Over the course of the last year, but particularly this year, we’ve noticed this strange phenomenon when looking at all our clients’ website traffic data: huge spikes in direct traffic, particularly on specific days. If you isolate the problem, typically these spikes are coming from one city on one single day.

The cities are places such as:

• • Ashburn, VA
  • Lanzhou, China
  • Singapore (the city-state)
  • Clifton, NJ
  • Council Bluffs, IA
  • Moses Lake, (WA)

To be clear, your website isn’t suddenly getting a rush of visitors from each of these places. This is non-human, often bot traffic.

How to Spot Bots/Non-Human Traffic in GA4

Look for these signs when reviewing your analytics:

• • A single-day spike from one city, followed by a sharp drop-off.
  • Traffic that appears across multiple client accounts from the same location.
  • Strange screen resolutions such as 800×600 or 720×1280.
  • Very low engagement time or no meaningful interaction.
  • Repetitive landing pages, especially the homepage or a few core URLs.
  • Geographic oddities that do not match your real audience, such as the cities above or Tehran, Iran, or other unexpected places.

When these signals appear together, the traffic is usually not a normal audience segment. It is often automated activity, proxy traffic, or technical validation traffic from data centers.

Why You Should Filter Out Bot/Non-Human Traffic

Bot traffic can make your site performance look better than it really is. If direct traffic climbs, you may think your brand is becoming more recognizable or that your SEO and offline efforts are paying off. In reality, the increase may be artificial and unrelated to real customer demand.

This matters because bot/non-human traffic can skew:

• • Direct traffic totals.
  • Average engagement time.
  • Bounce and engagement interpretations.
  • Landing-page performance.
  • Conversion-rate assumptions.
  • Channel attribution.

If you do not isolate bot traffic, you can end up making strategic decisions based on inflated data. That can lead to false confidence, wasted budget, and misleading reporting for stakeholders.

What Are the Types of Non-Human Website Traffic?

Bot traffic is usually an umbrella term for automated or scripted requests, but that doesn’t exactly encompass all non-human traffic to a website. There are several types:

• • Proxy traffic. A proxy is basically a middleman server between the user and the website. Proxy traffic is often used by bots, but a proxy itself is just an intermediary server. It can hide or change the apparent source of a request. Instead of the visitor connecting directly, the request goes through another server first, which can hide the original IP address (masked IP) and make the traffic look like it came from somewhere else. An example might be:
    • VPN/proxy use: A person or bot may, for example, connect through a server in Ashburn, so the analytics location points to Ashburn instead of the real user.

• • Automated checks. These are non-human requests made by software to test, verify, or inspect a website. These are usually bot traffic because they are machine-driven requests, but they can be legitimate if they come from uptime monitors, SEO tools, performance testers, security scanners, or platform-level checks. Example:
    • Data-center traffic: A crawler, app, bot, or monitoring system may originate from infrastructure hosted there, which also appears, for example, as Ashburn, VA.

• • Routing systems. Routing systems are the network paths that move traffic across the Internet. Sometimes a request from a cloud provider, content delivery network, or hosting service may be routed through a data center in a city like Ashburn or Council Bluffs, so that city shows up in analytics even though the “visitor” is really infrastructure, not a person. Routing systems are not bots by themselves. They are network infrastructure that can make traffic appear to come from a data center city instead of the end user’s real location. An example might be:
  • • Routing path: Sometimes the network path itself is what makes the location appear as Ashburn in analytics, even if no VPN is involved.

That is why these city names often appear together in suspicious traffic reports. The analytics location is often showing where the request entered the network rather than where an actual human is sitting.

What About LLMs Such as Claude, ChatGPT, Perplexity Etc. Does This Traffic Appear in GA4?

LLM bot traffic often does not appear cleanly in GA4 because many AI crawlers do not execute the JavaScript that analytics depends on, and GA4 also filters known bots automatically. Because not all AI bots are consistent, however, some traffic is reclassified. AI-bot-related visits can be lumped into Direct, Referral, Organic Search, or Unassigned rather than appearing as a neat “LLM” source.

In other words, with AI-driven traffic, analytics alone often cannot tell you whether the visit came from a human user or an automated crawler. Human clicks from AI tools can look like ordinary referrals, while LLM bots may be filtered, misclassified, or hidden unless you inspect server logs.

