Cancel

Jeff Chandler

Frederick Townes Confirms W3 Total Cache is Not Abandoned

W3 Total Cache is a free, caching WordPress plugin created by Frederick Townes that’s active on more than one million sites. While the plugin’s core functionality is free to use, there’s a handful of services users can buy from within the plugin such as enhanced support to server and site configurations.

W3 Total Cache Upsells
W3 Total Cache Upsells

A recent post by Scott Tuchman on the Advanced WordPress Facebook group cites that the plugin hasn’t been updated in more than seven months and is not compatible with WordPress 4.4.2. Out of 44 reports, 34 people say W3 Total Cache 0.9.4.1 doesn’t work with WordPress 4.4.2.

W3 Total Cache Not Compatible With WordPress 4.4.2
W3 Total Cache Not Compatible With WordPress 4.4.2

The plugin was recently updated with the only change being the readme.txt file which states it has been tested up to WordPress 4.5. A cursory glance at the support forums indicates some users are not pleased with the simple update.

A Rocky Year

In 2015, a disgruntled customer who purchased services from W3 Edge, described their negative experience in a post on the WordPress subreddit. The complaints include, a lack of communication, not receiving purchased services, and project delays. Mike McAlister, of Array Themes, told the Tavern about his experience with W3 Edge.

In 2014, I signed up for the premium version of W3 Total Cache to unlock some of the advanced caching features. The email confirmation said that the license would auto-renew unless cancelled, so I contacted them right away with my request to cancel the auto-renewal. I didn’t hear anything back and wrongly assumed this was taken care of.

Fast forward one year and I get an email out of nowhere that my W3 license had been renewed. I immediately contacted W3 with details of my order and politely requested a solution, but never heard a response.

I wrote a total of four emails and several tweets directed at W3 Edge and its founder, Frederick Townes, with not a single response throughout the ordeal. Like many others, I had to resort to calling the bank and filing a dispute to get my money back, which dragged the issue out for several months.

At the end of the Reddit thread, W3 Edge responded to the criticism and admitted that the company let users down with its lack of engagement. It also admitted to having a small team and the challenges involved with its support model.

“One of the largest lessons learned is that there’s a lot to get right especially with a small team,” W3 Edge said. “We’ve made some changes to the team recently and we’re working to rectify experiences like the ones noted here. Since it’s the customer experience that matters and not intentions, I apologize for the negative experiences. Our mission remains to enable independent publishers with great software and solutions.”

W3 Total Cache Status Update

The company’s official Twitter account has been inactive since April of 2015 and the last post published on its blog was 10 months ago. Combined with the lack of plugin updates, it’s understandable why customers and users are concerned with the plugin’s status.

Frederick Townes, founder of W3 Edge, issued the following statement to the Tavern.

Since the last update, development and other operations have been ongoing. There have been several hundred bug fixes based on user feedback, more than 100 improvements and numerous major improvements. We’ve added tens of thousands of unit tests for the various bugs and improvements in an attempt to allow us to release more updates faster in the future.

We’re also working on a new services and support model, which is more exclusive in an attempt to reduce the volume of service requests and misaligned customer / subscriber expectations.

Our goal is unchanged, to create easy ways for publishers and developers to increase WordPress performance and tune user-experience in self-hosted environments. As we work towards leaving beta and moving towards a 1.0 release (and making our GitHub repository public rather than private) our goal is to offer hosted services that use automation to simplify the performance optimization process in the future.

I spoke to Townes for an hour on Skype and he comes across as a sincere, concerned business owner who is struggling to find a system that works for the amount of support volume he and his team are dealing with. I asked what keeps him motivated to continue working on W3 Total Cache, “I love solving complex problems which is what W3 Total Cache does,” Townes replied.

I can only imagine how heavy the burdens are maintaining a free plugin that’s active on more than a million sites. Joost de Valk, founder of Yoast.com, and lead developer of WordPress SEO, a plugin active on more than one million sites shared some of the lessons he learned after users experienced issues upgrading to 3.0.

Have The Changes Worked?

