Traffic Bot Free Plans: How to Read the Feature List

Yes, and it will send visits. What no free offer gives you is capacity without a cap, so the useful question is where the cap sits. It sits in one of three places, and each shows up on a different line of the feature list:

• A monthly allowance. Permanent free tiers cap visits or page views a month, and usually cap how long each visit lasts. Both numbers sit on the plan card, which makes this the easiest cap to read.
• A trial window. Trials of paid tools give you the full feature set for a fixed time, such as 24 hours. The real cap is that window multiplied by the number of browsers the trial runs at once, and the second number is the one that tends to be missing.
• Your own computer. Freeware runs on your hardware, so its capacity is whatever your processor, memory and connection can hold. No page can tell you that number, because the vendor doesn't know it.

Whichever cap applies, four lines decide what the free visits look like once they reach your reports: concurrency, dwell time, referrer and device split. The sections below turn each one into a number or a test, starting with how rarely vendor pages give you one.

On September 10, 2026, we read the six pages ranking for this search, from five vendors selling hosted services and downloadable software. We counted how each handles the lines that decide delivery. Pages change, so treat this as a snapshot, not a verdict on any vendor.

A few other details stood out. Two of the six call their free traffic unlimited. Two make ranking claims, one of them promising that better engagement metrics "lead to better rankings", while one page, to its credit, says outright that no sales or ranking result is guaranteed. And one answers its own question, "Is It AdSense and Other PPC Programs Safe?", with "Yes!"

The pattern makes sense from the seller's side. A number can be checked, and a checked number can come up short. A feature name can't fail a test. Every line those pages leave vague, though, is a line you can put a number on yourself.

Concurrency is how many browser sessions a bot runs at the same moment. Some tools price on it directly: our Sparky Traffic Bot is sold by active browsers, from 10 on the LITE plan ($80 a month at list price) up to 1,000. Freeware inherits it from your computer. Free tiers of hosted services hide it behind a monthly visit count. It matters because each browser can hold only one visit at a time, so every other setting spends it.

A queueing result from 1961 makes that exact. Little's Law says the average number of items in a system equals the rate they arrive multiplied by the average time each one spends inside, written L = λW (Little, Operations Research, 1961). For a traffic bot the items are sessions and the system is the pool of browsers, which gives the one formula a feature list needs:

sessions per hour = concurrent browsers × 60 ÷ minutes per session

Minutes per session means everything a browser spends on one visit: loading the page, the dwell time you set, any extra pages, and the gap before the next session starts. Here's what that allows over a full day of nonstop running:

These are ceilings from arithmetic, not delivery figures from any product. Real campaigns land below them, because browsers sit idle between sessions, pages fail to load and schedules pause.

Read the table in both directions. A plan with 10 browsers can't deliver more than 2,880 five-minute visits a day, whatever else the page says. And a vendor promising 10,000 five-minute visits a day is promising at least 35 browsers running around the clock (10,000 × 5 ÷ 1,440 is about 34.7). If a page won't say how many browsers it runs, or at what daily pace a hosted plan delivers, its visit number has nothing underneath it.

No. "Unlimited visits" is true in one narrow sense: nobody meters the visits. The meter moved to the browsers. On a concurrency-priced plan, the most visits you can get is concurrent browsers × hours running × 60 ÷ minutes per session. On freeware, your hardware sets the same ceiling.

We sell one of these, so here's our own line read the same way. Every Sparky Traffic Bot plan lists "Unlimited visits", and every plan is priced on active browsers. LITE runs up to 10. At two minutes per session that's at most 300 sessions an hour, or 7,200 a day if the browsers never stop. Double the minutes per session and you halve the visits. "Unlimited" describes the billing, not the capacity, on ours and on everyone else's.

Two questions turn an unlimited claim into something you can plan around:

1. How many sessions run at once on the plan you'd buy, or on the trial you're about to start?
2. Is that limit per campaign or per account? If campaigns share one pool of browsers, "unlimited campaigns" splits the same capacity more ways. It doesn't add any.

On a spec sheet, "dwell time", "visit duration", "session duration" and "time on site" all mean one setting: how long the bot keeps a page open before it moves on or closes. Three things about that line matter more than the value you pick.

The name promises more than the setting does. Dwell time is an industry coinage, not a metric any search engine publishes, and Google has dismissed the idea in public: its Gary Illyes called dwell time and CTR theories "generally made up" in a February 2019 Reddit AMA (Search Engine Roundtable). A feature list that sells dwell time as an SEO lever is selling a claim no search engine has confirmed. What the setting changes, on the record, is your own analytics.

It spends concurrency. A browser holding a page for three minutes can't start another visit for three minutes. On a fixed number of browsers, longer visits mean proportionally fewer of them, which is the formula from the concurrency section at work.

A single number is a tell. A range is a spec. A bot that holds every page for exactly 60 seconds draws a spike in your duration report that real visitors don't produce. Ask whether the line takes a minimum and a maximum. Ours do: Sparky sets a minimum and maximum time on page and a minimum and maximum wait between actions, and paid Website Traffic plans randomize time on page across up to three pages.

