Picking A Proxy City That Fits The Account Persona
An account says it lives in Manchester. Its profile mentions a local football club, it posts in the evening UK time, and it follows a couple of Manchester restaurants. Its proxy exit is set to “United Kingdom”, and the rotating pool hands it an address in London one day, Glasgow the next and somewhere in Kent after that. At the country level, everything matches. At the level where a person actually lives, the account has no home.
Most proxy advice stops at the country. Match the exit country to the account and you’re done. That’s the minimum, and for some platforms it’s enough. But many platforms work at a finer grain than country, and many personas make claims about where they live that go well beyond it. Picking the city or region is its own decision, and it’s mostly about one thing: keeping the account’s location story stable and believable over months.
This piece is about the proxy side of that decision. The locale signals inside the browser, time zone, language and so on, have their own article. Here the question is which exit to pick and how to keep picking the same one.
What a platform can know about location
A platform builds its idea of where an account lives from many inputs, and the IP address is only one of them.
The IP gives a rough location through geolocation databases, which map address ranges to countries, regions and cities. The browser or app reports a time zone and a language. On mobile, apps with permission can read GPS or network-based location. Profiles often state a city outright. Content carries location too: the places an account mentions, the local pages it follows, the events it posts about. Marketplaces and delivery apps hold addresses. Payment methods have billing addresses. And behaviour has a clock, since the hours an account is active say something about its time zone.
Each of these can agree or disagree with the others. A platform doesn’t need any one of them to be precise. It needs them to tell roughly the same story. An account whose IP, time zone, stated city and activity hours all point at the same region is ordinary. An account whose IP says one city and everything else says another is a small puzzle, and if the IP keeps changing city, the puzzle is repeated every session.
So the city choice matters even when the platform never asks where you live. The IP is one witness among several, and it’s the one most operators never think to coach.
The good news is that you control it directly. You can’t easily change what a geolocation database says about an address, but you can choose which addresses an account uses.
Why city geolocation is fuzzier than it looks
Before choosing a city, it helps to know how imprecise IP geolocation really is, because it changes what “matching” should mean.
Geolocation providers are fairly reliable at the country level for most addresses. At the region or state level they’re usually decent. At the city level, accuracy drops a lot, and different providers often disagree about the same address. One database may place an address in the city centre, another in a suburb thirty kilometres away, another in the regional capital, because each is inferring from different registration records, routing data and user reports.
Mobile addresses are especially fuzzy. Carriers route mobile data through a limited number of gateways, and the public addresses often geolocate to the gateway’s city or to wherever the carrier registered the range. A phone in a small town may show up as the nearest big city, or as a city a long way off where the carrier has its core network. Residential broadband tends to geolocate closer to the real location, but not always.
This has two practical consequences. First, exact city matching is often impossible and not worth chasing. What matters is that the exit sits in a plausible region for the persona and stays there. Second, you should check what several geolocation sources say about an exit, not just one. If three lookup services put it in the right region and one puts it elsewhere, that’s normal. If they all put it somewhere unexpected, the exit doesn’t fit the persona.
Treat the city as a region with a centre, not a pin on a map.
Start from the persona, then check the supply
There are two ways to approach the choice, and one of them causes trouble.
The wrong way is to pick a city because it sounds interesting or because a client mentioned it, build a detailed persona around it, and then discover your proxy provider has almost no supply there. You end up either stretching to a distant city or reusing a handful of addresses across many accounts.
The better way is to start from what the account genuinely needs, check the supply, and let the two meet. If the account belongs to a real business in a real place, its city is fixed and you look for supply there or close by. If the persona is flexible, for example a general audience account for a country-wide niche, choose a city where your provider has deep supply and a large population, and write the persona around that.
Deep supply matters for two reasons. It means the exit can stay in that area even when individual addresses rotate or fail. And it means your accounts aren’t crowded onto a few addresses. Ten accounts all living in a small town that only has four addresses in the pool will cluster on those addresses whether you intend it or not.
Population matters because big cities give cover. A large metro area has millions of people, many of them on the same carriers and ISPs. A small town has fewer, and accounts from it are simply rarer on any platform.
Ask the provider what locations it actually has and how many addresses sit behind each. A location option in the dashboard doesn’t always mean meaningful depth behind it.
Time zones inside one country
In small countries, city choice barely touches the clock. In large ones, it decides the account’s time zone, and that’s where country-level matching goes wrong most visibly.
The United States spans several time zones. So do Canada and Australia, and several other large countries. An account that claims to live in New York but exits through Los Angeles is three hours off. Its browser time zone might say Eastern, its IP says Pacific, and its posting hours may fit one or the other but not both. Every session repeats the mismatch.
