An audit for a Romanian website covers everything an ordinary one covers, plus a layer automated reports do not touch at all: compliance.
The difference matters, because problems there do not show up in traffic. They show up in an inspection.
The technical layer
What crawlers see. If the site is a JavaScript application serving empty
pages, search engines see a <div> and AI sees nothing. It is the most common
structural problem and the most expensive.
Product structured data. In a market where price comparison matters, a shop without correct product data loses rich results — and the clicks that come with them.
Speed on a phone, on mobile data. Not on wi-fi. The difference between those two measurements is often threefold.
Canonicals, pagination, filters. Large Romanian shops generate thousands of URLs from filters. Without correct canonicals the index fills with variants of the same page.
The compliance layer
e-Invoicing. If you issue invoices to companies, the system has to submit them to the tax authority, not merely generate PDFs.
Consumer information. A shop selling to individuals has information duties and must display the links to alternative dispute resolution.
Company identification. Registration number, registered address, trade register details, visible and correct.
Real policies, not copied ones. Terms, privacy, returns — matching what you actually do rather than lifted from another site.
What actually gets checked
On every important page, with each crawler separately: response status, title and description, canonical, hreflang, how many words the crawler receives against what the browser sees, structured data blocks and whether they assert things present on the page, redirects, robots.txt and the per-agent rules, the sitemap and whether the URLs in it respond.
On top of that, for a shop: canonicals on filters and pagination, product data, declared availability and price against what is displayed.
The difference between this and an automated report is not the number of checks. It is that every result arrives with what it means for you and where it sits in the repair order.
What you receive
A document with both layers, each problem with its impact and its place in the repair order. Technical verification runs against real crawlers — Googlebot, Bingbot, GPTBot, ClaudeBot, PerplexityBot. The tool is public: crawlview.
The document is yours. It can be executed by the team that built the site, by someone else, or by us.
Questions
Why does the audit include compliance?
Because in the Romanian market those problems have real consequences, and no automated report checks them. A technically perfect site can carry a problem that is invisible in traffic.
Are you a Romanian company?
No, we are in Chișinău. We know the requirements because we work with clients there, but for legal questions you work with a lawyer, not a developer.
How long does it take?
A few days for an ordinary site. A shop with tens of thousands of pages takes longer, because canonical and pagination problems only become visible at scale.
What is not in the audit?
Editorial content and marketing strategy. The audit says what is technically broken and where you are not compliant, not what copy to write.