Web Synthesis Or WP Engine

Over the last couple of weeks I have been optimizing a number of websites and I ended up with a website hosted on Web Synthesis and another website hosted on WP Engine. This gave me some updated insight into how each of these “WordPress Optimized” Managed Hosts work.

As an optimization specialist it is very important for me – that I can efficiently carry out my work with the relevant level of access to alter configurations in order to achieve the highest possible score on Page Speed as well as ensuring the site loads as fast as possible within the confines of what is available to me.

This is the first time I have received work from two clients hosted on two competitors at the same time and so I am going to share my observations and experience of both hosting companies.

I will be looking at a number of key areas and some of them are standard expectations that every webmaster considers as well as the level of simplicity in regards to optimizing a website – which is something that I expect.

So let’s get started with some of the desirable features that concern all webmasters.

Note: I am not affiliated to either hosts so you will not find a single affiliate link in this post.

Pricing

While both companies have similar pricing structures – which has been mentioned before in similar comparisons, I felt that simply looking at price alone is not a realistic analysis – “it’s what you get for that price that counts”.

For this comparison I will be looking at the Professional plan as the price is very similar – $97 a month for Web Synthesis and $99 a month for WP Engine – so what do you get for your money.

Web Synthesis ($97)
WP Engine ($99)
Technical Specs
Number of Websites 2+ (up to 10) up to 10 WordPress Installs
Daily Traffic 20,000 Unlimited
Number of Visits Limited by Bandwidth 100,000
Monthly Bandwidth 650GB Unlimited
Local Storage 40GB 20GB
Memory 2GB N/A
Support
Daily Backup
Malware Scanning
Malware Removal Free Free
Firewall Unknown
SLA (Downtime Compensation) Unknown 5% off monthly bill per hour of DT
Telephone Support
24/7 Support
Managed Upgrades Unknown
One Click Restore Unknown
Migrations Fixed Fee Self Migrate or 3rd Party
Others
CDN Extra (3rd Party)
Brute Force Prevention Unknown
Uptime Management
Nginx
Realtime Monitoring Unknown
User Stats
Self Healing
SSL
Built in Cache Included with W3TC Pro
Marketing Tools

Both hosts have similar offers in terms of features. Features marked as unknown may change depending on who can verify them – so feel free to comment below on anything that I missed and please provide information regarding missing items.

The most notable point about “what you get for the price” is the number of installs allowed. I feel this is something that users will regard as an important factor when deciding which host to use.

Another major point is Storage Space – I feel that Web Synthesis has a more generous allocation of disk space “something I hope WP Engine will look into” as disk space is generally quite cheap.

The details above don’t necessarily give you the full picture – each host uses a different method to present their allocations and limitations within a fair usage policy so we will need to dig a little deeper before you can decide which one provides better value for money.

Storage Space

Storage space is a factor that every website owner needs to consider.

Web Synthesis offers 40GB of disk space and I think this is quite generous considering the price while WP Engine offers 20GB which I think is something they should certainly consider raising to match that of Web Synthesis.

Disk space is essentially the amount of storage space that you have to store files and content such as images, JavaScript, WordPress files etc…

If your site is going to contain a lot of Images then you obviously want more disk space. To give you an idea of how much 40GB and 20GB is in terms of storage space – let’s look at how many images you would be able to store.

Let’s assume that average Image on the web is 100kb.

Number of Images Required Disk Space
1 100KB
10 1MB
100 10MB
1000 100MB
10,000 1GB
100,000 10GB
200,000 20GB
400,000 40GB
1,000,000 100GB

Quite a lot of images right?

Given the above figures – it’s safe to assume that your not going to fill your disk space any time soon – assuming you follow standard web practices when uploading images to your site. I often see sites with images that are around 1mb – so when it comes to disk space allocation, it is important to remember to optimize your images to get the most out of your available space.

It is also important to note that WordPress will create (by default) 3 versions of every image you upload – more if you have custom image sizes as well.

Bandwidth

Web Synthesis allocates 650GB of Bandwidth per month compared to WP Engine who have unlimited bandwidth – but there is a monthly limit to the total number of visits (100,000 per month) so we will need to do some maths to work out which is a better deal.

For this test we shall assume the average page size to be 1.7MBPosts are often embedded with Images which increases page size and this number is the current web page size average according to HTTP Archive.

