Why two people in one household get flagged

you and your partner sign up to the same survey panel. you both answer honestly, you each use your own email, and you both do a few studies on the couch wifi. a week later one of you has a payout on hold, an account under review, or a feed that has gone strangely quiet. neither of you did anything wrong. the panel’s duplicate check just saw two accounts that look like they came from the same place and guessed they were one person.

this is one of the most common ways honest people run into trouble with a panel, and it is almost never explained at signup. i’m writing this from the operator side, in plain language, so you know what the system is looking at before it costs you a balance. if you want the wider picture first, why survey accounts get banned covers the other routes to the same outcome.

what it is

a household flag is what happens when a panel’s fraud or duplicate-account check links two or more accounts because they share signals that come from the same home. the signals are things like the public IP address, the device or browser, a payout account, a postal address or a phone number.

the flag is a link, not a verdict. the system has connected the accounts and is now asking a question: is this one person running two accounts, or two people who live together? on its own it usually cannot tell the difference, and that is the whole problem. someone quietly farming extra accounts and a couple sharing a router leave almost the same footprint.

panels care because they sell answers. a research buyer is paying for opinions from distinct people. if the same person answers twice, the data gets worse and the panel gets complaints. so the check leans towards suspicion, and households get caught in it.

what happens after the link varies, and no panel publishes its exact rules. in general it comes in three strengths:

passive: you see fewer invitations or get screened out more often, with no message at all
active: an account review, a request for ID or an explanation, or a delay on a withdrawal
hard: a balance is held or an account is closed, sometimes both accounts at once

how it works

start with the IP address. your ISP gives your router one public address. every phone, laptop and TV inside the home gets a private address (the ranges are set aside in RFC 1918, for example 192.168.0.0/16) and the router translates all of them to that single public address on the way out. to a survey site, two people in the same flat look like the same visitor.

it gets messier. some providers, mobile networks especially, run carrier-grade NAT, so several unrelated households share one public address. the IETF’s RFC 7021 walks through how that breaks applications that assume one address means one user. a sensible fraud system knows this, so it treats an IP as a weak hint and not as proof. a weak hint is still something, though.

the IP is only the first layer. the signals that carry more weight usually stack up like this:

ip address: the same public address, or the same network, on two accounts in overlapping time windows
browser and cookies: if you both log in from one browser profile, the site can set a cookie on the first login and read it back on the second. that is a direct link, no guessing needed
device fingerprint: a summary of what a browser reveals, such as screen size, time zone, language, fonts and how it draws graphics. two people on one family laptop produce the same fingerprint. the EFF’s Cover Your Tracks test shows you how identifiable your own browser is, and fingerprinting products sold into the survey industry, RelevantID from Imperium being one name you will see, exist to do this at scale
payout details: the same PayPal address, bank account or gift card destination on two accounts. in my view this is the strongest link of all, because it is where the money lands and it is the hardest thing to explain away
profile details: the same home address, phone number or surname, or one account signing up through the other’s referral link
behaviour: two accounts completing the same studies within minutes of each other, or giving near identical answers
history: if one linked account was already closed, the other can inherit some of the suspicion

none of these is a switch. they get combined into a risk score, and i won’t pretend to know any panel’s weights or cut-offs. what you can reason out is the direction. one shared IP adds a little. a shared IP plus a shared laptop plus one payout account adds a lot, and that trio is exactly what an honest household tends to have, which is why it happens so often.

if you are two real people, the practical version is short:

each of you gets your own account, your own email, your own phone number and your own payout account
use separate devices, or at least separate browser profiles, and never log in to each other’s accounts
fill in your profile as yourself. if your answers differ because you are different people, good. profile surveys, the unpaid work that decides your paid work explains why those answers get read so closely
tell support up front that two people in your home use the panel. most panels have a help page or a contact form, and a two line message costs nothing
do not use a VPN or proxy to make the two of you look like different places. that is the pattern the checks exist to catch, and it usually breaks the panel’s terms

why it matters

four reasons this is worth understanding before you sign up.

the cost lands on the wrong person. the usual penalty for a linked account is a held balance or a closure, and the person who did nothing wrong loses the same money as the person who did. a household link is one of the ways honest accounts end up in the same bucket as farmed ones.
the quiet version is hard to notice. a linked account rarely gets a notice. it gets fewer invitations and more dead ends. if you keep landing in screen-outs, survey routers and why you get bounced between studies covers the ordinary reasons, so you can rule those out before blaming a link.
flags stack. if you already guess at prescreeners to get into studies, a household link can make a manual review harder to pass, because a reviewer sees two connected accounts and answers that do not line up. why answering to qualify backfires covers that habit.
it is not personal, and that helps. panels are protecting data quality for their buyers. knowing that lets you respond calmly. a short honest note to support is a better move than opening a third account or arguing about it on a forum.

common misconceptions

“same IP means automatic ban.” no. an IP is a weak signal, and providers that share addresses across homes are common enough that a fraud team cannot ban on it alone. it becomes a problem when it lines up with the stronger signals, a shared device or a shared payout account.
“a different email address makes us two people.” email is the cheapest thing to change and the least useful to the panel. a matching device, network and payout account can still link you, however clean the two inboxes look.
“a VPN fixes it.” it tends to create a second problem. fraud tools can tell home broadband from hosting and VPN ranges, and those exit addresses are shared with a crowd of strangers. and if the aim is to hide that two accounts sit close together, that is the behaviour the rules are written against.
“only one person per household is ever allowed.” the rule differs by panel. some allow several members, some cap it, some are strict. read the terms of the one you use, whether that is Swagbucks, Toluna, Prolific or something else, and ask support if the terms are vague. i’d rather you ask and hear no than guess and lose a balance.

where to go from here

if you want to keep going, these are the four i’d read next:

why survey accounts get banned: the wider list of reasons, of which household links are one
filling prescreeners properly: how to answer honestly without screening yourself out
getting paid faster: how payout checks work and what tends to slow them down
five dollar survey math: how to measure your own hourly rate, so you can judge whether juggling any of this is worth the hassle

the full list of guides is on the blog index.

Written by Xavier Fok

disclosure: this article may contain affiliate links. if you buy through them we may earn a commission at no extra cost to you. verdicts are independent of payouts. last reviewed by Xavier Fok on 2026-09-20.