Website Visitor Bot Free: The Engagement Data It Can't Fake

Yes. We run two, so here they are with their limits, taken from our own plan data:

• Website Traffic, free NANO tier. 6,000 page views and 2,000 unique visitors a month, up to 3 pages per visit, each visit capped at 30 seconds, and no card. Each session loads a page, waits, then loads the next URL on its list. It doesn't scroll, click or move the mouse.
• Sparky Traffic Bot, 24-hour free trial. Full access for a day. It runs real Chrome windows through a flow you script: page loads, scrolls by count or depth, clicks on internal, external or random links, waits, refreshes, and back and forward. Every session gets a new user agent and screen size, and clears its cookies and storage when it ends.

The version of Website Traffic that works on the page, Realistic Traffic, has no free tier. Its visits scroll down in short steps, move the mouse on a curve, highlight a passage and click a link, and it starts at $29.99 a month for 60,000 page views, exactly three times the Website Traffic price for the same volume.

Free or paid, ours or anyone else's, a visitor bot does one job well: showing you how visits register in your own analytics. It can't make a site look engaged to someone who checks, and the reason is structural, not a question of how clever the bot is.

A visitor bot controls one thing: a browser tab, and what that tab does while your page is open in it. Your GA4 tag runs in the same tab. So every engagement number GA4 builds from tag events (sessions, engaged sessions, scrolls, clicks, site searches, key events) is written by whatever is driving the tab, a person or a script.

Other engagement data is written somewhere else. An order is written by your payment processor, a reply by a person into your inbox, a Search Console click by Google Search itself. A bot can't reach those systems through your page, so they only move when something real happened.

That leaves one question to ask of any engagement metric: which system writes it? If the answer is only your tag, a bot can produce it. If a second system has to agree, a bot can't.

The top half is where every visitor bot does its work, ours included. The bottom half is what anyone doing due diligence on a site can ask for, and it's where the rest of this guide spends its time. If you're still checking whether a free bot's visits reach GA4 at all, start with what GA4 filters before you see it.

This is the gap that turns any visitor bot into a conversion counter, and Google's own help describes how to build it. For the case where "you want to register a key event whenever a user lands on a confirmation page", it suggests creating a custom event whenever page_view fires with a page_location matching your confirmation URL, then marking that event as a key event (Google).

That rule counts arrivals at a URL, not signups. Anything that loads the confirmation page trips it: a bookmark, a refreshed tab, a crawler or a visitor bot. And because a key event is one of GA4's three routes to an engaged session, the same page load makes the session engaged as well.

Our Website Traffic engine only visits the URLs you enter, up to three per campaign, and never clicks, so it trips a rule like this only if you give it the confirmation URL. A bot that follows links or reads your sitemap can find the page on its own.

Three fixes, strongest first:

1. Fire the key event on the action. Tie it to a successful form submission, or to a purchase that carries a transaction ID, not to a page view.
2. Keep confirmation and receipt URLs out of every campaign, and load your receipt page once with no order behind it to see what your tags send.
3. Reconcile weekly. Compare the key event count with the system that ought to agree: the CRM for leads, the payment processor for orders. A gap means a tracking fault or automated visits, and both are worth finding.

Each of these needs a system outside the tab to agree, so each one moves only when something real happened.

