Know what Google and AI bots actually did on your site today.
First-party crawler intelligence for Google and AI search. See how bots access your site, where they fail, and what needs fixing.
See what is crawling your site
No tracking script. No cookies. First-party server data.
No signup.
Works with your existing infrastructure
- Built for agents
- No tracking pixel
- No JavaScript
- Every request, bots and humans
See what every crawler is doing
right now, not yesterday
Every search and AI crawler request, verified and explained.
Built by search and infrastructure practitioners who have worked with
Who uses Salience
Search and organic growth
“I deployed sitemap changes yesterday. No idea if Googlebot has seen them. Search Console won't tell me for 48 hours.”
Search crawler intelligenceAI and content teams
“Which AI systems are actually reading our content, is it training or retrieval, and are they even genuine?”
AI crawler intelligenceSecurity and platform teams
“Everyone claims to be Googlebot. Half aren't. Meanwhile the real one is getting rate-limited by my WAF.”
Bot threat intelligenceEngineering and commerce ops
“We shipped a URL change at 14:00. Googlebot hit 15,000 404s before anyone noticed.”
Site reliability monitoringAlso used forE-commerce intelligence Publisher intelligence
Connected in about ten minutes, with nothing added to the page.
Connect your source.
Cloudflare, Vercel, CloudFront, Netlify, Kinsta, Shopify, Apache, Nginx, or a raw log drop.
- Cloudflare
- Vercel
- CloudFront
- Kinsta
- Shopify
- Apache
- Nginx
Salience verifies and classifies every request.
Runs server-side, so nothing changes for your visitors.
In the dashboard in 5 to 10 seconds.
See issues and get alerted.
Errors, blocks and impersonation land in your dashboard or in Slack.
16 detectors on a 5-minute cycle.
Your analytics cannot tell you what bots actually received.
- Crawlers do not execute the JavaScript tag, so GA4 never sees them.
- A crawl simulator shows what a bot could fetch, not what the production stack served it.
- User-Agent strings are trivially spoofed, so allowlists and blocklists work from unreliable data.
- The answer is in the raw logs, and nobody has time to read them across a CDN, an origin and a platform with no raw access.
- By the time rankings drop or a ticket comes in, the crawler has already been served thousands of errors.
See recorded AI crawler activity.
Recorded requests grouped by crawler classification. See who requested your pages; requests alone do not establish training, indexing, answers or citations.
Illustrative AI Crawlers snapshot for demo-site.example, covering 24 hours. 5,307 LLM requests, 9.7% of total traffic. 1,961 requests from training and other crawlers, 3,178 from AI search crawlers, 168 user-triggered AI fetches. Recorded requests are classified by agent. They do not prove training, indexing, answers or citations. Top crawlers: Amazonbot 2,950 requests, Verified; ClaudeBot 1,095 requests, Verified; AIWebIndex 470 requests; ByteSpider 392 requests, 392 Unverified; PerplexityBot 178 requests, Verified; DuckAssistBot 93 requests.
Alerts that tell you what to do.
Not just a notification.
When Googlebot starts hitting 404s or GPTBot triples overnight, your inbox names the crawler, the URLs and the next step. Not “something happened, check your logs.” Every alert checks the site is reachable first, so you are never told it is down when it is not.
The Alerts page for demo-site.example: Know what needs you. Needs attention 2, Worth knowing 8, History 10. 15,000 404s on /products/* from Googlebot since the 14:00 deploy. Suggested next step: Roll back or fix the URL structure. ClaudeBot refused with 403 on 32 URLs under /collections since the WAF rule change. Suggested next step: Review the rule if you want those pages in AI answers.
Fits how you already work
For agents
- REST API
A personal API key reads your data on every plan, including Free.
- CLI
Install it and query the same record from a terminal or a script.
- MCP server
Connect Claude, Cursor or ChatGPT to your crawler data and ask it questions.
For your team
- Dashboard
A live view per site, with shared read-only links for clients and colleagues.
- Alerts
Email, Slack and webhooks, each naming the crawler, the URLs and the next step.
- Reports
A weekly summary per site, and CSV exports of the record behind it.
This is the internet hitting your website.
Search engines, AI crawlers, scrapers and real people arrive at once. Salience classifies each request as it lands, checks who each crawler claims to be, and shows you what needs acting on.
The Bots and Crawlers page for demo-site.example, last 24 hours: 20,562 bot requests, 37.8% of all 54,435 requests · same filters. 12,078 verified, 2,845 failed verification, 5,639 other status. By category: Search Engines 2,001, Social Media 126, AI Crawlers 5,307, Monitoring 325, Scrapers 6,706, Other Bots 6,097. Bot traffic over time in fifteen-minute buckets, stacked by category, with one bucket of 3,185 requests from a blocked probe storm.
From raw request to recommended action.
See what search engines, AI systems and people actually request from your infrastructure, then verify, explain and act on it.
Signals from search crawlers, AI bots, Search Console, Google Analytics and your own CDN and web-server logs feed the Salience engine, which produces four capabilities: Observe, Verify, Explain and Act.
