Skip to content

NewExtra 20% Off on 3-Month Plans! πŸŽ‰ Order your VDS server for 3 months and get an extra 20% off in the cart!

Blog

Tutorials, guides, product news and engineering insights from the team that runs your hosting.

Guides

Why Is Your Website Slow? 8 Hosting-Related Causes

10 October 2026 18 min read
Share this post
Reading time: about 12 minutesLevel: Beginner - IntermediateTopic: Performance and infrastructure

When you notice your site loading slowly, the first reflex is usually to blame the plugins. You're often right; but not every slowdown comes from your site. Sometimes what's slowing down is the infrastructure hosting the site itself: shared resources, slow storage, an old PHP version, a missing caching layer or a server far from your target audience.

In this guide, we cover 8 hosting-related causes that can make your website slow. For each cause, we explain the symptoms, how to verify it on your own and what you can do. The goal is to help you answer the question "should I change hosting, or optimize the site?" with measurements.

Short answer

Measure first. If the server's initial response time (TTFB) is high even with caching enabled and your site stripped down, the problem is most likely one of the causes in this article. If even a static file arrives late, the culprit is almost certainly the infrastructure.

Measure First: Where Does the Slowness Start?

"The site is slow" isn't a diagnosis; it's a symptom. The time it takes for a page to load is made up of several separate stages, and each stage has different causes: resolving the domain name, connecting to the server, setting up the secure connection (TLS), the server generating the page and downloading the content. Finding which stage takes longer determines whether the problem is in the hosting or the site.

A timing breakdown with a single command

The command below requests the page and prints the cumulative time from the start of each stage, in seconds. Replace example.com with your own domain. In Windows PowerShell, use curl.exe instead of curl:

● Terminal
curl -o /dev/null -s -w "DNS: %{time_namelookup}\nConnect: %{time_connect}\nTLS: %{time_appconnect}\nTTFB: %{time_starttransfer}\nTotal: %{time_total}\n" https://example.com

Because the values are cumulative, interpret the stages by subtracting them from each other:

  • If the DNS time is high: Domain name resolution is slow. The DNS provider or record configuration should be reviewed.
  • If Connect βˆ’ DNS is high: Reaching the server takes a long time. Physical distance or network routing is a factor.
  • If TLS βˆ’ Connect is high: Setting up the secure connection is slow. It could be an old TLS configuration or a distant server.
  • If TTFB βˆ’ TLS is high: The server is generating the page late. Hosting resources, PHP, the database and caching are decisive here.
  • If Total βˆ’ TTFB is high: The content is downloading slowly. Page size, compression and bandwidth should be reviewed.

Most hosting-related problems show up in the TTFB βˆ’ TLS difference. web.dev considers a TTFB under 0.8 seconds good and over 1.8 seconds poor. Don't rely on a single measurement; repeat it a few times at different hours. For browser-based measurements, tools such as PageSpeed Insights and WebPageTest also let you test from different locations.

Info

The time for a page served from the cache and one generated from scratch is very different. Repeat the measurement both with an address that has never been cached (for example by adding a random parameter to the end) and with a frequently visited page.

Cause 1

Shared Resources Are Insufficient or Overloaded

On shared hosting, the CPU, memory and disk resources of the same physical server are divided among many accounts. Providers set per-account limits on CPU, memory, I/O and concurrent processes for fair use. When a limit is exceeded, requests queue, slow down or return errors. In addition, a busy neighboring account on the same server can squeeze shared resources. This is called the "noisy neighbor effect" and varies with the density of accounts placed per server.

Symptoms
  • Pages are fast at night but slow during the day or at peak hours.
  • Occasional slowdowns even though visitor numbers haven't increased.
  • On some platforms, 503 or "Resource Limit Is Reached" (508) errors.
  • The CPU, memory or concurrent process limit graphs in the panel look constantly full.
How do you verify it?

Measure the TTFB of the same page 10 times each at different times of the day. The loop below does this in a single command:

● Terminal (Linux / macOS)
for i in $(seq 1 10); do curl -o /dev/null -s -w "%{time_starttransfer}\n" https://example.com; done

Note the results along with the times. If the values rise noticeably during peak hours and coincide with the same periods in the panel's resource graphs, resource sharing is a strong suspect.

What can you do?
  • Reduce resource consumption by cutting down PHP and database work with caching.
  • Remove unnecessary plugins and heavy scheduled jobs.
  • Ask your provider in writing which limits your account is hitting and what the limit values are.
  • If the limits are constantly exceeded, consider moving to a higher plan or to a VPS or VDS where resources are dedicated to you.
Cause 2

Slow Storage and a Disk I/O Bottleneck

A dynamic site reads files on every request, and the database keeps its data on disk. So disk latency is reflected directly in the server's response time. Mechanical disks (HDDs) aren't as fast as SSD and NVMe disks at random reads and writes. In shared environments, heavy I/O load can queue the operations of everyone using the disk.

