Site Analytics Available
How many? Traffic, sources, devices and geography for every site you run, plus the business events your site already pushes to a dataLayer.
Site Analytics in detail →Site SEO Available
Can a search engine read your site?
Skomi walks your site the way a search engine does, reads what every page says about itself, and reports what would stop it being indexed. It sits beside your analytics, so the faults are ordered by the traffic those pages actually get rather than alphabetically.
What you get
A report you can act on, not a spreadsheet.
A crawler that lists every fault it can think of gives you a thousand rows and no idea where to start. This one names the fault, says what to do about it, and puts the pages people actually read at the top.
The busiest broken pages first
Of four hundred pages with no description, ten of them matter and the rest can wait. Because the crawl sits beside your analytics, it knows which ten. No other SEO tool on your site knows what your traffic does.
The contradiction neither file admits
Your sitemap submits a page for indexing and your robots.txt forbids it. Each file is perfectly consistent on its own, which is why this is the fault that survives longest, and it is the one this looks for first.
Pages nothing links to
A page reachable only by typing its address is a page search engines struggle to find and visitors never will. The crawl walks your links, so it knows which pages the site itself forgot.
Titles, descriptions and headings
Missing, duplicated, or longer than a result page will show. Two pages claiming the same title is a fact about the site rather than about either page, so it takes a crawl of the whole thing to see it.
Redirect chains and dead links
A link that goes to a redirect that goes to another redirect costs you on every visit and every crawl. Anything that answers 404 or 500 is reported as what it is.
It runs itself
Once a week by default, on a schedule you set per site. There is no desktop crawler to remember to run and no export to import anywhere.
The part that is hard
Crawling somebody else's server, politely and honestly.
A crawler is a program that hits your machine a few thousand times. Skomi reads your robots.txt first and obeys it, honours a crawl delay if you set one, and never leaves the hostname you registered. A redirect off your site is followed no further than the fact that it left.
The honesty is the other half. A crawl that stopped early has seen part of a site, and two of these checks are about the whole of it. Rather than guess, the report leaves those checks out and says it did. A finding that fires on a page you have already got right is worse than one that never fires at all.
- Reads first
- robots.txt, then your sitemaps, then the pages themselves.
- Stays home
- One hostname, the one you registered. Links off it are recorded, not followed.
- Paced
- A handful of requests at a time, and one at a time if your robots.txt asks.
- Identifies itself
- A user agent naming Skomi and a page explaining what it is doing there.
Worth knowing
What it does not do.
No keyword research and no rank tracking
This reports what is wrong with your pages, not what to write on them. The queries people actually used to reach you, with their positions and impressions, come from Search Console and Bing and are read on the Analytics side.
It reads your markup, not your writing
It can tell you a page has no description, two headings where it should have one, or barely any text. It has no opinion about whether the text is good.
Your site only
There is no competitor crawling and no backlink index. Skomi crawls the sites in your own account and nothing else, and it obeys their robots.txt. This is a report on your site, not a way to study somebody else's.
- Installation
- None of its own. It uses the script that is already on the site, and it is the one product that needs no visitors: it works on a site nobody has been to yet.
- How often
- Once a week by default, set per site.
- How many pages
- A ceiling per site, so a crawl can never cost more than you expect.
- What it keeps
- One row per page, replaced by the next crawl. Findings are worked out when you read them.
The rest of the platform
Same script, same account.
Every product reads the site you have already registered. Switching one on is a toggle, not another installation.
Site Behavior Available
What did they do? Heatmaps of where people click and how far they scroll, replayable recordings of real visits, and a live view of who is on the site now.
Site Behavior in detail →Site Feedback Available
What did they say? Let visitors rate a page and say what is wrong, pinned to the exact element they were looking at and triaged in one inbox.
Site Feedback in detail →App Analytics Available
How is the app doing? Sessions, screens and retention for an app people install, with crash-free rates and errors alongside them, so a bad release is visible in the same place as a quiet one.
App Analytics in detail →Find out what the crawler finds.
Switch SEO on for a site and the first crawl starts straight away.