Search crawlers
- GooglebotGooglebot
- BingbotBingbot
AI / LLM bots
- GPTBotGPTBot
- ClaudeBotClaudeBot
- PerplexityPerplexityBot
Platform signals
- Search ConsoleGoogle Search Console
- AnalyticsGoogle Analytics
Primary data
- CDN / server logsCDN / Web server logsPrimary source
SalienceAI and searchcrawlerintelligence
Turns raw requests into answers.
- Raw
- Normalised
- Enriched
- Understood
- Observe
See what is actually reaching the server.
Analyse the requests arriving through your CDN or web server. Server-side data, no browser script, no sampling.
- Verify
Know which bots are genuine.
Check supported search, AI and automated identities against official provider IP ranges where verification data is available.
- Explain
Understand why behaviour changed.
Connect each request to the path, status code, site section, historical pattern, sitemap, robots.txt rule or index state.
- Act
Move from evidence to action.
See the affected bots, IPs and URLs, the likely issue and a recommended next step for your team to take.
Trust & data protection
Privacy and data protectionYou are the controller
We process only on your instructions. GDPR Art. 28 DPA on every account, nothing to sign.
UK data residency
AWS eu-west-2 (London). S3, DynamoDB and Aurora are encrypted at rest. See the Security Overview for the query-cache exception.
Server-side collection
No browser tracking script and no client-side pixel.
No sale, no pooling
Your logs are never sold, never used for advertising, never shared between customers. DPA, sub-processor list and security overview available.
Crawler intelligence from $19/mo.
Priced on requests, not seats. Unlimited users on every plan except Solo. Start on the free tier and upgrade when your traffic does. Free trial, no card needed.
- 500K requests/mo
- 1 site
- 30 days history
- Unlimited users
- Real-time analytics, bot and AI detection
- 20M requests/mo
- 5 sites (+5)
- 1 year history
- Unlimited users
- Real-time analytics with bot and AI-crawler verification
- 100M requests/mo
- 15 sites (+15)
- 2 years history
- Unlimited users
- AI allowance: ~600 answers or ~60 reports a month
- 500M requests/mo
- 50 sites, no ceiling
- 4 years history
- Unlimited users
- AI allowance: ~3,000 answers or ~300 reports a month
- 1B requests/mo
- Unlimited sites
- Custom history
- Unlimited users
- SSO / SAML and audit log
Common questions.
How is Salience different from Google Analytics?
GA4 is an event analytics platform: it measures users, sessions and events collected by its tag or Measurement Protocol, and it filters known bots out by design. Salience reads server / CDN access logs, so it sees every request that reaches the site: bots, crawlers, AI agents, scrapers and traffic that never executes analytics tracking. Use GA4 for human behaviour and conversion; use Salience for what search engines and AI crawlers actually do on your site.
Does Salience replace my WAF, CDN or observability tools?
No. Salience reads the access logs those systems already produce and sits alongside them. Your WAF decides what gets through, your observability tool watches the application, and Salience tells you what search engines and AI crawlers requested, what they got back and whether they were genuine. It replaces none of them and needs nothing added to the page.
Can I import historical log files?
Yes: drag-and-drop Apache, Nginx or CloudFront logs, or an export from another log analyser. Auto-detected, deduplicated against any real-time data, merged cleanly into your dashboard. Imports don't count against your monthly request quota.
Where is my data stored, and is it GDPR-friendly?
AWS eu-west-2 (London). Encrypted at rest and in transit. We don't sell data, we don't aggregate across customers, we don't run ads. DPA available on request.
Can Salience detect AI crawlers like GPTBot and ClaudeBot?
Yes, and it verifies them against the bots' published IP ranges, so you can tell legitimate GPTBot/ClaudeBot/PerplexityBot traffic from scrapers using their User-Agent string. Per-bot crawl frequency, top URLs and verification status live on the Bots & Crawlers page.
Do I need an engineer to connect a site?
Not for Cloudflare, Vercel or Kinsta: each is a guided setup of about ten minutes or less. Apache and Nginx need someone with root or sudo for about ten minutes. Any setup can be delegated to an engineer from inside Salience without sharing the account.
Does Salience touch my origin or slow the site down?
Nothing is added to the page: no JavaScript tag, no pixel. CDN sources need no agent on the origin. Apache and Nginx use a read-only agent that reads the log file and sends it over HTTPS; it is not in the request path. Where a Worker collects the logs, on Cloudflare and Shopify, it adds a small amount of edge compute.
Are all requests ingested, or only crawlers?
Every request that reaches the source, humans and bots, with no sampling by Salience. Each one is classified human or bot, and 200+ crawler identities are recognised. CloudFront real-time logs arrive at the sampling rate you configure, and
Don't see your question?
Ask Salience AI directly. The same brain as the chat bubble, answering inline.
Your logs already show what Google and the AI crawlers are doing.
Free plan, no card, about ten minutes to your first request.