These values are approximate values for easier reading.

Bandwidth (MB) Bandwidth (GB) Page Loads
1.7 0.0017 1
17 0.017 10
170 0.17 100
1700 1.7 1000
17000 17 10,000
170000 170 100,000

So according to these metrics you would get a better deal in terms of bandwidth from Web Synthesis – but hold on there a second!

Read what it says – 100,000 visits / month

So how do you define a visit?

This is part of what WP Engine defines as a visit.

  1. When a human being first arrives on the site and loads the page, staying there for 31 seconds, that’s a visit.

  2. If that same human then clicks a link and sees another page, that’s not a new visit; that’s part of the same visit.

  3. If that same human doesn’t have cookies or JavaScript enabled, still all that should count as one visit.

  4. If that same human loads the site with a different browsers, that’s still not a new visit; that’s part of the same visit.

  5. If that same human bookmarks the site, then 11 days later comes back to the site, that is a new visit…

You can read the rest of the article here.

This changes everything – the metrics now are not as simple as before. As far as WP Engine is concerned once a user visits your site – their IP is logged and that user can return as many times in a day on any browsers and view as many pages as they want – and it still counts as one visit.

So with an average page size of 1.7MB a user would need to visit your site 3.8 times (approximately)  in order to match the bandwidth allocation provided by Web Synthesis.

But then the question is what if your site isn’t the average website?

What if your average page size is 2.2MB which I see commonly across the web – then you would need users to on average visit 2.95 pages per day.

The way that WP Engine counts visits is quite interesting. Another example is your house – if you have 3 computers and everyone visits a website hosted on WP Engine – it still counts as one visit. This also applies to offices or communal spaces where multiple users are connected to the same router sharing the same IP. Theoretically a thousand people could connect to the site from the same IP and it would still count as one visit.

Web Synthesis – although you get 650GB of bandwidth per month (Which is a more than enough for most) – you are limited to 20,000 visitors per day. Why this is the case I am not sure as you would usually be allowed as many visitors in a day as long as you remain within your Bandwidth allocation.

Don’t get me wrong, 20,000 visits per day is a lot! But my question would be what happens if your website is to be featured on a  very popular website read by millions. 20,000 visits no longer seems out of reach – so would users no longer be able to reach your website after that point?

If so that would certainly have an affect on those who tried to visit your site – “first impressions count” right?

Bandwidth Conclusion

I think that both Hosts offer a reasonable amount of Bandwidth and both certainly fall within what should be expected for the vast majority of websites. You can vastly improve your Bandwidth allowance by optimizing your website.

I optimize websites on an almost daily basis and in almost all cases I can reduce page size by 50% or more.

Each host (for the less than a $100 price mark) offers a good amount of resource availability and and each has their pros and cons – but you can rest assure that most of you will not exceed your limitations each month.

Security

Both WP Engine and Web Synthesis offer security as a part of their package. How effective their measures are can only be summed up by long terms users of either host. However I do know that WP Engine will fix your site if it is hacked which is always reassuring. If Web Synthesis offers the same, please let me know below.

WP Engine does have a firewall which blocks all known attacks.

Both hosts scan for malware but Web Synthesis includes Securi as part of their packages.

If you have experienced a malicious attack on your site on either host please feel free to comment on how each host dealt with your situation.

Support

There isn’t really much to say here – from reviews and testimonials, both hosts seem to provide excellent support to their clients and that is a very important factor.

Performance & Efficiency

So far each host appears to be about the same in terms of what the user can expect. Things get a little more interesting when we begin to look at performance and efficiency.

Those who visit my website often know that I am an avid inciter of improving performance and so for me – this is the most interesting part of my comparison. You can learn a lot about a host by how they configure their servers and believe me I was surprised by the results.

The best way for me to look at the performance of each host is to look at their clients testimonials and test their sites.

The conditions for testing are as follows:

  • Tests will be carried out on the Front Page
  • The sites Front Page must be devoid of adverts (they screw up results)
  • The sites Front Page must be devoid of social sharing (they also screw up results)
  • Each test will be performed from Vancouver, Canada – this will test performance at a distance.

I will test 4 sites from each host.

