Cloaking for affiliate marketing means using a filter that sits between the ad and the landing page and, for every click, decides whether to let the visitor through to the offer or show them a neutral page. Media buyers and teams use it so they do not pay for bots and click fraud, to hide their funnel from spy tools, to cut off-target geos and to get statistics they can actually make decisions on.
If the term itself is new to you, start with what cloaking is. Here the discussion is practical: which media buying problems a filter really solves, where the line with ad platform rules runs and what the workflow looks like day to day.
Cloaking in affiliate marketing: what we are talking about
Media buying means purchasing ads on one platform and reselling the result to an affiliate network on a CPA basis (pay per action: a lead, a purchase, a deposit). The media buyer pays for clicks now and gets money from the network later, and only for confirmed conversions. This is where a cloaker comes in.
People use the word "cloaker" in two senses:
- A traffic filter: every visit is checked by network, device, browser, country and the ad click ID. Bots, scanners and spy tools go to the White Page (a neutral page), people go to the offer.
- A way to hide the offer from the platform's review. This is exactly what ad platform rules prohibit, and we will be honest about it below.
The first sense is a working tool without which money leaks out of media buying. The second is a violation that gets accounts banned. This article is about the first.
Terms. An offer is an advertiser's proposal that the network pays for an action on. A funnel (or bundle) is the combination of source, creative, landing page and offer that makes a profit. A postback is a message from the network that a click produced a conversion.
Why affiliate cloaking matters: five media buying problems
Bots and click fraud eat the budget
Any paid traffic contains clicks no human made: auto-clickers, scrapers, third-party quality checks, incentivized clicks from farms. On large platforms the share is usually lower; on cheap push, popunder and some native networks it is noticeably higher. You pay for every such click, and the network will never pay you for it.
A filter does not return money already spent, but it does two things: bots do not reach the offer and do not damage your relationship with the network through junk leads, and you see which placements and subnets they come from, so you can switch off the source. How to spot inflated clicks is covered in click fraud.
Spy tools steal the funnel
Spy tools collect other people's ads and follow them to save the landing page and the offer. If your working funnel ends up in such a database, a week later ten teams are running it in the same geo, the cost per click goes up and the creative burns out. Spy tool robots usually come from data centers and automated browsers, so a good filter sends them to the neutral page. How this works is explained in protecting landing pages from spy tools.
Off-target geos and devices
An offer accepts leads from one or two countries, and platform targeting is not perfect: some impressions leak into neighboring regions, and some users sit behind a VPN. The network will reject a lead from the wrong country, but you have already paid for the click. Filtering by country, language, device and schedule cuts these visitors before the offer.
Clean statistics for decisions
A media buyer makes decisions based on numbers: which creative to switch off, which placement to scale, where to raise the bid. If a third of clicks are bots, CR is understated and EPC (earnings per click) lies. Filtered traffic gives numbers you can cut and scale on.
Tracking, postbacks and money in one place
The filter already knows everything about every click: source, geo, device, campaign tags. It makes sense to link the click to the conversion there too, via click ID and the network's postback, pull in spend and calculate profit. That is why a modern cloaker for CPA almost always comes with a tracker. The difference between the two is covered in cloaker vs tracker.
What ad platform rules say about cloaking
Filter vendors usually keep quiet about this, so let us say it plainly.
Large ad platforms (Meta, Google, TikTok and others) prohibit showing the ad review system content different from what the user will see. At Google this falls under the circumventing systems policy, at Meta under landing page requirements and review circumvention; other platforms word it differently, but the meaning is the same. Rules change, so check the current version in the platform's own help center.
What a violation leads to:
| Consequence | What it means in practice |
|---|---|
| Ad disapproval | The mildest: the ad does not run |
| Account suspension | Often without prior warning for serious violations |
| Suspension of linked accounts | Business managers, profiles, ad accounts sharing a payment method or domain |
| Advertiser ban | The platform refuses to work with the person or company |
| Domain blocklisted | Links to it stop passing review |
Hence three honest conclusions for an affiliate:
- A filter does not make a prohibited offer allowed. If a platform does not allow a topic or requires a license, that is solved with documents and certification, not technology.
- No filter guarantees your account will not be banned. The platform's decision depends on the account history, payment method, creatives, domain, offer and user complaints. Promises of "100% ban-proof" are a reason to walk away.
- The legitimate value of a filter is real. Filtering bots, click fraud, spy tools and wrong geos, clean statistics and money tracking work just as well when the ad and the page comply with the rules.
How Google words this rule and what breaking it leads to is covered in Google Ads circumventing systems policy.
Cloaker for CPA: who uses it and how
Scenarios differ by scale.
- A solo media buyer tests funnels on small budgets. What matters is that a bot does not eat the test and that the log shows why a click went the wrong way.