Symptoms
  • The dashboard and database-heavy pages are especially slow.
  • The site slows down noticeably while a backup is running or a large file is being uploaded.
  • Pages that load quickly thanks to the cache are very slow when the cache is empty.
How do you verify it?

On shared hosting, you don't have access to the server, so ask your provider about the storage type (HDD, SSD, NVMe) and I/O limits. If you use a VPS or VDS, you can monitor CPU wait time and disk utilization on the server with these commands (iostat comes with the sysstat package):

● Terminal (on the server)
vmstat 1 5
iostat -x 1 5
  • The wa value in the vmstat output shows the share of time the CPU spends waiting on the disk. If it's consistently high, there's a disk bottleneck.
  • In the iostat output, a %util value consistently close to 100% and a high await time show that the disk can't keep up with the work.
What can you do?
  • Move to a service that offers SSD or NVMe storage.
  • Clean up unnecessary files, old backups and log files. Large numbers of small files strain both I/O and the file count (inode) limit.
  • Use caching to reduce disk access on every request.
Cause 3

Old or Misconfigured PHP

WordPress, WooCommerce and many web applications run on PHP. New PHP versions are usually noticeably more efficient than old ones; versions whose support has ended don't receive security patches either. Besides the version, configuration matters too. If OPcache is off, PHP files are recompiled on every request. If the number of PHP-FPM workers is insufficient, concurrent requests wait in a queue.

Symptoms
  • The dashboard and processing-heavy pages (cart, checkout, search) are slow.
  • Occasional 502 or 504 errors under load.
  • Only old PHP versions can be selected in the provider's panel.
How do you verify it?

You can see the PHP version in the hosting panel or on the Tools > Site Health > Info screen in WordPress. If you have command-line access, list the version and key settings with these commands:

● Terminal (on the server)
php -v
php -i | grep -e opcache.enable -e memory_limit -e max_execution_time

The command-line PHP settings may differ from the settings the web server uses; for exact values, check the phpinfo() output or the panel's PHP settings screen.

What can you do?
  • Move to the most recent PHP version compatible with your theme and plugins. Take a backup before the switch and, if possible, try it on a test copy.
  • Make sure OPcache is enabled.
  • Confirm that the memory limit (memory_limit) suits your site's needs.
  • If your provider doesn't offer current PHP versions, treat this as a serious reason to change hosting.
Cause 4

Database Server Bottleneck

Most dynamic pages query the database while generating a response. Slow queries, huge tables and missing indexes are usually caused by the site. But bottlenecks can also arise on the hosting side: the database server is shared and the number of connections is limited, memory settings are low, or the database is on a separate machine from the application, which adds network latency to every query.

Symptoms
  • Static files arrive quickly, but the first response of dynamic pages is slow.
  • The dashboard, search and filtering are especially slow.
  • "Too many connections" or "Error establishing a database connection" errors.
How do you verify it?

If you use WordPress, a developer plugin such as Query Monitor shows which queries take how long. On your own server, you can inspect running queries and the slow query log with a privileged user using these commands:

● MySQL / MariaDB (on your own server only)
SHOW FULL PROCESSLIST;
SET GLOBAL slow_query_log = 'ON';
SET GLOBAL long_query_time = 1;

The first command lists the queries running at that moment. The next two commands enable logging of queries that take longer than 1 second. These settings reset when the server restarts and require privileges.

What can you do?
  • First reduce the database load from your site: clean up old revisions, expired transients and unnecessary autoloaded options.
  • If you can use an object cache (Redis or Memcached), reduce the load of repeated queries.
  • If you need to change database settings and that isn't possible in a shared environment, consider a VPS or VDS where you can customize the configuration.
Cause 5

Missing Server-Side Caching

Caching is the most effective and cheapest way to speed things up. Page caching serves ready-made HTML directly, skipping PHP and database work entirely. Object caching keeps query results in memory. OPcache stores compiled PHP code. Some providers offer these at the server level (such as LiteSpeed cache, Nginx FastCGI cache or Varnish). Without server-level caching, all that's left is plugins; these often consume more resources and are less efficient.

Symptoms
  • When you request the same page twice in a row, the second response isn't faster.
  • The page is generated from scratch for every visitor, and CPU usage rises in direct proportion to traffic.
  • No caching information appears in the response headers.
How do you verify it?

Inspect the response headers. Headers such as Cache-Control, Age or a provider-specific X-Cache indicate that caching is working:

● Terminal
curl -sI https://example.com

Measure the same address twice. If TTFB drops noticeably after the first request, caching is working.

