a single load of the home page. The domains examined and each site's results are kept in a private file and are not published; they are kept for up to 12 months after the study is published and then deleted. The aggregate results are in the published study.

How to ask it not to visit your site

Add this to your robots.txt:

User-agent: MarTechConsentStudy
Disallow: /

Or write to hola@cristhianvelasquez.com and I will exclude you.

Full user-agent

MarTechConsentStudy/1.0 (+https://cristhianvelasquez.com/es/estudio-consentimiento/; hola@cristhianvelasquez.com)