Web Synthesis

jaybaer.com

jaybaer-report

thetomkatstudio.com

thetomkatstudio

pioneerclubs.com

pioneerclubs

prettydarncute.com

prettydarncute

WP Engine

bourncreative.com

bourncreative

wpmudev.org

wpmudev

wp101.com

wp101

soundcloud.com

soundcloud

Although performance and efficiency is largely down to the user – there are some aspects of performance that are dependent on the servers configuration and there are a couple of issues that I noticed with Web Synthesis.

The two issues that cropped up time and time again on websites hosted with Web Synthesis were:

  • Specify a Vary: Accept-Encoding header
  • Gzip compression

It’s no coincidence that the two are related because they are part of the same configuration.

Specify a Vary: Accept-Encoding header & Gzip Compression

The Vary Accept-Encoding header is missing from Web Synthesis and I know this because I looked at the core configuration file to see what was in there.

gzip  on;
gzip_disable "MSIE [1-6]\.(?!.*SV1)";

This is the bare minimum required to enable Gzip compression as well as disable it for older Internet Explorer browsers that don’t support compression.

What was striking to me was that according to the back-end Gzip Compression is enabled – yet it does not appear to be working.

Prettydarncute.com could save 271.5KiB if Gzip Compression was working.

pretty darn cute gzip compression

I found this issue to effect most of the sites I tested from Web Synthesis – which means that each site will consume more bandwidth than it needs to, after all compression is about saving bandwidth by sending compressed resources across the web rather than the standard uncompressed version.

Things got even stranger when I looked at the headers.

main css header with gzip

Figure I – this response clearly shows that I was sent a compressed version of the file.

Figure II - This header shows that I was sent an uncompressed version of the same resources.

Figure II – This header shows that I was sent an uncompressed version of the same resources.

What in the hell is going on?

Here is what we know:

  • Gzip is turned on – according to the config
  • The First header (Figure I) clearly shows that the main style sheet was sent compressed
  • The second (Figure II) shows it was sent uncompressed

At first I thought – perhaps GTmetrix is doing a number on me and flagging false negatives – after all it does that for serving scaled images on websites wider than 1024px.

So I checked Page Speed.

pagespeed results for pretty darn cute

Page Speed only confirmed what GTmetrix was saying.

Basically the server is being a little picky about whether it should send compressed or uncompressed resources – but why is this the case?

Well it is fairly simple-ish…

Simply enabling Gzip compression is only one half of the story – the clients browser must also ask for it when making a request via the Vary Accept-Encoding Header. If this is not included in the servers configuration file then you will get these random responses in regards to whether the server will send resources compressed or not.

The precise solution for Web Synthesis is not something I can write about because I would need to know all the details in regards to how the servers are set up – but I do know that the configuration is not correct and should be rectified as soon as possible.

I pondered why Web Synthesis didn’t notice this issue and the only logical answer I can give is that by default gzip_vary is disabled.

According to the documents from Nginx they say gzip_vary

Enables response header of “Vary: Accept-Encoding”. Note that this header causes IE 4-6 not to cache the content due to a bug.

Perhaps this bug is the underlying reason, but that doesn’t make sense because IE 4 – 6 are all but dead (if not dead already).

But they include

gzip_disable "MSIE [1-6]\.(?!.*SV1)";

Which is basically saying disable Gzip compression for IE 1 – 6 as well as Service Pack 1 so that doesn’t make sense either.

Why it isn’t configured correctly isn’t the most important thing – what’s important is that it gets corrected.

Conclusion

As far as I can tell both hosts provide a Good service – but I am inclined to say that at the time of writing WP Engine seems to be the better of the two.

The built in CDN and Caching as well as the server being configured very well makes my job of Optimizing websites hosted on WP Engine a whole lot easier as i don’t need to configure the server, caching or CDN. That means I can focus solely on the front of the site ensuring the client gets the very best results within a reasonable amount of time.

Unfortunately I cannot say the same for Web Synthesis at the moment – the simple mistake of misconfiguration with the possibility that users may of been charged (couldn’t find an over usage policy) for over usage when they could have been within their limits is simply not acceptable to me.

Looking at the 4 sites I tested above you can see the impact of this misconfiguration on efficiency in general.

Before anyone asks – I am aware that Yoasts website is very efficient and loads fast on Web Synthesis – Yoast values the importance of an optimized site and so I would not expect anything less than high performance from him. His site also uses a CDN so his resources are coming from the CDN and not the server – which explains why his site isn’t flagged for the issue raised above.