The fix is to choose the proxy region within the correct time zone first, then the city. If the persona lives on the East Coast of the US, the exit should be on the East Coast, and ideally in or near the persona’s metro area. The browser’s time zone should be set to match, and the account’s activity should follow that clock.
Posting hours are the part people forget. A persona in Sydney who is always active at what is 3am Sydney time looks strange regardless of the IP. If the operator works in a different time zone from the persona, the schedule has to follow the persona’s day, not the operator’s. That’s partly a scheduling problem, but it starts with picking a region the operator can realistically keep hours for. There’s no point giving an account a home in a time zone where its natural active hours land in the middle of your night unless you have someone who can work them.
For countries with a single time zone, this whole section collapses into choosing a sensible city. For the big ones, it’s the first decision.
Why hopping regions breaks the story
The most common city-level problem is making no choice at all. Rotating pools set to a country hand out addresses from anywhere in that country, and the account lives everywhere at once.
Real people move, but in patterns. Most of the time they’re at home or at work, which in most lives means the same metro area. They travel sometimes, and when they do, they go somewhere, stay a while and come back. A phone on a mobile network moves around its home region through the day, but it doesn’t wake up in a different part of the country each morning.
An account that appears in a different city every session produces a location history no person has. Individually, each session is plausible. Together, they describe someone who teleports. Platforms that track login locations, and most large ones do for security reasons, see this as a stream of unusual location changes. That can trigger security checks, “was this you” prompts or verification requests, and repeated triggers add to the account’s risk.
The fix is to restrict the account’s exits to one region. Many providers let you set a city or state for a rotating pool, or give you a sticky session that keeps the same address for a period. For a long-lived account, the best arrangement is usually a dedicated or sticky exit in the persona’s region, with any rotation happening within that region.
If your provider can only target by country, that’s a real limitation for location-sensitive accounts, and it’s worth knowing before you commit a fleet to it.
Some movement is normal
The opposite mistake is freezing an account onto a single address forever and treating any change as a disaster. That’s safer than hopping, but it isn’t quite how real connections behave either.
Home broadband addresses change occasionally when ISPs reassign them. Mobile addresses change frequently, sometimes several times a day, because carriers rotate them from shared pools. A person who uses both their home Wi-Fi and their phone’s mobile data logs in from at least two networks, often two different ISPs. Someone who commutes appears from a work network during office hours.
Platforms know all this. A login history showing a handful of addresses within one metro area, from one or two carriers or ISPs, is completely ordinary. What stands out is movement between distant regions, or between very different kinds of network, like a residential address one day and a hosting provider the next.
So the goal is a stable region, not a single frozen address. For a mobile persona, a mobile exit in the right metro area that rotates within the carrier’s local pool looks like a phone. For a desktop persona on home broadband, a residential exit that stays put for weeks and occasionally changes within the same ISP looks like a household.
And keep the type consistent with the story. A persona who is supposedly at home on a laptop every evening shouldn’t be arriving through a mobile carrier some nights and a residential ISP on others without some reason in the story, such as a phone hotspot while travelling.
Matching the rest of the persona
Once the region is chosen, everything else about the account should agree with it, and the proxy is only one of the things to align.
The browser or device time zone should be the region’s. The language settings should fit what people in that region use. In a bilingual region, either language works, but the choice should stay constant. The profile’s stated location, if it has one, should be in the region. Content should make sense for someone living there: local references, local seasons, local events at the right times. A persona in Melbourne posting about a heatwave in July has a problem the proxy can’t solve.
Commerce details matter on platforms that use them. A marketplace account’s shipping address, a delivery account’s service area and an advertising account’s billing address all make location claims that should match the exit region. Payment methods issued in one country and used from another are a well-known risk signal on their own.
Local features give the story away too. Platforms show local trends, local listings, local news. An account that consistently interacts with content from a different region than its exit is telling a mixed story, even if every technical signal matches.
None of this needs to be elaborate. A persona sheet with the region, time zone, language, active hours and any addresses the account uses is enough. The discipline is checking the proxy choice against that sheet, rather than choosing the proxy first and hoping the rest fits.
Checking an exit before you commit
Before assigning an exit or a region to an account, check what the outside world thinks of it. A few minutes here saves an account from months of living in the wrong place.
Open the exit in a browser profile and look it up on several IP geolocation services. Note the country, region and city each one reports. You’re looking for agreement at the region level. Disagreement at the city level is normal.