On free tiers this line is often the tightest cap on the page. Our free NANO tier stops each visit at 30 seconds, against up to 5 minutes on paid Website Traffic plans. For choosing values that resemble your real audience, see how to set visit duration and pages per session.

A referrer line ("Custom referrers", "968 different types of referrals", "Direct / Organic / Social / Referral") can describe three different mechanisms, and none of the six pages we read explained which one it uses. Here's where each lands in Google Analytics 4 (GA4) and in Search Console:

GA4's default channel group puts a session in Organic Search when its source is on Google's list of search sites or its medium is exactly "organic" (Google Analytics Help), so the first and third rows look identical in your acquisition report. Search Console is where they part: it counts a click when someone clicks your listing on a results page (Search Console Help), so a typed referrer never shows up there.

The third row is the only one with a search behind it, and it's an automated search. Google's spam policies define machine-generated traffic as "the practice of sending automated queries to Google" (Google Search Central). That definition covers any search-entry feature pointed at Google Search, ours included: Sparky offers keyword search entry alongside a referrer you set and direct navigation. Know which mode a campaign runs before you read its results.

Check whether a free tier offers this line at all. On our NANO tier it doesn't: referral, organic and social traffic types are paid features, so you can't choose the channel free visits arrive in. How the Referer header itself works, and why browsers strip it, is covered in referrer control explained.

A device line ("Device simulation", "desktop, mobile, and mixed") describes a browser profile, not a handset. The bot picks a user agent string and a screen size, and your analytics reports what those say. In GA4, Device category is "the type of device: Desktop, Tablet, or Mobile" and Screen resolution is "the screen resolution of the user's monitor" (GA4 API schema). No phone has to be involved for a session to count as mobile.

Four of the six pages named devices, and none gave a ratio. Three questions fill the gap:

1. Mode or ratio? "Mixed" can mean a fixed split, a random draw for each session or a default nobody documented. Ask which.
2. Per campaign or per session? A campaign-level setting sends all of one device unless it offers a mixed mode.
3. Can you see it during the free period? Your server log shows each visit's user agent as it arrives, and phone browsers usually include the word Mobile in theirs. GA4's Tech details report catches up once the data processes.

Ours don't escape the first question. Sparky and Website Traffic both offer desktop, mobile, tablet or mixed, and neither page states a fixed ratio for mixed, so count it in your log before you rely on it. Choosing the split itself, from your own GA4 and Search Console data, is covered in desktop, mobile or mixed.

Here's the whole method on one invented spec sheet. It isn't a real product, and every result below is made up to show how the reading works:

Free 24-hour trial. Unlimited visits. 5 browsers at once. Visit duration 60 to 180 seconds. Custom referrer. Mobile, desktop or mixed.

A decoder that only works on other people's pages isn't worth much. Here are our two free offers, line by line.

One unit makes the two comparable: browser-minutes. The NANO tier's 2,000 visits of up to 30 seconds add up to at most 1,000 browser-minutes a month, about 33 minutes of one browser a day. A 24-hour trial on a single browser holds 1,440 browser-minutes, so a one-day trial on 10 browsers would hold 14,400, more browser time than a year of the free tier. That's worth knowing before you choose where to run a test.

Neither free offer can show you Search Console clicks or promise positions, and the NANO tier can't show you country targeting, which starts on paid plans. Both send visits you can check against the lines above in your own log. The paid plans are on the Sparky Traffic Bot and Website Traffic pages.

Give each line two points if the page states it in terms you can check, one point if it names the feature without a number, and zero if it's missing.

Eight to ten points: worth a trial, and you'll know what to count. Four to seven: test before you trust anything else on the page. Under four: the page is selling feature names. The six pages we read scored between 3 and 6. Run it on ours as well. Neither of our pages scores ten: both lose a point on device split, because neither states a ratio for mixed.

Four lines can't be backed by any feature list, so treat them as promises to walk away from:

• "Undetectable." A vendor can't see the checks your host, CDN, analytics tools and ad networks run, so it can't promise how they'll classify a visit.
• "Improves rankings." No bot controls what a search engine concludes. Bounce rate isn't a ranking factor, and Google has publicly dismissed dwell time theories.
• "AdSense safe." AdSense's program policies prohibit "automated click and impression generating tools and the use of robots or deceptive software", and list paid-to-surf and autosurf programs among the traffic sources publishers can't use (AdSense program policies). The honest version of an ads line describes a mechanism you can check, such as ad requests being blocked so the ads never load, which is how we describe Website Traffic. A promise that your account is safe can only come from the ad network, and its policy has already answered.
• "Real visitors", with no noun after "real". A real browser isn't a real person. Ask which one the line means.

A page that passes this scorecard has earned a test, not trust. The questions a feature list can't answer go to the vendor directly, and 12 questions to ask a traffic provider covers them.