One verdict on every visit, rolled up per source, worst first — down to the placement you buy.
Two jobs, usually sold as one
Invalid traffic creates two separate problems. The first is knowing: which visits were invalid, how many, and which source sent them. The second is stopping: making sure the network stops serving, or stops charging for, that traffic.
A blocking tool does both at once. It decides a visitor is invalid and immediately acts, usually by adding an IP address or an audience to an exclusion list in the ad platform. A report-only tool does the first job and hands you the second: here are the sources, here is the evidence, the cut is yours to make.
Neither is simply better. They answer different questions, and which one you need depends on where you buy traffic and how you want exclusion decisions made.
What automatic blocking does well
On the platforms it connects to, mostly Google Ads, Microsoft Ads and Meta, automatic blocking saves real work. A repeat offender clicking from the same address gets excluded within minutes, around the clock, without anyone opening the account. For a small advertiser on one Google Ads account, with a competitor clicking the same keywords every morning, that is exactly the right tool.
It also changes behavior: once an address is excluded, the ad simply stops showing there, so the next click is never paid for. A report tells you what happened; an exclusion changes what happens next.
Where blocking misses and misfires
- The lever is small. Google Ads allows up to 500 IP exclusions per campaign. Automated traffic that rotates through residential or mobile addresses rarely comes back from the same one, so an IP list mostly records where invalid traffic has already been.
- Shared addresses catch real people. Mobile carriers and many offices put large numbers of users behind a few public addresses. Excluding one of those addresses can hide your ads from real customers along with the bot.
- False positives are silent. A blocked customer does not show up in any report. They simply never see the ad, so a blocking tool’s mistakes are the hardest thing to measure about it.
- Most networks have no lever to automate. On native, push and pop networks, exclusion is by site, zone or widget, made in the network’s own dashboard, and few of those networks expose an API for it. There, automatic blocking has nothing to push to.
- The same tool grades its own work. When one system both decides what to block and reports how well blocking is going, there is no independent check on either.
Judge your own traffic like this — join the early-access waitlist.
Evidence before exclusion
A report-first workflow trades speed for precision:
- Measure before cutting. Judge every visit for a period long enough to see each source’s pattern, days rather than hours on smaller budgets.
- Rank sources, not visits. Sort zones, sites and widgets by the share of their visits judged invalid, compared with your clean sources. One bad visit proves nothing; a source whose visits are mostly invalid is a decision.
- Cut at the unit the network allows, from the worst down. Where each network lets you do it, and the limits, is in blocking by network.
- Keep the record. The same export is the evidence for a credit request, and the deadlines are short. Refunds by network lists what each network accepts.
- Re-measure. Sources rotate. A cut is checked by whether the next week’s traffic from the rest of the campaign looks clean.
ValidVisit is built for this workflow. It checks every visit against 100+ independent data points, judges it Valid or Invalid, with its evidence, and pins it to the zone, site or widget that sent it. The verdict never blocks or turns a visitor away, and ValidVisit never writes to your ad account; you decide what to exclude.
Rules for what you already know
Not every unwanted visitor has to be measured before you act. Some you can name in advance: a country you don’t sell to, connections from hosting and data-center servers, where almost no customer browses from, or VPN and Tor exits if your market doesn’t use them. Judging those visits one at a time tells you nothing new. ValidVisit’s traffic rules handle them instead: you list the countries and connection types once, on any plan, and every visit that matches is tagged and counted in the dashboard.
The split is simple. Rules are for what you already know you don’t want. The verdict is for what you have to measure: automation on home connections, click farms, a placement that went bad last week. The verdict never blocks anyone or sends anyone away; it judges, attributes and ranks, and the cut stays yours. Rules don’t look at the verdict, so a rule can’t misfire on a customer because of it.
A rule can act at three points, and each one stops something different:
- Send them to another page. Once your landing page has loaded, matched visitors are sent on to an address you choose, such as a not-available page or a regional store. The visit is recorded and tagged first. This saves nothing on that click: it was bought, and the page has already loaded. What it does is keep those visitors out of your forms, your checkout and your retargeting audiences.
- Block them at your firewall. For the countries on your list, the settings page writes a firewall rule you paste into your own CDN or firewall. Blocked visitors never see the page. The ad network has still charged for the click, and because the page never loads, ValidVisit stops seeing those visits too, so the matched count shrinks to what still gets through.
- Exclude the location in the ad platform. The same page gives you the country list to paste into Google Ads or Microsoft Ads location exclusions. This is the only one of the three that stops you paying, because the ad is no longer shown there. ValidVisit doesn’t make that change for you; you paste the list into your own account.
When each fits: exclude the location wherever the platform lets you, because it’s the only step that saves money. Add the firewall rule when traffic from those countries keeps arriving anyway, through sources you don’t buy or targeting you can’t tighten further. Use send-away when you’d rather show something than an error page, and for data-center and VPN matches, which the firewall rule doesn’t cover. Leave everything else to the verdict: a rule is a blunt instrument, right for a country you never ship to and wrong for anything you need evidence to judge.
When to block first, and when to run both
Block first when your spend sits on one or two platforms with automated exclusion, when the problem is a small number of repeat clickers, and when nobody has time to review sources every week.
Report first when you buy on native, push or pop networks, when you need evidence you can take to a network or a client, or when a false positive costs more than a wasted click.
Many advertisers run both: a blocker on Google Ads for the repeat offenders, and independent verdicts across every network to see what still gets through and where it comes from. The two do not conflict: ValidVisit sits on the landing page rather than in the click path, and never writes to the ad account the blocker manages. How the tools compare covers the blockers individually.