For most B2B teams, company-level identification such as Dealfront / Leadfeeder covers the job: it tells you which account read pricing. Person-level tools such as RB2B and Warmly add a named visitor, which helped us mainly at reply time.
Your shortlist
Editorial selectionThese products cover different parts of this job. Use the fit and buying model to decide which belong in your evaluation; display order is not a performance ranking.
Teams evaluating person-level identification of website visitors for sales follow-up.
Read profile →B2B teams using website account activity to prioritize follow-up.
Read profile →Teams evaluating visitor identification with chat or outbound engagement.
Read profile →Revenue teams combining buyer and account signals into a shared prospecting view.
Read profile →Combine account-based data, buying-group context and activation across the revenue team.
Custom pricing; quote on request
Vendor source ↗Pricing read 2026-11-12 · Currency not specifiedWebsite visitor identification software matches anonymous visits to the companies behind them, and some tools claim to name the person. We ran RB2B on a B2B software site from August 2025 to August 2026 and scored the full export. The other tools below are described from their own sites and pricing pages.
Company level or person level
Company level matches a visit to a business, mostly from IP ranges and company databases. It answers "which account is on our pricing page", which is what a team that routes on account owners needs.
Person level claims to name the individual and attach an email or LinkedIn profile. RB2B's site says its person-level matching is US-only, with company-level matching outside the US. A wrong name sends an email to the wrong person, so accuracy matters more here.
Decide which job you're buying for before you compare vendors:
| You want to | Level that worked for us |
|---|---|
| Alert a rep when a target account reads pricing | Company |
| Keep customers and open deals out of outbound | Company, matched on domain |
| See what a person read before they replied | Person |
| Build a one-time LinkedIn connection list | Person |
| Find a steady supply of new accounts | Neither, if your market is already in your CRM |
In our year of person-level data, the account and the page visited were what we acted on. A company in our market reading a page only a buyer reads was worth a rep's attention that day, and the CRM already said who owned the account. The named person helped in two places: when a prospect replied, the page history showed what they had read, and visitors with LinkedIn URLs made one small connection list. Of about 1,000 rows with a LinkedIn URL, a little over 100 survived filtering and a read by hand.
How accurate is visitor identification?
Check the company fields before anything else. More than half (55%) of the RB2B rows that carried both a work email and a website disagreed on domain. That figure comes from our own test log and overstates the error a little, because parent companies, subsidiaries and rebrands count as mismatches. Reading samples still turned up plenty of plain wrong matches, so we kept the name, LinkedIn URL, page and timestamp and rebuilt the company from the email domain. Apollo resolved nearly 9 in 10 of those domains to a company.
Two of the gaps in our data were our own doing. We hadn't put the tag on a docs subdomain buyers read, and the export went close to empty for two months in early 2026 for reasons we never traced. Put the tag wherever buyers go, and export on a schedule so a quiet month shows up fast.
How many new leads will it find?
Fewer than a filtered sample suggests. We wanted to know whether the visitor pool was worth a dedicated outbound campaign. Research agents read the sites of 76 promising visitors, companies that passed our fit filter and had reached a pricing, demo or product page, and 42 were real fits (55%). Projected onto the companies nobody had reviewed, that promised more than 200 new accounts.
The 76 had passed two filters the rest of the pool hadn't, and the pool had been worked: each company we had verified as a fit was already in our CRM. So we ran every one of the several hundred unreviewed companies through the same crawl-and-judge step we use for sourcing, which cost under $2. About two-thirds were clear misses, about a sixth had too little public information to judge, and a couple of dozen were competitors, customers or the wrong segment. A handful were new, qualified accounts: 1.2%. In our calibration the model's "no" held up and its "yes" needed a person to confirm, so even a generous share of fits among the ones it couldn't judge leaves a few dozen accounts, far short of 200.
To size your own pool:
- Export a month or more of identified visitors, all of it, not only the highlighted rows.
- Rebuild each company from its email or site domain and match against customers and open deals.
- Take a random sample of what's left, not the best-looking rows, and judge it by hand.
- Write down the filters your sample passed through, and project only onto rows that passed the same ones.
- Price the full check before you plan around the estimate. Ours cost less than two dollars and an afternoon.
Keeping customers out of outbound
About one in 25 identified companies (4%) was a customer, an open deal or already qualified. Those visits are good news for an account manager and a bad cold email. Match on domain against the CRM before any list leaves the tool, and treat a visitor whose email domain differs from the site domain as unresolved until a person looks. The lead routing guide has the precedence order we use.