Caution

Cart, checkout, account and personalized pages must not be cached. When you enable caching, make sure these pages are excluded; otherwise one user may see another's content.

What can you do?
  • Find out about your provider's server-level caching options and enable them.
  • Use a compatible caching plugin with the settings your provider recommends.
  • Add object caching on sites with a high database load.
  • If your provider offers no caching solution at all and you can't get enough improvement with plugins either, that's a reason to change infrastructure.
Cause 6

Web Server and Protocol Configuration

How the page travels from the server to the browser also determines speed. HTTP/2 and HTTP/3 carry many files in parallel over the same connection, while HTTP/1.1 is slower on pages with many files. Gzip or Brotli compression reduces the size of HTML, CSS and JavaScript files. TLS 1.3 shortens secure connection setup. These settings are made on the server side and depend on the provider's configuration.

Symptoms
  • In the Network tab of the browser's developer tools, the protocol column shows http/1.1.
  • Text-based files download uncompressed, at full size.
  • There's a large difference between the total time and TTFB.
How do you verify it?

The first command below prints the HTTP version used (1.1, 2 or 3). The second command shows whether the server responds with a content-encoding header when you request compression:

● Terminal
curl -sI -o /dev/null -w "%{http_version}\n" https://example.com
curl -sI -H "Accept-Encoding: gzip, br" https://example.com
What can you do?
  • Ask your provider to check or enable HTTP/2 or HTTP/3 and compression support.
  • Use a CDN for static files to benefit from both protocol and distance advantages.
  • If these basic features aren't supported on your provider's infrastructure, consider new infrastructure.
Cause 7

Server Location and Network Latency

When a page loads, several round trips happen between the browser and the server: DNS resolution, the TCP connection, the TLS handshake and then the actual request. Each round trip takes as long as the distance and network routing between you and the server. If most of your visitors are in Turkey and your server is in a distant country, this latency repeats at every step. Routing quality is as important as distance: a geographically close server can be reached over a poor route.

Symptoms
  • The page content downloads quickly, but the first response is delayed.
  • Measurements taken from different cities and countries show large differences.
  • In the timing breakdown, the connect and TLS times are high.
How do you verify it?

Look at the connect and TLS stages with the timing breakdown command we gave at the start. In addition, you can see the time to reach the server and the route with these commands (on Windows, use ping -n 5 and tracert):

● Terminal (Linux / macOS)
ping -c 5 example.com
traceroute example.com

Some servers don't respond to ICMP requests; in that case ping may show no results. Services that test from different locations are more reliable in this respect.

What can you do?
  • Choose a service hosted in a data center close to your target audience.
  • Use a CDN for static files. A CDN doesn't fix the server response time of dynamic pages on its own; it's effective for cacheable content.
  • Use a reliable and fast DNS provider.
Cause 8

Background Load: Bots, Backups and Scheduled Jobs

Your hosting resources aren't used only by real visitors. Search engine bots, measurement and crawling tools, malicious bots, password attempts against the login page, backup jobs and cron jobs consume the same CPU, memory and disk resources. Without bot filtering or rate limiting at the provider level, this load is reflected directly in the speed real visitors experience.

Symptoms
  • Resource usage is high while visitor numbers are low.
  • The slowdown happens at the same times every day, for example during backup or cron hours.
  • In the access logs, many requests from a single IP and heavy wp-login.php or xmlrpc.php traffic.
How do you verify it?

If you can access the access logs, list the IP addresses making the most requests. The command below is for the Nginx log; on Apache the path is usually /var/log/apache2/access.log, and paths vary by installation:

● Terminal (on the server)
awk '{print $1}' /var/log/nginx/access.log | sort | uniq -c | sort -nr | head -n 10

On shared hosting, you can usually download raw access logs from the panel. Also check backup and scheduled task times in the panel.

Security

Heavy, irregular requests may be a sign of a password-guessing attack or malicious scanning. Protect your login page with strong passwords, two-factor authentication and attempt limiting; before blocking suspicious IP addresses, make sure you're not accidentally blocking search engine bots.

What can you do?
  • Move backups and heavy cron jobs to low-traffic hours.
  • Filter malicious and unnecessary bot traffic at the web application firewall or CDN level.
  • Remove unnecessary scheduled jobs or run them less often.
  • Ask your provider about its bot filtering and rate limiting options.

