Shared hosting is by far the most popular option for most people when creating their first website. It’s generally cheap, but still provides your site with decent performance while it grows. However, this type of hosing often starts struggling to keep up when faced with serious traffic.
For this reason, you may want to look for a more robust hosting plan once your site becomes more successful. A good affordable option is managed WordPress hosting, or if your budget allows, consider a dedicated hosting provider. Of course, making the decision to upgrade will depend on the kind of traffic increase you have (or are expecting to see). To give you an idea of how much traffic shared hosting can handle in general, we’ll conduct a few performance assessments using Load Impact – a tried and tested online tool.
Let’s go!
As you may already know, shared hosting plans get their name from the fact that they split server resources among multiple customers. If you’re using shared hosting, your website will be located on the same server as several other sites.
The main upside to this approach is that you can save a lot of money on hosting. Providers can offer you low prices because you’re not using dedicated hardware. However, problems sometimes arise when hosting providers oversell their shared servers. After all, each server only has a limited amount of resources. If there are too many websites running on the same hardware, and they’re each receiving a decent amount of traffic, their performance can suffer.
Your experience with shared hosting will ultimately be shaped by the provider you choose. The best ones in the business will limit the number of users on each server, so they can offer you excellent performance. Even so, as your website grows in popularity, there will likely come a point when it can benefit from a more robust form of hosting.
What’s more, if one website starts to receive a significant amount of traffic, that can impact performance for other sites on the same server. In that case, your website would also start taking longer to load, which can lead to a higher bounce rate and unhappy visitors. If you ever find yourself in this situation, we recommend upgrading your plan to a Virtual Private Server (VPS), which is a step up from shared hosting.
Of course, it can be hard to gauge when it’s time to upgrade from shared hosting performance to something with more resources. After all, there are a lot of reasons your website might feel a bit sluggish, aside from your hosting provider. This means it can be helpful to know just how much you can expect your shared hosting plan to handle before performance begins to suffer.
To give you a clearer picture of what to expect from shared hosting, we decided to run some tests. Our goal was to figure out how well a few popular WordPress shared web hosts fare under stress. We settled on three of the best web hosts that offer affordable shared plans: GreenGeeks, A2 Hosting, and SiteGround. We chose comparable shared hosting plans on each, and set up three basic and identical WordPress sites.
After that, we used Load Impact to test how our sites fared under stress. Load Impact enables you to see how long a website takes to load under traffic, by simulating multiple simultaneous visits:
For each test, we chose a set number of visitors and a duration. For example, if you tell Load Impact to test your website using 20 visitors over five minutes, it will add one visitor at a time until it reaches a cap of 20. Then, you’ll be able to see how long your site took to load at every point along the way.
Here’s an example of a Load Impact result graph:
Our goal was to find out if there’s a predicable threshold where most shared host plans start to buckle. For example, if a website’s loading times start to spike when it gets five simultaneous visitors, then you know your web host isn’t up to par. On the other hand, if your shared plan can handle 50 concurrent users, then you probably won’t need to upgrade to something better for a long time.
We settled on three rounds of testing for each host, using these specifications:
- 15 users over a period of five minutes
- 30 users over a period of five minutes
- 50 users over a period of five minutes
Each time we ran one of these tests, Load Impact gave us a detailed breakdown of how loading times varied during the entire duration. Plus, we got access to average loading time information:
In the next section, we’ll analyze the results we saw. For now, suffice it to say that because most websites have peaks and dips in traffic, you should only take these results as benchmarks. They can, however, provide a rough idea of what a quality WordPress shared hosting plan can handle.
It’s also important to remember that in real-life scenarios, there are plenty of other things you can do to improve performance beyond upgrading your hosting plan. For example, you can always use a Content Delivery Network (CDN) for your website, or set up browser caching.
Upgrading your plan can cost a lot in the long run, so it’s not a decision to be taken lightly. You may want to try a few other solutions first, and see if they make a meaningful difference. If they don’t, it might be time for a change of pace.
If you crunch the numbers, you’ll realize that a website receiving 15 hits every five minutes would be getting 64,800 visitors per month. That’s a massive amount of traffic for most sites, which means that even our initial test could have been enough to strain a basic shared hosting plan. At least, that’s what we thought when we began the testing phase.
GreenGeeks Test
At first, our GreenGeeks shared hosting plan performed admirably. With up to six simultaneous visitors, our website loaded in less than 0.30 seconds each time. However, that time quickly began to rise, and we were looking at loading times of over one minute once we moved past ten simultaneous users:
This is a clear example of a situation where our plan couldn’t handle the amount of traffic we were throwing at it. As you would expect, the results held true for further tests. We’ll show you all the results side by side in a minute, but for now, take a look at our results graph for the 50-users test:
Here, we’re seeing the same trend as before. The performance began to drop with more than six concurrent users. That’s not good news if your website experiences a spike in traffic.