Ad audiences are the other trap. A company list uploaded to LinkedIn targets the staff of each company; one of ours came to about 16 million members. The LinkedIn Ads write-up covers that.
What it costs
| Tool | Level | Price we saw |
|---|---|---|
| RB2B | Person (US), company elsewhere | $79 a month entry paid tier, July 2026 |
| Dealfront / Leadfeeder | Company | Free Lite plan, November 2026 |
| Warmly | Person, plus chat and ads agents | $900 a month, July 2026 |
| Common Room | Signal platform with web visits | $2,100 a month, July 2026 |
Put the price against a count you can measure in a trial: companies in your market that reached a buyer-only page. For us that was about 16% of identified companies.
RB2B
Person level. RB2B says it "identifies the actual people (and companies) visiting your site" and pushes records to Slack, Teams, the CRM and Clay. It fits a US-focused team that wants named visitors in Slack at a low entry price. What we saw over a year: useful same-day alerts and reply context, company fields we had to rebuild, and a handful of new qualified accounts from the pool. The RB2B review has the detail, and RB2B alternatives covers what to switch to.
Dealfront / Leadfeeder
- Company-level visitor intelligence
- Website activity context
- Audience building
Is the result an account, an individual or an inferred match?
Check product scope ↗Company level. Leadfeeder describes itself as software "that reveals the B2B companies visiting your site", built on an IP database it says it keeps current. Its site lists syncs to CRMs such as Salesforce and HubSpot, ad audiences for Google and LinkedIn, and Slack. It fits teams that route on accounts and owners, and coverage outside the US is part of its pitch. The free Lite plan shows recently identified companies, which is enough to see whether your traffic matches. RB2B vs Leadfeeder compares the two.
Warmly
WarmlyPublished capabilities- Website visitor identification
- Account activity context
- AI chat
Which identification and engagement features are in the contracted product?
Check product scope ↗Person level with engagement on top. Warmly says it "de-anonymizes website visitors at the person level" and then engages them through AI chat and ads, with separate agents for inbound visitors and account-level outbound. We priced it and passed, because we only needed identification. Ask which package includes identification and which adds the agents. RB2B vs Warmly has the side-by-side.
Common Room
- Buyer identification
- Intent and account context
- Signal aggregation
Can your team inspect the signals behind a score and correct a bad identity match?
Check product scope ↗A wider signal platform. Common Room says it identifies people across CRM, product, marketing and engagement signals, enriched from a contact directory it puts at 400M+, and lists web visits as one of those signals. Its homepage doesn't explain how a visit is tied to a person. It fits a team that wants community, product and web signals in one place. We passed on price.
What to check in a trial
- Install the tag on each subdomain buyers visit, including docs.
- After a few weeks, export the full pool.
- Compare the email domain with the website domain on rows that have both, and read a sample of the mismatches.
- Match the companies against customers and open deals by domain before anyone sends.
- Count companies in your market that reached a buyer-only page and put that number against the price.
- Hand-check a random sample of the raw pool before you estimate how many new accounts it holds.
Questions buyers ask
What is the best website visitor identification software?
Dealfront / Leadfeeder for company-level routing, RB2B for low-cost person-level alerts in the US, Warmly if you also want chat and ad agents. We ran only RB2B; the others are judged from their own sites. In our year with RB2B, the account and page visited were what we acted on.
Should I buy company-level or person-level visitor identification?
Company level, unless a named contact changes what your rep does next. The named person helped us mainly when someone replied and for one LinkedIn connection list.
How accurate is website visitor identification?
Check the company fields first, because that is where ours went wrong. 55% of RB2B rows with both a work email and a website disagreed on domain, so we rebuilt companies from the email domain.
Is visitor identification worth it for finding new leads?
Only if much of your market isn't in your CRM yet. Our researched sample projected more than 200 new accounts; checking every one of the several hundred unreviewed companies found a handful (1.2%).
How much does website visitor identification cost?
From free to thousands a month. We saw RB2B's entry paid tier at $79 a month, Warmly at $900 and Common Room at $2,100 in July 2026; Leadfeeder had a free Lite plan in November 2026.
Does RB2B work outside the US?
RB2B's site says person-level matching is US-only; outside the US it identifies companies. Teams selling mainly in Europe may get more from a company-level tool such as Leadfeeder.