A final point

Web Synthesis does not allow Thesis 2 on their servers. I am not aware of any technical reasons for this and their site doesn’t clearly state any reason for this exclusion which leads to people simply guessing. I would assume it to be a marketing ploy and a means to promote Genesis over Thesis 2 which would make sense because Web Synthesis and Genesis have the same owners.

If this is true – then I would never be able to recommend Web Synthesis as most of my client use Thesis 2 – but from a personal point of view I find this to be reminiscent of the days of AMD and Intel – where Intel attempted to aggressively market AMD out of the market and it almost worked. If this is the case then the only people that would suffer is the end user. Market domination is never a good thing as it often leads to a slowdown of innovation. Competition is healthy and good for you and me – it allows for a steady stream of progression towards better ideas at affordable prices.

The statement on Web Synthesis:

(Note: Synthesis does not currently support Thesis 2.0+. We do support Thesis versions prior to the 2.0 release.)

Implies to the unknowing user that something is wrong with Thesis 2 and that is why we can’t allow it on our servers. I think that this is unfair and it is just one more reason why I would choose WP Engine over Web Synthesis as WP Engine doesn’t limit you in any way when considering which Theme/Framework to use.

I firmly believe that hosting companies should never interfere in another market for there own gain. It is not for one market to determine who is the top dog of another market – no matter who owns what.

Meet the Author

Matthew Horne

Matthew Horne is web developer who specializes in optimized development. He also builds custom solutions instead of reverting to plugins. Matthew Has a strong understanding of PHP, JavaScript, jQuery.

21 comments… add one
  • Pedro Lima Dec 26, 2013, 3:50 pm

    I have to say that I am a bit surprised with that issue of Web Syn(anti)thesis. I have been a big fan of Genesis and it is a really good platform but using this dirty tricks its not only unfair but a shot in the foot. Thanks for flagging that up Matt!

    • Matthew Horne Dec 27, 2013, 3:30 am

      It surprised me when I first heard about it, it appears to surprise anyone who knows about it.

      The practice of exclusion is something not permitted in Europe and my American colleagues seem to dislike it too.

      • Pedro Lima Dec 27, 2013, 5:14 pm

        You might find this interesting after I inquired with the man himself, the genesis mastermind and the man behind Web Synthesis, Brian Gardner (@bgardner):

        http://pedroli.ma/twitter.jpg

        I wonder now what explanation is Web Synthesis preparing about this?

  • Chris Langille Dec 27, 2013, 1:05 am

    Awesome review man! I can’t wait for you to optimize my sites!

  • Cam Dec 27, 2013, 5:05 am

    Hey Matt,

    This is probably the first real word non-biased and factual comparison that I have seen between these two premier hosting solutions.

    Very nice write up and I enjoyed the read immensely!

    I recently after 4 years of using HostGator said adios to them due to the spiraling downward trend that they have taken since being acquired by EIG .

    I have always wanted to use WP-Engine but they are a tad rich for my blood and like you I prefer to be hands on.

    I ended up going with Site5 and so far, I couldn’t be happier with the speed of the server and being able to choose your data center location is a big plus when it comes to optimization and localized S.E.O.

    How is life in the server optimization arena been treating you? Keeping busy Matt?

    Really liked the article man.

    Cam

    • Matthew Horne Dec 27, 2013, 5:19 am

      Thanks man, Yeh Hostgator went downhill fast since being acquired by EIG (same group that owns Bluehost and about 70 other companies), was a shame because they were good.

      Server Optimization and Site Optimization is going well, but there is alot of work to be done – It wouldn’t be to farfetched to say that 99% of websites are in need of some tuning. The recent pingdom stats highlighted just how much wasted bandwidth there is.

      29.9 years of time is wasted on waiting for websites to load everyday according to pingdoms annual stats. That’s pretty amazing if you ask me and that only includes sites that were tested on Pingdom – so the total amount of time is unknown.

      • Cam Dec 27, 2013, 5:29 am

        Yeah I know it’s a pathetic situation out there considering how easy it is to maintain correct server load times and reduce HTTP request by just being conscious of what you install on your website and upload to your server.

        Sad that most people wont take the time to educate themselves about maintaining their site correctly.

        forget the waiting times accrued and lost on a yearly basis, lets look at the bigger picture and assess what kind of impact this has on the environment due to excessive and squandered power resources.