A2 Hosting Test
Next up, we tested A2 Hosting, one of the best budget-friendly WordPress hosts. Our site took a little longer to load from the get-go, with times around of 0.80 seconds for the first few users. However, the website held up admirably as we added more traffic. Likewise, loading times only spiked on a couple of occasions:
The same trend held true for our 30-user test. In fact, we only experienced a single (albeit high) spike in loading time during that test:
We expected A2 Hosting to start flagging a bit once we moved to the 50-user mark. However, to our surprise, it was also able to handle that load without much trouble:
These results were a welcome surprise after the first test. So we moved on to the SiteGround test, to see if it held up as well.
SiteGround Test
In our first test with SiteGround, the initial loading times crept a bit above the one-second line. However, they quickly became stable at a lower point:
We didn’t have a single spike in traffic during the 15-user test. When we moved on to 30 users, there was a single instance when loading times rose a bit. Otherwise, our results were almost a flat line:
Finally, we expected at least a couple of spikes during the 50-user test, but the results were still consistent and fast:
As you can see, we didn’t experience a single dip in performance during the whole test.
In Summary
Before we talk more about our results, let’s put them all side by side. That way, you can more clearly see the difference in shared host performance. All times are listed in seconds:
GreenGeeks | A2 Hosting | SiteGround | |
---|---|---|---|
15 users min loading time | 0.15 | 0.69 | 0.55 |
15 users max loading time | 120 | 1.68 | 1.31 |
15 users average loading time | 29.8 | 0.87 | 0.61 |
30 users min loading time | 2.26 | 0.73 | 0.55 |
30 users max loading time | 120 | 7.47 | 1.60 |
30 users average loading time | 76.0 | 0.96 | 0.61 |
50 users min loading time | 0.15 | 0.68 | 0.55 |
50 users max loading time | 120 | 1.77 | 0.78 |
50 users average loading time | 62.0 | 0.84 | 0.60 |
We went into these tests expecting to see a clear difference in performance between hosting providers. However, we also expected the plans we chose to show a bit of strain under large amounts of traffic. Instead, we found that these days, shared hosting can handle more traffic than you might imagine.
This is good news, since it means that your shared hosting plan may be sufficient even as your site grows more popular. Overall, we can conclude that if you’re launching a new website, your choice of web host is more important than whatever plan you pick at first. You can always upgrade your plan to something better down the line, if the need arises.
Conclusion
Shared hosting gets a bad rap sometimes, but it can get you a long way on a budget. More importantly, your choice of web host will significantly impact the performance your plan offers. In some cases, spending more for a better provider will save you money over time. After all, you’ll be able to put off upgrading your plan for longer.
During our tests, we found that popular providers such as A2 Hosting and SiteGround offered excellent shared hosting performance. This held true even with up to 50 concurrent users. Smaller web hosts may also do a decent job, but we recommend sticking with well-known providers just to be safe.
Do you have a shared hosting plan, and how well has it held up over time? Share your experiences with us in the comments section below!
Article thumbnail image by Nobelus / shutterstock.com
Everybody has this doubt in mind..but there is no way to actually get some numbers. Every search would bring us to answers like it depends on the website content, the website optimization etc. I do understand that it depends on them, but having some avg number of how shared hosting performs is always great to know before actually selecting a plan.
Thanks for sharing this wonderful post.
Siteground is simply a beast! Couldn’t be much happier!
Thanks for this informative and useful article. What about bluehost? Any comments or remarks about this hosting recommended by wordpress.org…..?????
Hello Raza. Thanks for your question.
You can read our thoughts on BlueHost in our blog – WordPress Hosting: The Ultimate Guide to Choosing The Best Option.
You can read it via the following link:
https://www.elegantthemes.com/blog/resources/wordpress-hosting-the-ultimate-guide-to-choosing-the-best-option
Hope this helps.
This is fantastic. Thanks for sharing the results. I host on SiteGround so was pleased to see they performed really well.
I was a little disappointed that you didn’t actually answer the question you posed in the title. So if SG can handle 50 simultaneous users, where do they begin to fail. Not that I have this problem yet (I wish) but I always wonder if my sites will be able to handle a sudden spike in growth.
Thanks again for doing the hard work on this,
Ben
Hello Ben. Thanks for your question.
It is advisable to use a Content Delivering Network (CDN) service, as well as caching, to better handle spikes in traffic as and when they occur. Combining those two approaches should be more than enough to adequately handle a decent amount of traffic until you start receiving large amounts of site visits on daily basis. At this point, it would be advisable to upgrade to a Virtual Private Server (VPS), just in case.
You can find out more about caching and CDN services via the following links:
https://www.elegantthemes.com/blog/tips-tricks/best-cdn-services-for-wordpress-users
https://www.elegantthemes.com/blog/tips-tricks/how-to-manually-configure-wordpress-browser-caching
Hope this helps.
Can you test at 100 simultaneous users or more to see when actual failure occurs?