Check the network that owns the address, which the same lookups usually show. It should be the kind of network the persona would use, a local mobile carrier or a residential ISP that operates in that region. An address in the right city owned by a hosting company is the wrong kind of address, whatever the city says.
Check the time zone the lookups infer for the address, and compare it with the profile’s time zone setting.
If the provider gives you a rotating pool restricted to a region, sample it. Take ten or twenty addresses over a day and check where they geolocate. If most land in the region and a few stray elsewhere, decide whether that’s acceptable for the account. If they’re scattered, the region filter isn’t doing what the dashboard implies.
Record the results with the account. When you later wonder why an account got a location prompt, you’ll want to know where its exits were supposed to be and where they actually were.
Different platforms care about different grains
How much the city matters depends heavily on what the platform does with location, and it’s worth sorting your accounts by that before spending effort.
Local services care most. Marketplaces with local listings, delivery and ride apps, dating apps, local classifieds and anything showing “near you” results use location as part of the product. An account on one of those whose exit sits in a different city from its stated area is visibly inconsistent, because the platform is constantly comparing where the user says they are with where the results should come from. For these, the exit should be in the same metro area as the persona, and on mobile the device location should agree too.
Social platforms care about location mostly for security and for content. They track login locations to spot account takeovers, and they use location to tailor feeds and suggestions. Country and region consistency matters a lot. City precision matters less, except where the account’s content makes a strong local claim.
Advertising and payment platforms focus on country and billing details. The billing address, the currency, the tax settings and the country of the payment method carry more weight than the exact city of the IP. But a billing address in one city and logins from a region far away are still a mismatch worth avoiding.
Search and review platforms tied to physical businesses, where an account manages a listing for a real shop, are the strictest of all. The business is in one place, and the people who manage it usually are too.
So rank your accounts. The ones on local services get dedicated exits in the right metro. The ones on broad social platforms get stable regional exits. The rest can often live with a stable country-level choice.
When a client decides the city
Agencies and freelancers managing accounts for real businesses don’t get to choose the persona’s city. The client is where the client is, and the accounts need to look like they’re managed from there or nearby.
That’s the easy case in principle and the hard one in practice. In principle, the answer is fixed: exits in the client’s city or metro area, time zone set to the client’s, activity during the client’s business hours. In practice, your provider may have little or no supply in a mid-sized city, and the client’s real team may already be logging in from their own office, which means the account has an existing location history you need to fit.
Start by looking at that history. If the account’s owner and staff already log in from the business’s own connection, that address is the account’s home. Anything you add should be close to it. A regional exit in the same area is plausible, since an agency or an employee working remotely nearby is ordinary. An exit on the other side of the country is less so.
If supply in the right area doesn’t exist, consider alternatives before settling for a distant exit. The client’s own office connection, used through a device on their premises, is the most natural option where the client agrees to it. A team member based in that city is another. Being honest with the client about the limitation is a third, since some businesses would rather have fewer managed actions than risk their account.
Whatever you do, keep one arrangement and stick to it. A business account that alternates between the owner’s office and three different agency exits in three cities has a location history nobody can explain.
What changing an existing account’s region costs
Everything above assumes you’re choosing a region for a new account. Existing accounts are different, because they already have a location history, and that history carries weight.
An account that has logged in from one region for a year has a home as far as the platform is concerned. Moving it to a different region, even one that fits the persona better, creates a location change the platform will notice. Depending on the platform, that might trigger a login challenge, a security email to the recovery address or a temporary limit while the account proves itself.
So before changing an established account’s region, ask whether the current one is actually causing a problem. If the account has been stable, a region that’s slightly off from the persona may be better left alone than corrected. Consistency with the account’s own history often matters more than consistency with a persona sheet written later.
If the change is necessary, make it once and make it complete. Move the exit, the time zone and any profile location together, at a moment when the account would plausibly move, and then leave it. Expect friction for a while, and don’t stack other changes on top, such as a new device or a new browser profile, in the same week. One change at a time lets the account settle, and it lets you tell which change caused any trouble that follows.
The moving an account to a new setup guide covers the device side of this. On the location side, the rule is simple: moves should be rare, deliberate and permanent.
Small countries and city-states
For accounts in small countries, most of this simplifies. Singapore is effectively one city, so the city question disappears. The same goes for many small European states, Gulf states and island nations. The exit country is the city, and a single time zone covers everything.
What remains is the carrier and ISP question. In a small country, there may be only three or four mobile carriers and a handful of broadband ISPs. The persona’s network should be one of them, and it should stay consistent. Someone on one carrier’s mobile network doesn’t usually switch carriers every week.
