Latency is the time delay between an action and its response. In networking, it refers to how long data takes to travel from one point to another, typically measured in milliseconds (ms).
Lower latency means faster communication, while higher latency can make websites, online games, and business applications feel less responsive.
For example, when a user clicks a link, the request travels to a server, and the server returns a response. Network latency contributes to this delay, alongside server processing and other operations.
How Is Latency Measured?
Network latency is commonly measured using round-trip time (RTT), which represents how long a signal takes to reach its destination and return.
Two common diagnostic tools are:
- Ping: Measures RTT using ICMP messages. A ping result of 20 ms means the round trip took approximately 20 milliseconds, not that the one-way delay was 20 ms.
- Traceroute: Identifies network hops between two points, helping investigate routing paths and potential delays. Individual hop timings do not always reflect actual traffic latency.
For applications, response-time metrics provide a broader view because they also include processing delays. Teams often track P95 and P99, representing the times within which 95% and 99% of measured requests complete, respectively. These metrics help reveal slow requests that averages can hide.
What Causes High Network Latency?
Several factors contribute to higher latency, affecting how quickly data travels across a network and how efficiently systems process requests.
- Physical distance: Data traveling over longer distances takes more time to reach its destination.
- Network congestion: Heavy traffic causes packets to wait in queues, increasing latency.
- Routing inefficiencies: Suboptimal network paths add unnecessary travel time.
- Processing delays: Routers, firewalls, and servers require time to inspect and process data.
- Transmission medium: Wireless interference and packet retransmissions can introduce additional delays.
- Protocol overhead: Connection setup, encryption handshakes, and repeated network exchanges add time to requests.
Latency vs. Bandwidth vs. Throughput: The Differences
Although closely related, latency, bandwidth, and throughput each describe a different aspect of network performance.
| Metric | Definition | Common unit | What it tells you |
|---|---|---|---|
| Latency | Time taken for data to travel or a request to receive a response | Milliseconds (ms) | How quickly communication happens |
| Bandwidth | Maximum theoretical data transfer capacity | Mbps / Gbps | How much data a connection can carry |
| Throughput | Actual amount of data successfully transferred per unit of time | Mbps / Gbps | How much data is actually delivered |
Why Does Latency Matter?
Latency plays an important role in the performance of digital services, affecting how smoothly users interact with websites and applications and how efficiently online services operate. High latency can affect user experience, business outcomes, and overall application performance.
Degraded User Experience
Low latency enables responsive websites, smoother video calls, and faster interactions in online gaming. When latency is high, users may experience delays when opening pages, submitting requests, or interacting with applications. These delays are especially noticeable in real-time activities, such as video conferencing and online gaming, where even brief interruptions can affect the user experience. For businesses serving users across regions, reducing latency makes digital services more responsive and consistent.
Reduced Conversions and Revenue
Slow response times can disrupt user interactions, delay transactions, and increase the likelihood of purchase abandonment. For e-commerce websites, delays during product browsing, checkout, or payment processing can interrupt the purchasing journey and discourage customers from completing a transaction. Similar issues can affect online services that rely on users completing forms, booking appointments, or submitting requests. Reducing latency helps minimize these points of friction and creates a smoother experience that supports customer engagement and conversion.
Increased End-to-End Response Times
In modern applications, a single request may involve multiple services, APIs, and databases. Each interaction can introduce additional network delays and processing time, which may accumulate when requests must be completed sequentially. As a result, even small delays across individual services can increase the total response time experienced by users. High latency can therefore affect the responsiveness of the entire application, not just one component, making efficient communication between services important for maintaining performance.
How to Reduce Latency
- Use a content delivery network (CDN): A CDN caches content on servers closer to end users, reducing the distance data travels and helping lower latency.
- Optimize network routing: Choose efficient network paths and reliable connectivity to minimize unnecessary delays.
- Reduce connection overhead: Minimize redirects and redundant requests, and reuse connections to avoid repeated setup time.
- Optimize application processing: Improve database queries and reduce unnecessary service dependencies to speed up responses.
- Monitor latency: Use ping, traceroute, and application response-time metrics such as P95 and P99 to identify performance bottlenecks.
For businesses reaching users across different regions, the location of CDN infrastructure and the quality of local network connectivity can make a significant difference.
With over 3,000 PoPs across more than 90 countries and regions, CDNetworks’ CDN helps deliver cached content from edge servers closer to users, reducing the need for long-distance data transfers. Its network presence and local ISP partnerships in emerging markets, including Southeast Asia, Latin America, Africa, and the Middle East, also help improve content delivery across regions where connectivity can be more complex.
FAQs
Is high or low latency better?
Low latency is better because it means data travels between systems or requests receive responses with less delay. High latency causes slower interactions and can negatively affect real-time applications, online gaming, video calls, and business transactions.
What is a good latency?
A good latency depends on the application and network path. Under 20 ms is excellent for many interactive uses, while under 100 ms is generally acceptable for many online activities. Business applications should also evaluate end-to-end response times and P95/P99 targets.
Can latency be zero?
Latency cannot be truly zero because data requires time to travel through physical media and systems need time to process requests. Even within a local network, transmission, propagation, and processing introduce some delay.
Is latency the same as ping?
Latency is the delay involved in data transmission or system response, while ping is a diagnostic tool that commonly measures network round-trip time. Ping results help estimate network latency but do not capture complete application response time.
Why is my latency high even with fast Internet?
High latency can persist despite high internet bandwidth because of network congestion, distant servers, inefficient routing, Wi-Fi interference, or slow application processing. Bandwidth determines data capacity, whereas latency measures delay, so the two can vary independently.