        Cam

        • Cam Dec 27, 2013, 5:32 am

          Hey Matt,

          I have a couple of questions that I would like to pick your brain about but it is off topic. What is the best way to hit you up to ask what I have in mind?

          Cam

  • Jerod Morris Dec 27, 2013, 4:35 pm

    Matthew,

    Thank you for taking the time and effort to further the online dialogue about WordPress-optimized managed hosting. While we appreciate your detailed post, and commend your method of data collection, there are quite a few assumptions and missing facts. We are always open to a phone call to provide further clarification if you are interested in providing a more factual follow-up to this post.

    In short, a few comments are:

    — Hard drives at MicroCenter are cheap … but managed, RAID10 storage hooked to gigabit connectivity is the single most expensive element of a good stack.

    — Pageviews and bandwidth cannot really be compared, and we do not have hard limits for either, just guidelines. Our goal at Synthesis is that no customer is charged for overages. We don’t believe in penalizing content publishers for having success with their site. As explained on the Plans & Pricing page, “We won’t throttle or block access to your site based on traffic, but do know that if your site consistently and flagrantly exceeds the data transfer allocation for your plan, you may be assessed overage charges ($0.15/GB).”

    — Gzip is one tool, one element to look at. We solve for problems and deliver outstanding performance while striving for high uptime, compatibility, etc. To do this, we’ve chosen conservative base settings but do provide for customization. This is why tuners generally love us, because we provide root access so they can customize to their liking. You can read more about our philosophy on site performance here: http://copy.bz/19HF1Wd … The key is that JayBaer.com, for example, has a sub-1.0 second load time.

    — We fully support session variables in ALL of our cache offerings as well as full support for mobile caching. That is a big comparison point you failed to mention and the reason several top reference customers have left our competitors and chosen Synthesis. Others have left due to pageview-based overages.

    — Your conclusion on our support for Thesis is incorrect. Discouraging Thesis 2.0 is not a slight toward DIY’s work nor their technology choices, as we do allow previous Thesis versions. Rather, our position with Thesis 2.0 is a business decision that derives from the work it takes for us to really certify and understand a theme framework, plus the overall size of the market share of DIY versus Woo, our own Genesis, and others. We just haven’t had the traction to justify it for Thesis versions 2.0 and above. If we were going to use evil marketing tactics, we’d disallow Woo Themes. 🙂 As you can see, we fully support Woo and enjoy a good relationship with their support teams.

    — We do take full responsibility for cleaning/fixing any infected site.

    And finally, a few corrections for your chart:

    — Our Professional Plans support up to 10 WordPress installs, not five. The first two are included. Each additional install beyond that costs $5/month to cover daily backups and malware scanning.

    — We do not provide phone support for Professional Plan customers. What we do provide is emergency response times within a guaranteed window and 24/7/365 monitoring of our support Help Desk by full-time employees who are both WordPress and hosting experts.

    Thank you for providing a forum for clarification here in the comments. Again, if you would like to chat to clear up any other questions you have, just shoot me an email and we’ll happy to. Happy holidays everyone!

    Jerod Morris
    Director of Content, Copyblogger Media
    Co-founder of Synthesis

    • Matthew Horne Dec 28, 2013, 4:23 am

      Hi Jerod, Thank you for taking the time to read.

      Please note my intentions are not to slate your service in anyway as I found your service through reviews and through clients to be good – with few complaints.

      Ref – Gzip compression – It is enabled by default, my point was that it is enabled but without the Vary Accept-Encoding Header it might as well not be as the site would be sent (randomly) compressed or uncompressed. It is either on or off but at the moment it is somewhere in between.

      I understand your support is very good as is with WP Engine.

      Ref – Thesis 2 – my point here wasn’t to state that you are playing dirty – It would be more along the lines that simple making a note is very much open to debate and I speculated what I thought some would think and indeed because I have been asked about it before. Questions like – is there something wrong with Thesis 2, because Web Synthesis don’t support them. Others have stated marketing reasons etc…

      The point there is that you should clarify – especially with the history between Brian and Chris. That makes for even more random assumptions. I would like to see further clarification as to why Thesis isn’t supported and that way people cannot assume if it is written down in an FAQ or something.