Rule Out Non-Hosting Causes First

Before changing hosting, check the common causes that come from the site itself. Otherwise, you'll move the same problem to the new server:

  • Unoptimized images: Large images, or images not converted to modern formats (WebP, AVIF), increase page weight and worsen LCP. Google considers an LCP of 2.5 seconds or less good.
  • Many heavy plugins: Each plugin can add extra queries, extra scripts and extra PHP work.
  • Third-party scripts: Ad, analytics, chat and tracking scripts slow down loading on the browser side; no matter how fast the server is, it can't hide this.
  • A heavy theme or page builder: Setups that generate a lot of JavaScript and CSS lengthen both download and processing time.
  • Outdated software: Old core, theme and plugin versions create both security and performance problems.
Info

A quick way to tell them apart: if most of the page appears quickly in the browser but TTFB is high, the problem is on the server; if TTFB is low but the page takes a long time to settle, the problem is in your site's content.

Order of Priority: What to Do When?

Not every improvement takes the same effort and cost. Start with the cheapest and most effective steps and move toward the infrastructure if needed:

1. Quick wins (today)

  • Enable caching and configure it correctly.
  • Upgrade PHP to the most recent compatible version.
  • Confirm compression and HTTP/2 support.
  • Remove unused plugins and themes.

2. Medium-term improvements (this week)

  • Compress images, convert them to modern formats and apply lazy loading.
  • Clean up the database and add object caching if needed.
  • Use a CDN for static content.
  • Move backup and cron times to low-traffic hours.

3. Infrastructure decision (if the measures aren't enough)

  • Ask your provider in writing about resource limits, storage type and caching options.
  • Consider moving to a higher plan, to a server with dedicated resources or to a different provider.

When Should You Change Plans or Hosting?

If most of the following are true, the improvement measures aren't enough and it's time to review the infrastructure:

  • Even with caching enabled and the site stripped down, TTFB is above 1–1.5 seconds for many hours of the day.
  • Even static files arrive slowly at peak hours.
  • Resource limit errors happen regularly.
  • Basic features such as current PHP, server caching or HTTP/2 aren't available from the provider.
  • The provider doesn't respond to problems with a measurable solution.

The options vary with your needs. If you're looking for a better environment without the management burden, web hosting, or WordPress hosting if you use WordPress, may be suitable. If you want to control the configuration yourself and have resources dedicated to you, you can explore our VPS or VDS server options. These products require server management knowledge; for the initial setup, you can use our Ubuntu VDS initial setup guide.

If you're thinking of changing providers, we covered which symptoms are a real reason to switch in detail in our article 7 Signs It's Time to Switch Hosting Providers.

Frequently Asked Questions

What should a good TTFB be?

web.dev classifies TTFB values under 0.8 seconds as good and those over 1.8 seconds as poor. On a well-configured server with caching, values far below that can be seen. Don't interpret the number on its own; interpret it together with the cache status and measurement location.

Will a CDN fix the slowness?

A CDN serves static files (images, CSS, JavaScript) from a point close to the user and reduces network latency. It doesn't, on its own, fix the time it takes the server to generate a dynamic page. If the problem is a high TTFB, you need to improve the server side first.

Is a caching plugin enough?

On many sites, it makes a big difference. However, server-level caching is usually more efficient and uses fewer resources. If TTFB is high despite the plugin, the problem isn't the plugin; it's the infrastructure.

Can I speed up my site without changing hosting?

Often, yes; caching, the PHP version, image optimization and database cleanup bring significant improvements on many sites. If your measurements are still poor after these measures and resource limits are constantly exceeded, an infrastructure change is a realistic option.

Does site speed affect search rankings?

Google uses metrics such as Core Web Vitals among its page experience signals. However, speed alone doesn't determine rankings; the relevance and quality of the content remain more important. Speed matters especially because of its direct effect on visitor experience and conversion rates.

When should I move from shared hosting to a VPS or VDS?

A VPS or VDS is a sensible step if you hit resource limits despite optimizing, if you need custom software or configuration that the shared environment doesn't allow, and if you have the knowledge to manage a server. If you don't want management responsibility, first consider a hosting plan with more resources.

Conclusion

The eight possible hosting-related causes of a slow website are: insufficient or overloaded shared resources, slow storage, old or misconfigured PHP, a database bottleneck, missing server-side caching, outdated web server and protocol configuration, a distant server location and background load. Each one shows up with a different symptom and can be distinguished through measurement.

The right approach is to proceed with data, not guesswork. Get the timing breakdown, find the stage where the problem starts, rule out non-hosting causes and apply the cheap, quick improvements first. If your measurements show the problem is still in the infrastructure, moving to a higher plan or a more suitable hosting solution gets results far faster than trial and error.

If the Cause Is the Infrastructure, Explore the Alternatives

If your measurements show the problem is in the infrastructure, you can explore Netiyo's web hosting and server solutions.

Explore Hosting Solutions
Never Miss a Post

Get our latest tutorials, guides and product news in your inbox.

You're subscribed to our newsletter.