# Location Reporting Accuracy Explained

## How Delivr estimates location, why accuracy can vary, and how privacy-first fallbacks work.

Get a clear view of where your audience is engaging with your content. This article explains how Delivr estimates location, from highly accurate GPS-based detection (when users opt in) to IP-based geolocation for broader insights. You’ll learn why reported locations can sometimes vary — due to VPNs, corporate networks, or device settings — and how Delivr’s smart fallback system ensures reporting stays reliable. Plus, we highlight our privacy-first approach and compliance practices so you can trust the data you see.

## How accurate is IP-based location reporting?

Delivr uses trusted, industry-leading IP geolocation databases to estimate where scans and visits originate. These databases are highly reliable at the country level, with accuracy of up to 99.8%.

Accuracy decreases as location granularity increases and varies by region:
- **Country**. Very high accuracy
- **State / Province**. Moderate accuracy, varies by country and network
- **City**. Lower accuracy, often unreliable

Additionally, in some countries and regions, digital infrastructure is less developed or frequently shared across neighboring areas. This can further reduce the accuracy of IP-based geolocation compared to more digitally advanced regions.

These limitations are inherent to IP-based location tracking and represent an industry-wide challenge — not something specific to Delivr.

## What IP-based location is (and isn’t).

IP-based location identifies the network routing location of a request — not the precise physical location of a person or device.

**Delivr**:
- Reports location based only on available signals
- Does not infer or guess precise locations
- Does not attempt to fingerprint users
- Does not track individuals across sessions or platforms

Location data is used for aggregate reporting and analytics, not for identifying or following individual users.

## Why IP-based location can be inaccurate

An IP address represents a connection point on the internet, not a physical address. Several common scenarios can cause IP-based location to differ from real-world location.

## VPNs, proxies, and anonymization services.

When a user accesses a link through a VPN (Virtual Private Network), proxy, TOR exit node, or anonymizing service, their traffic is routed through a remote server.

**As a result**:
- The detected location reflects the VPN or proxy server
- The user’s true location is intentionally masked
- The server may be in a different city, state, or country

## Corporate, institutional, and shared networks.

Company Wi-Fi, universities, hospitals, and other large organizations often route all traffic through centralized gateways.

**This can cause scans to appear as if they originate from**:
- A single headquarters location
- A regional data center
- A different country than the user’s actual location

## Mobile networks vs desktop connections.

Most QR scans happen on mobile devices.

**Mobile carriers frequently**:
- Route traffic through regional hubs
- Rotate IP addresses
- Share IP ranges across wide geographic areas

This makes city-level IP location less reliable on mobile than on fixed broadband connections.

## Automated security scanners and link preview services.

Email providers, messaging platforms, and enterprise security systems often automatically follow links to check for malicious content.

**In these cases**:
- Visits originate from the service’s infrastructure, not a human
- IPs may appear repeatedly
- Some services re-scan links on a scheduled basis

These automated checks can appear as scans or visits even though no user actively engaged.

## How time affects IP accuracy.

IP geolocation databases are continuously updated, but they are not real-time.

**Over time**:
- IP ownership and routing change
- Mobile carriers reassign IP ranges
- Cloud providers move infrastructure

This means an IP that mapped accurately yesterday may resolve differently today. Delivr reflects the **best available data at the time of access**, without retroactive adjustment.

## What happens when device location services are turned off?

If a user has location services disabled on their device:
- **Country-level location** may still be available, depending on the device and network
- **State and city data will not be available**
- Delivr automatically falls back to **IP-based geolocation**

This behavior is controlled by the operating system and device manufacturer — not by Delivr.

## How Delivr handles location accuracy and fallbacks.

Delivr follows a clear, conservative hierarchy and never guesses when data quality is low:
1. GPS-based location (only if explicitly enabled and approved)
2. IP-based country-level location
3. Unknown / not reported when signals are insufficient

When higher-accuracy data is unavailable, Delivr reports less granular results rather than inferring precision.

## Decision Flow

## Optional GPS-based location (higher accuracy).

Delivr supports **GPS-based location detection** as an optional feature.

## How it works

When enabled and approved by the person scanning:
- iPhone and Android devices use built-in **GPS receivers**
- Location is calculated using **trilateration**, based on time-coded radio signals from at least four satellites
- Devices determine precise **latitude and longitude** (and altitude when available)

Both platforms use **Assisted GPS (AGPS)**, which combines:
- Satellite data
- Cellular networks
- Wi-Fi positioning
- Bluetooth signals (when available)

This significantly improves speed and accuracy compared to satellite-only GPS.

## Consent, privacy, and user control.

GPS-based location **always requires explicit opt-in**.
- Users are clearly prompted for permission
- If permission is granted, GPS data is used for that interaction
- If permission is denied, Delivr immediately falls back to IP-based location
- No location data is collected silently or without consent

## What IP-based vs GPS-based location is best suited for.

**IP-based location is suitable for**:
- Country-level audience insights
- Regional trend analysis
- High-level reporting and analytics

**IP-based location is not suitable for**:
- Store-and facility-level attribution
- Determining exact physical presence
- Compliance decisions that require precise location

**GPS-based location is suitable for**:
- Use cases requiring higher precision
- Experiences where the user explicitly opts in
- Scenarios where accuracy matters more than coverage

## Accuracy Ladder

### Most Accurate

### GPS-based Location (Opt-in)
- Device-level precision
- Latitude / Longitude
- Requires explicit user consent

### IP-based Country
- ~99.8% accuracy
- Reliable for market-level insights

### IP-based State / Province
- Variable accuracy
- Affected by networks and carriers

### IP-based City
- Often unreliable
- Frequently misleading on mobile

### Unknown / Private / Not Reported
- Used when signals are insufficient
- No guessing or inference

### Least Accurate

## Data handling and compliance (legal safe)

- Delivr processes location data in a privacy-first, compliance-oriented manner:
- Location data is anonymized and aggregated
- Delivr does not store continuous GPS trails
- No attempt is made to identify or re-identify individuals
- Location data is used solely for analytics and experience delivery
- Processing aligns with applicable data protection regulations, including GDPR and CCPA, depending on customer configuration

Customers remain responsible for determining whether optional GPS-based features are appropriate for their specific use case and regulatory environment.

## Summary

- IP-based geolocation is highly accurate at the country level
- Accuracy decreases at state and city levels due to network routing, VPNs, mobile carriers, and automated scanners
- Device settings and time-based IP changes affect results
- Delivr never infers or guesses precise location
- GPS-based location offers higher accuracy, but only with explicit user consent
- When signals are limited, Delivr safely falls back or reports no location at all

Last Updated 6 Feb 2026