- An affiliate team has several media buyers working with shared sources and offers. It needs roles and access: a buyer sees their own flows, the team lead sees everyone, the finance person sees the money.
- An agency or growth team buys traffic for its own product or a client. For them the filter is an anti-bot tool and a tracker, and the client report must open without access to the dashboard.
In all three cases the filter does the same thing: a real person passes through to the offer quickly and unnoticed, a bot or off-target visit does not, and every decision is explained.
What working with ArtisanClo looks like
ArtisanClo is a cloud service for filtering ad traffic with a built-in tracker. Your site and domain stay with you: you connect with a JS tag on the page or a PHP file on your server, and there is a plugin for WordPress. The workflow looks like this.
- Add a traffic source. The platform template maps its parameters and SubIDs to the ad link's tags and picks protection suited to the platform.
- Create a flow. Cloaking, Tracker or both, the offer URL and a White Page: your own or a ready one from the design catalog for your topic.
- Choose strictness. Soft, Balanced or Strict and the countries you need. Balanced cuts VPNs, proxies and data centers and suits most campaigns.
- Connect the landing page: the JS tag as the first line in
<head>or the PHP file, then Check my page. How these architectures differ and which to choose is covered in server-side cloaking vs JS tag. - Set up the postback. The affiliate network reports leads and sales, and Offer link parameters rules put our
{click_id}into the parameter the network expects, for examplesub1. More in postback URL and S2S tracking. - Launch traffic and read the click log. Every visit has a decision reason, one of dozens of clear labels such as "VPN or proxy blocked" or "Country not allowed".
- Count the money. Spend, revenue, profit and ROI by source, offer and geo; conversions can be sent back to ad platforms.
This is what an offer URL looks like once the rule "Set sub1 = {click_id}" has put the click ID into the network's parameter:
https://track.partner.com/click?offer=55&utm_campaign=nutra_de&sub1=C-01JB8Q3Z7XKP4M2R9TVN6WDY5H
Tip. Do not judge a filter by its filtered share. On paid traffic with a verified click ID, the pass rate to the offer can be high, and that is normal. What is worrying is a very low pass rate: it means a rule is cutting real people along with bots. The click log will show which one.
For teams, ArtisanClo has roles with per-section permissions, Projects for separating flows and guest links to reports. The full feature list is on the features page and the terms on the pricing page. Payment is in cryptocurrency, and there is a free trial.
How to tell the filter is helping, not hurting
The first days after launching a campaign through a filter are the time to test the filter itself. It can err in both directions, and the second mistake costs more than the first: a missed bot costs one click, while a blocked buyer costs a conversion.
| What to look at | Where | Good sign | Warning sign |
|---|---|---|---|
| Reason breakdown | Click log | Filtered visits are data centers, robots, wrong geo | Many filtered visits from residential and mobile networks of the target country |
| Conversions from passed clicks | Flow report | Leads come from clicks that passed the filter | There are clicks but no leads at all |
| Gap with the ad account | Click comparison | The difference is explained by bots and repeats | The difference grows for no visible reason |
| Pass rate | Dashboard | Stable for this source | Dropped sharply after a rule change |
If you see a warning sign, start with the log: the reason on filtered visits usually names the rule that fired. Most often real people are cut by a click-per-IP limit that is too strict, blocking IPv6 on mobile carriers or requiring a referrer where the platform's in-app browser drops it. Typical cases are covered in why a click went to the White Page.
In ArtisanClo, Diagnostics handles this check: it replays past visits through the current settings and shows how much traffic each filter cuts. And with the PHP file connection you can turn on Shadow mode for a while: every visitor goes to the offer, and the log shows whom the filter would have cut. That way a rule is tested before it starts cutting traffic.
What a cloaker will not do in affiliate marketing
It is worth aligning expectations in advance, so you are not disappointed in the tool and do not blame it for other people's mistakes.
- It will not make a funnel profitable. If the offer does not convert in this geo, the filter shows it faster but does not fix it.
- It will not replace the creative and the landing page. Conversion depends on what the person saw, not on who was kept out.
- It will not cancel the platform's rules. The ad and the page must comply with the requirements.
- It will not catch every bot. Every filter makes mistakes both ways; a good one shows where, so you can adjust the rule.
In short
- Cloaking for affiliate marketing is a filter between the ad and the landing page: bots, click fraud, spy tools and wrong geos go to the White Page, people go to the offer.
- The main value in media buying is a saved budget, a protected funnel and statistics you can make decisions on.
- Showing the platform's review a page different from what the user sees breaks the rules and risks suspension of linked accounts. A filter does not change that and does not cancel a ban.
- The working setup: a flow, connection with a tag or PHP file, a log of reasons, a postback and money tracking in one place.