1. An order that settles. A purchase event is a tag event. The money is recorded by your payment processor, and no page load puts it there. Our own Website Traffic FAQ says it plainly: "It does not make purchases."
2. A lead who replies. A form submission is an event. A reply to your follow-up, or a call that actually takes place, is written by a person into your inbox or CRM. None of our three engines has a typing step (Sparky's flow steps are page loads, scrolls, clicks, waits, refreshes, and back and forward), so none of them fills in a form at all.
3. A newsletter click. A bot can load https://www.sparkcliks.com/example-offer/?utm_source=newsletter&utm_medium=email, and GA4 will credit your newsletter with the session. Your email platform only counts clicks on the tracked links it sent to subscribers, so its report won't move.
4. A signed-in user. User-ID "lets you associate your own identifiers with individual users so you can connect their behavior across different sessions and on various devices and platforms" (Google). A bot has no account with you, so none of its sessions carries one. Sparky also clears cookies and storage after every session, so GA4 sees each of its sessions as a new user. How to spot bought traffic in analytics explains what that does to your new versus returning split.
5. A Search Console click. Search Console counts "How often someone clicked a link from Google to your site" (Google). A bot that loads your URL clicked nothing in Google Search, whatever referrer it sends. A bot that runs a search first is a different case (Sparky has a keyword search entry), and whether a search engine counts that click is the search engine's decision. If real searchers' clicks are what you're after, that's our SERP Clicks, where paid people run the search and click your listing: desktop only, no free tier, starting with LITE at $47 for 120 clicks a month (60 if geo-targeted, at 2 credits a click).
6. Age, gender and interests. Google says that with Google signals on, "reporting data is enriched for users who have turned on Ads Personalization and signed into their Google accounts, such as demographic and interest data" (Google). A fresh, signed-out automated profile has nothing to contribute through that route.
7. Chrome's field data. The Chrome UX Report only collects from people who enable usage statistic reporting, sync their browser history and have no sync passphrase set (CrUX methodology). A throwaway automated profile does none of that. What a free traffic bot does to your logs, bills and reports covers the indirect route, where bot load slows the pages real visitors report on.

The check that separates the two halves of the table takes a week and needs nothing from any vendor. Every figure below is illustrative, invented to show how the reading works. None of it is SparkCliks data or a measurement from a real site.

Setup. A site that normally gets around 300 sessions a day runs a free NANO campaign on two pages for seven days, so every delivered visit views two pages. Nothing else ships that week: no redesign, no email blast, no tag changes.

Pull the same numbers for the campaign week and the week before:

1. GA4. Explore, then Free form, with Sessions, Engaged sessions, Engagement rate, Events per session and Key events as values. Add a second tab filtered to Event name exactly matching scroll, and divide its Event count by Sessions.
2. Search Console. Performance, then Search results, then the date filter, then Compare, with the two weeks side by side. Read Clicks.
3. Your payment processor. Successful charges, by day.
4. Your CRM or inbox. New leads with at least one reply.
5. Your email platform. Link clicks across every send that week.

Reading the top half. Sessions rose by 460, and filtering to the two campaign pages puts about 450 of them there. Each of those passed the two-page rule, which is why engagement rate climbed seven points. In the same week, scroll events per session and events per session both fell, because the new sessions were engaged without scrolling or doing anything beyond loading two pages. Engagement up while event depth goes down is the pattern to notice: sessions that clear the engaged bar while doing less than your average reader.

Reading the bottom half. Every second-system row moved by a handful, well inside what an ordinary week does. That's the signature of any visitor bot, ours included: what your tag writes moves, and what a second system writes doesn't. Key events stayed flat here because they fire on a form submission. Had they jumped along with sessions, the first thing to check would be whether one fires on a page view.

If your numbers swing more than this from week to week, compare two weeks on each side. The reading doesn't change.

Nobody can honestly promise that, and there's a structural reason to doubt it. GA4 engagement rate is calculated inside your own property from settings you control: raise the engaged-session timer and it drops, mark another event as a key event and it can climb, with no visitor behaving any differently. Google's John Mueller has said, "We don't use Google Analytics in Search" (Search Engine Roundtable), and bounce rate, which GA4 defines as the share of sessions that weren't engaged, is not a ranking factor either. Search engines also publicly deny using click data as a direct ranking factor, a record weighed in click data as a ranking signal. What a visitor bot can honestly deliver is visits you can see in your own reports.

Run one if:

• You want to see how automated visits register in your own GA4 property, and how your engaged-session timer treats them
• You want to prove that a comparison or segment built on the campaign pages strips them out of the reports your team reads
• The pages you'd target carry no ads, no affiliate links and no running A/B test

Skip it if:

• You need engagement that holds up when someone checks a second system, whether that's a site buyer, a sponsor paying per lead or your own finance team
• Your key events fire on page views and you can't change that before the campaign starts
• The pages carry ads. Automated traffic counted as ad impressions is invalid traffic under every major ad network's rules, and the penalty lands on your account, not the bot's
• You're hoping for rankings or positions, which no traffic tool can promise

If you land in the first list, the Website Traffic free tier and the Sparky Traffic Bot trial are two free ways to run that test. Point either at pages nobody's decisions depend on, then read the top half of your table against the bottom half.