      Hopefully this post will help to create a greater sense of clarification so that the tides of past don’t get mixed in with the real reasons. I wrote this post in response to precisely the lack of details around the decisions to exclude Thesis 2 but also because some of my clients sites who are hosted with you lacked Gzip and I noticed this was something that was present on almost all sites.

      If Gzip is to be optional then I think for those who are less technical like my client (who would not of known about it) would perhaps benefit from an option to enable it when signing up or form their dashboard or something. I like the fact you give users choice, but unfortunately not everyone has the technical knowledge to change configurations – and in the case of those who have never configured a server, they likely wont know about it anyway.

      I left some details open to debate so I will change the number of installs. The reason I assumed up to 5 is because that is what I could see from the plans and pricing. 2+ with the next plan being 5+. So I think some clarification may be needed there too.

      The post is written as a general users from what can be ascertained from what can be seen. If I contacted you before hand – then it would not of gotten the point across about clarification.

      All the load tests were to Canada to include some distance as fast load times would be expected if the test location is close to the server. Speed is only one piece of the picture – Efficiency is another. Google expects compression to be enabled and for the correct headers to be included.

      I hope you understand I am not trying to harm you in anyway – I just want greater clarification.

      Happy Holidays and I’m open to discussion.

      Regards

      • Derick Schaefer Dec 29, 2013, 4:16 pm

        Matthew,

        I appreciate your detailed and data driven approach.

        I think your question around Thesis could have actually been positioned in the form of a section on restrictions and policies–this would have been of great value to your readers. Like our competitors, we have a number of policies driven by both technological and business criteria. Our policies on Thesis are business related driven by our desire to exceed customer support expectations. We have similar policies for multisite. On the flip side, if a customer brings in Thesis 2.0, we will not remove their site. We do have some Thesis 2.0 implementations hosted with us. Still, we do not want to set expectations we provide incredible support for it as the truth is we haven’t invested in it. From a strengths perspective, we do not remove or disable plugins automatically. We do allow Linux cron and SSH access which is well received by our customer base.

        Your performance review is detailed and I commend you taking the time to gather the data. Since the beginning we’ve taken the approach of using data to solve problems. We aren’t big on generic grades; the topic actually inspired this white paper http://copy.bz/19HF1Wd . After all, your first example of ours is a website that has sub second load times.

        Varys is topic we’ve taken a conservative approach on in our default settings. As explained here http://stackoverflow.com/questions/1975416/trying-to-understand-the-vary-http-header , varys is designed to facilitate upstream proxy and cache server communications–not browsers. We don’t use upstream caching proxies by default. Still, even Frederick and I share differing opinions on this topic. In the end, it is one of many levers available to a site owner or tuner; it is not “the determining factor” in site performance. We do change this default on a site by site basis. There are also risks involved with aggressive gzipping in today’s widening browser landscape.

        Last, we use CDNs as a tool and not as a marketing ploy that insinuates that it will make a site automatically faster. The majority of CDN implementations that come in our door slow site performance when compared to our native ability to deliver static objects. A minority of sites see improvements. As content publishers, however, our largest concerns are image URL changes and SEO. We were the only managed WordPress host left off of this exact criticism in a present at WCSF this ( http://wordpress.tv/2013/07/28/mark-jaquith-confident-commits-delightful-deploys-2/ minute 7:05) . We do think these topics through.

        Thanks for taking the time in including us in your research.

        • Matthew Horne Dec 30, 2013, 6:01 am

          Hi Derek, Thanks for the reply.

          The point about Thesis 2 was made because of questions I had recieved asking me why Synthesis doesn’t support Thesis. I could not answer them in a meaningful way. One person asked if there was something wrong with Thesis 2 – this inference lead me to wonder whether others thought the same thing.

          Because the point about Thesis 2 is simple a note on your site, it does not clarify and therefor it is open to debate and likelihood of it being interpreted the way you intended is quite slim. For that reason I would ask of you to ensure clarity on that issue as it could inadvertently be seen in the wrong way.

          Ref Vary – The article linked is quite old now and of the 100s of sites I have optimized this year I haven’t come across any issues related to including the vary “accept-encoding” header for gzip.

          The point is Gzip is enabled on your servers, but without that header, there remains the possibility that a user will download resources for a site twice. This is why I received both non compressed and compressed headers in my findings. If your user uses cache, then there must be away to determine whether the user has already stored a cached version of the resource.