What Each City May Indicate About Bot Traffic in GA4

Ashburn, VA, Traffic in GA4

Ashburn is one of the biggest data-center hubs in the United States, so traffic from there is often infrastructure-related rather than human. Requests may come from cloud services, crawlers, automated checks, or routing systems tied to hosting providers.

If you see a one-day burst from Ashburn, treat it as a technical signal first. It is usually not evidence of a real surge in consumers from Northern Virginia.

Lanzhou, China, Traffic in GA4

Traffic from Lanzhou is often a sign of proxy behavior, routing noise, or automated requests rather than genuine local visitors. A sudden spike from Lanzhou usually needs scrutiny because it rarely matches the behavior of a normal customer audience.

If the sessions are brief, repetitive, and non-converting, the traffic is probably not commercial demand. It is more likely part of a bot pattern or infrastructure path.

Singapore, Traffic in GA4

Singapore is a major hosting and cloud region, so traffic from there can reflect technical infrastructure, scanning, or automated services. That does not make every visit suspicious, but it does mean the traffic should be checked carefully in context.

If Singapore appears in an isolated spike with weak engagement, it is often a bot or network activity pattern. If you truly serve Southeast Asia, compare it against campaign activity and conversion quality before assuming it is real interest.

Clifton, NJ, Traffic in GA4

Clifton, NJ, is another location that often shows up in suspicious traffic reports because of network and hosting activity. It can be tied to automated systems, proxies, or data-center routing rather than end users.

When Clifton appears as a sudden spike, it is usually better to think “infrastructure” than “new market.” The key question is whether the sessions behave like real visitors.

Council Bluffs, IA, Traffic in GA4

Council Bluffs is one of the most common bot-traffic locations because it is tied to major Google infrastructure and data-center activity. In Shopify environments, Council Bluffs often appears because of automated visits, testing behavior, and system-level requests associated with the platform’s ecosystem.

This is why many marketers see Council Bluffs in analytics and assume they suddenly got traffic from Iowa. In reality, it is often Shopify bot traffic, Google infrastructure traffic, or other automated activity unrelated to real users.

Indeed, if you run a Shopify store, Council Bluffs deserves special attention. Shopify merchants have long reported unexpected traffic from Council Bluffs in analytics, and it is often connected to bot activity, platform testing, or automated requests rather than shoppers.

What this usually looks like:

• • A sudden spike in direct traffic.
  • Sessions from Council Bluffs concentrated on one day.
  • Little or no conversion activity.
  • Weak engagement and repetitive page views.
  • No corresponding ad campaign or promotion.

How to interpret it:

• • It is usually not a sign that your store is suddenly gaining traction in Iowa.
  • It is more often a reporting artifact or automated traffic.
  • It should not be used to judge brand growth or campaign success.

Moses Lake, WA, Traffic in GA4

Moses Lake, WA, can also be tied to infrastructure and data-center activity rather than human visitors. Like the other cities on this list, it often appears in traffic reports because of automated systems, cloud routing, or technical checks.

If Moses Lake shows up as a one-day spike, it is usually part of the same broader bot pattern. The main thing to look at is whether the traffic is shallow, repetitive, and disconnected from conversions.

How to Isolate Bot Traffic in Google Analytics (GA4)

Use a structured process to separate likely bot traffic from your core reporting:

• • Create a city report and isolate the recurring suspicious locations.
  • Compare those cities against engagement metrics, not just sessions.
  • Review screen resolution, device category, and browser patterns.
  • Check whether traffic arrives in a single-day burst.
  • Compare direct traffic against conversions and revenue.
  • Build a watchlist of repeat offenders such as Ashburn, Council Bluffs, Clifton, Singapore, Lanzhou, and Moses Lake.

Remember, the goal is not to eliminate every bot. The goal is to stop bot traffic from controlling your business decisions. If a city appears repeatedly and behaves like infrastructure, it should be treated as noise until proven otherwise.

Conclusion

Bot traffic in GA4 can make your site look busier, broader, and more successful than it really is. If you are seeing strange direct traffic spikes from data-center cities, the safest assumption is that the data needs filtering and context before you trust it.

The more carefully you isolate bot activity, the more accurate your reporting becomes. That gives you a better read on what real users are doing, what marketing channels are working, and where your actual growth is coming from.

If you need help in this area, don’t hesitate to contact us at Marketing Nice Guys for a free consultation.