A friend for example had used a shared host for 18 years but only recently reached its limits because his website was serving 200,000 pages a day and 600 GB a day.
What are SiteGround’s actual limits?
John, Thanks for the informative article.Thanks for sharing hosting performance test result.
No problem. Thanks for your comment. 🙂
I read your post and I really like it, Thanks for sharing.useful information
You’re welcome, Nupur. 🙂
Someone should create a WordPress site with some fairly complex elements such as a large slider that uses video backgrounds, and make that a standard speed / load test site that ordinary people could deploy to a sub-domain on their own sites to compare these results. This way everyone could compare their own host to that of others in a controlled manner. Testing three major hosts is an admirable step, but without being able to test my own, I don’t know how it would stack up.
I do take issue with the recommendation to stick with the well-known hosting companies. I’ve been using the small but extremely customer-focused BigScoots hosting out of Chicago for some time and a lot of their customers are seeing big speed increases after switching to them. I switched because they are extremely fast and helpful in responding to support questions. I am not affiliated with them except I pay my money and get their service.
Conversely, I had a customer who is on the most well-known large hosting company, and making changes to his website took forever because their servers are so overloaded that they can’t handle it. I decided that if another customer insisted on using that company, I would charge them double to account for the truly massive amount of time that is wasted after each little change. When the customer contacted their tech support, their “fix” was always to try and talk him into a more expensive plan that is on a less crowded server.
To avoid getting sued I won’t mention the name of that company but their initials are, fittingly, GD.
A2Hosting’s biggest drawback is that it has a 30 process limit, and when you count PHP zombie processes that can take a long time to be cleaned up, it’s extremely easy for just about any traffic spike to start resulting in error pages about resource limitations being hit or even for the DB to be too slow to respond, resulting in random 404s for pages that do exist. A2 just isn’t reliable for anything more than a personal blog.
I run a small hosting business and one of the major issues I find is that 90% of the traffic to most WordPress sites is actually inbound bruit force attacks. That is no exaggeration at all, the mere fact that your site is WordPress is a massive honeypot to hackers and in many cases, poor site performance is almost entirely the result of continuous unwanted ‘spam’ traffic.
So one of the most important factors in performance is fending off these attacks. If your hosting company can do this before they hit your site, that will make a major difference. If not, then you need a firewall solution that will do so without much load. Generally, ‘in-app’ firewalls will work but they, too are subjected to load.
Hey John,
I had this very bad experience with Hostgator as my blog grew in traffic. The customer service didn’t help at all so I ran out of patience.
Helpful stats you have shared here. I’ll have to move over to to test the performance of my current host.
Do have a wonderful weekend ahead
Hi Enstine. Thanks for your comment. Best of luck with your current host. 🙂
Hi John,
I found SiteGround one of the best hostings out there. They provide the latest technology and in-house SuperCacher for WordPress and Joomla.
I use their GrowBig plan on my blog.
When its the right time to upgrade to personal or private hosting? Please assist me.
Hello Rupak. If your site is experiencing slow performance (in relation to your hosting plan) or excessive downtime, it may be time to upgrade to private hosting. If you’re experiencing a steady rise in site traffic, and require more space and resources, it may also be a good time to upgrade.
We recommend running a performance test for your site using Load Impact to gain an overview of its current performance:
https://loadimpact.com/
Hope this helps.
Once you’ve exhausted your efforts to improve performance application side, you may want to consult about scaling for resources.
This is the exact reason we developed a new kind of WordPress hosting with Convesio. Our hosting isolates your site in a container, and allows us to easily spin up a new instance of your site on another server if the server you are currently on starts to slow due to traffic. This also helps us isolate your site from database errors and other resource errors that can be common in shared hosting.
So, are you using the LXC (Linux Container) system?
Is this unlike Mediatemple’s Grid shared hosting? They allow scaling db, too. But they require actual migration of u need to move to a dedicated or scalable vps option.
I have used these same three hosts, and had very similar results. I find Siteground is a great performer/value for 99% of my small business clients. My own site was on GreenGeeks for years (I like their “green aspect” much more than their performance!) I finally moved over to Cloudways, a slightly different option. I am very happy with them, too.
Hi John,
Great article. Thanks for this! One question, what shared plans did you test from each provider? I ask because I know A2 offers their Turbo plan which has less users as well as more resources.
Thanks!
Hello Jared. Glad you found the article useful. We used the following plans.
A2 Hosting’s Turbo Plan: https://www.a2hosting.co.uk/web-hosting
SiteGround’s Grow Big plan: https://www.siteground.co.uk/shared-hosting-features.htm
GreenGeeks WordPress hosting plan: https://www.greengeeks.com/wordpress-hosting
Hope this helps.
Have started experimenting with SiteGround’s shared plans about half a year ago and was completely surprised by the performance and ease of use.
Now I’ve ported most of my websites to SiteGround and I get better results than from an expensive $90 / month package.