Small countries also have smaller address pools, which makes clustering easier to trigger. If many of your accounts share a country with a small population, they’re drawing from a limited set of addresses, and the chance that two of your accounts end up on the same exit rises. That argues for dedicated exits per account, or at least careful tracking of which account used which address.
Neighbouring countries create a subtler issue. A persona in a small country near a border might plausibly show up from the neighbouring country sometimes, since people cross borders for work or shopping. That’s fine occasionally. It shouldn’t be the default because the provider happened to have cheaper supply next door.
And for small countries, geolocation databases sometimes place addresses in the wrong country entirely, especially for newly allocated ranges or ranges that moved between carriers. Check the country, not just the city, on every lookup.
When the account travels
Sometimes a persona genuinely needs to be somewhere else. A travel account, a business owner attending a trade show, a person who really does spend a month abroad. Handling travel well is mostly about making it look like travel.
Travel has a shape. The account is at home, then it’s somewhere else for a stretch of days, then it’s home again. It doesn’t flicker between the two. During the trip, the time zone would change on a phone that updates automatically, and activity hours would shift to the new local day, at least partly. Content usually reflects the trip.
If you move an account’s exit for a trip, move the device or browser time zone with it where a real device would change, and keep the account in the new place for a believable period. Then move it back and leave it there.
Don’t use travel as a cover for supply problems. If an account needs to be in its home region and your provider has no supply there this week, it’s usually better for the account to be inactive for a few days than to appear from somewhere else with no reason.
Expect security prompts when an account appears in a new country, and be ready to answer them. A login from abroad is one of the most common triggers for a verification check, even on genuine accounts. Our article on logging in from a new country goes through what that check involves and how to handle it.
Keeping the assignment stable across a fleet
Choosing a region for one account is simple. Keeping every account in its region for months, across a team, through provider changes and outages, is where the real work is.
Keep a record per account: its region, time zone, exit type, provider and any sticky session or dedicated address. That record is what everyone on the team checks before opening the account.
Configure the region in the profile itself, not in someone’s memory. If the proxy string carries a region parameter, it belongs in the profile’s saved settings. If a rotating pool needs a region filter, that filter lives in the configuration, not in a note that says “remember to pick Texas”.
Watch for silent changes. Providers sometimes retire locations, merge pools or change what a region code means. A rotating pool that used to stay in one state may start handing out addresses from a neighbouring one. Periodic sampling, like the checks above, catches that before a platform does.
Plan for outages. When the provider’s supply in a region runs short, decide in advance what happens: pause the account, or fall back to an adjacent region that keeps the same time zone. Letting the pool fall back to “anywhere in the country” by default is the one option that reliably causes trouble.
And when an account moves permanently, for example when a client relocates, treat it like a real move. Change the region once and update the time zone and profile details at the same moment, then let the account settle in its new home without bouncing back.
Mistakes worth avoiding
A few patterns cause most city-level trouble, and they’re easy to catch once you know them.
Setting the proxy region to the operator’s city instead of the persona’s. It happens when someone sets up profiles quickly and picks the familiar option.
Choosing a small or unusual city for flavour, then putting several accounts there. They end up clustered on the same few addresses.
Leaving rotating pools at country level for accounts whose personas claim a specific city.
Setting the proxy region carefully while leaving the browser time zone on automatic detection from the machine, which reports the operator’s time zone.
Running accounts on the operator’s schedule when the persona lives several time zones away, so the account is active at 3am local time every night.
Using an exit in the right city owned by a hosting provider, and assuming the city makes it fine.
Changing an account’s region to test whether it fixes a problem, then forgetting to change it back. Every change is a location event the platform records.
The honest limit
Picking the right region makes an account’s location story consistent. It doesn’t make the account look real on its own, and it won’t rescue an account that fails on behaviour, content or connections. Platforms also vary in how much they care. Some use location mainly for security prompts. Others use it for content, advertising and fraud scoring. Which signals a given platform weighs most isn’t something anyone outside can know precisely.
What’s reliable is the direction. Location signals that agree with each other and stay stable over time are ordinary. Signals that disagree or keep moving are unusual, and unusual gets looked at. Choosing a region the persona can live in, one your provider can supply, and a clock you can keep, removes location from the list of things that make an account unusual.
I run Singapore mobile proxy farms myself, and for Singapore personas the city question is simple because the country is the city. Most of the real work is keeping each account on the same carrier and the same kind of exit month after month. For the rest of the stack, from locale settings to picking the right proxy type for each platform, start on the Multi Account Ops home page.
Get new guides and videos first — join the Telegram channel.