          Without Vary “Accept-Encoding” Header the user may re-download the same resource again and it can produce a random response as to whether they get a compressed or uncompressed version of that same resource.

          You can also specify the type of resources to gzip.

          I agree with you that there is no one size fits all when it comes to optimizing a website and server. Every site is unique and requires its own attention.

          CDNs have benefits, I live 7000km away from my server. The nearest Edge server is about 400km away, there is a big difference for me when I turn my CDN off. Also CDNs allow you take advantage of parallel loading. As you state in white paper linked above. A browser can load 6 resources at the same time. However for a site with lots of images, 6 resources may not be enough, With a CDN you can increase this to 24 by making 4 simultaneous connections with each connection capable of downloading 6 resources in parallel.

          That’s not to say they don’t have their downsides.

          It all depends on the users requirements. Who are they tailoring for etc. Of course a CDN wont automatically make your site faster, but it depends on where your user base is relative to your server.

          You say Jay Baer site is sub 1 second. But to Canada it is not. Outside the US it is not, a CDN could help speed up the site to those users outside the US or those inside the US who are further away from the server.

          Speed is one thing, Efficiency is another. I think we share many points and seek similar goals. Its my job to help people further understand the importance of efficiency and load times as they do play a part in the success of a website.

          Regards

    • Thomas Zickell Feb 10, 2014, 7:28 pm

      as a customer of WP engine, WebSynthesis, GetFlywheel, Pagely, PressLabs & Pressable
      I can tell you that web synthesis is a great hosting company. However there very slow to respond. In fact today I had a question for them I asked around 9 AM I’m still waiting for an answer, and this is to actually give them more business.

      The issue with ability not to host Thesis 2.0 I believe stems from a much bigger problem than just them not wanting to. I having a lot of experience with managed WordPress know that Thesis 2.0 causes issues on WP engine as well as other hosts listed. Why because of permissions and vulnerabilities that it creates with 777s. It’s a hassle

      I have no biased whatsoever.
      As far is a content delivery network vs a website without a CDN well of course they’re in a be hosts that are better than others, and I do believe that having a hosting company that knows how to cache content so that the site is faster is obviously going to be a huge plus. However WP engine and every host above knows how to cache the idea of copy blogger using net DNA and very fast sites like Yoast prove content delivery networks can add great amounts of speed.
      I believe that Yoast & copyblogger are hosted on their enterprise offering
      the idea that a system like edge cast’s ADN would slow down there sites is ridiculous.
      However the person who buys hosting at blue host and then believes they can slap the CDN on it is insane if not fixing the problem it’s putting a Band-Aid on it.

      WP engine has a better caching system in my opinion.

      However both hosting companies are excellent and very much above par I would recommend using whichever one you prefer.

      However, I have had service with both of these companies and everyone above that I’ve mentioned for over 6 months at least.

      You really can’t go wrong picking either one.

      I wish that when you sent in a ticket to synthesis that it was actually replied to within a short amount of time. Something that is done at WP engine and GetFlywheel I don’t blame the guys and synthesis at think it’s hard to get people with that type of expertise still synthesis might want to spend money on hiring more customer service people.

      You’ll be happy with any of the choices above.

      • Enrico L'anglais Feb 15, 2014, 1:08 am

        “However I do know that WP Engine will fix your site if it is hacked which is always reassuring. If Web Synthesis offers the same, please let me know below.”

        With Web Synthesis you have to PAY! It’s not free like WpEngine!

        • Enrico Feb 15, 2014, 1:17 am

          Also, about their support, you can notice that wpengine has a LIVECHAT 24/7/365 which is definitely better!

        • Matthew Horne Feb 15, 2014, 5:56 am

          Ok thanks for the info, I wasn’t sure whether you had to pay or not.

        • Tiago Hillebrandt Feb 27, 2014, 2:37 pm

          Hi Enrico,

          At Synthesis, we have Sucuri.net doing server side scanning every site we host several times a day. Sucuri.net is an industry leader for site scanning and malware removal, and we are glad to have them on board! If infected or exploited files are found on your site we will get them cleaned up.

          All of this is handled on our side, at no cost.

          Regards,

          Tiago Hillebrandt

  • Dee Mar 14, 2014, 11:23 pm

    Hi Matthew

    Great comparison. I’ve been doing some research as to which one to choose and this made my decision a bit easier. WP Engine it is.

    Thanks

    Dee

Leave a Comment