While researching for this article, I requested feedback from thousands of people to learn what recent customers are experiencing with the company but no one responded. This can mean any number of things but it’s difficult to determine whether the changes mentioned in the Reddit thread have improved the situation without getting feedback from recent customers.

If you’ve recently purchased services from W3 Edge, please share your experience with us in the comments.

Category:
Tags: ,
64

64 responses to “Frederick Townes Confirms W3 Total Cache is Not Abandoned”

    • If you mean WP Super Cache, that is really a different thing.
      WP Super Cache only does caching of html pages and consists of maybe 10.000 LOC iirc.
      W3 Total Cache does a ton of extra stuff, and is closer to 100.000 LOC iirc.
      If you just need a page cache WPSC probably fits your needs. If you need al the other stuff, W3TC is what you are looking for.

      • Correct. that’s the problem, switching to WPSC will not replace W3TC. There is nothing out there anywhere near as good (in my research thus far) so it’s not like there is an alternative…

  3. That’s good news. I hope that they are going to finally implement a mechanism for dynamic caching (something like the modular system provided by Comet Cache). That feature is really useful in many cases, without it the caching logic is rigid.

  4. I dropped W3TC a while back after many issues with multisite WordPress, and switched to WP Rocket. W3TC got to complicated and very hard to configure, and I understand how it can be a problem to support. Plugin needs to go through some very complex changes to be made usable and easy to use.

  5. It’s sad, that if the developer of WP plugin with million + active installs clearly scammed a user (I remember well when Mike tweeted about that issue) … nobody care.
    You Jeff, asked for feedback … almost nobody care.
    If Automattic does something helpful for users, everywhere pop-ups conspiracy stories, if someone spend a time to build something new and try to implement that to the WP core (like emojis), again a lot of negative feedbacks.
    Somewhere on the internet some polls what developers prefer to work with … Drama with a lot of noise again.
    But if it’s something related to millions of WP sites, the audience is quiet.
    This approach of WP users is poor, and sooner or later the results of this will come back like a boomerang.

    • @Peter Cralen:
      You know emojis in itself are just great, but they should never have been implemented into CORE. Why? At least 98% of active WP sites simply do not use them, still it occupies space and some milliseconds will be wasted in vain. That is a trade-off for nothing for the majority. That is why emojis should have been left as a plugin.

        • Well…it is not about me but about the majority. Being that said let us not hijake this topic :)

      • FYI, The emoji implementation was also a way to quietly patch a massive security vulnerability.

      • In addition to the security fix Sam alluded to, adding utf8mb4 (emoji) support also enabled WordPress to handle a bunch of Unicode character sets that many non-English languages rely on.

        In a time when the number non-English downloads is higher than English ones, this enhancement was vital and long overdue. Yeah, it was primarily promoted as emoji support but there was a lot more to it :-)

        • Drew, maybe you are confusing two related things that are related only because they were “sold” together. I want to believe that no one complains about the change in the encoding of the table, the complains are toward the shim that is the worst type of hack, it relies on an non standard quirk and unsurprisingly proved to be both unreliable and buggy. It tries to implement something that a browser (chrom) do not support which is stupid unless core want to start contribute to browser development as well.

          For years web developers were having a really bad time supporting IE6 and core never suggested to, for example, autoload some JS to solve alpha channel problems with PNG images, so if there was a change in policy maybe core should include more useful polifills that people actually care about.
          If the policy was that core includes polifills then another one to support imoji would have been less of an issue, but selecting to support polifill no one wants but no other? I don’t think anyone can actually figure out the reason for that which is probably part of the reason people keep ranting about it.

      • agreed – that’s why this has become default in every site I now build:

        /*= Remove Emoji Support
        **********************************************************/
        remove_action( ‘wp_head’, ‘print_emoji_detection_script’, 7 );
        remove_action( ‘wp_print_styles’, ‘print_emoji_styles’ );

      • hmmm, and there is some bug with the comments, I thought to reply to Peter and changed my mind but there was no link to get the comment box to the min thread, and a page refresh seemed to help, but obviously didn’t

  6. W3TC may as well be abandoned ever since Townes ignorantly declared that APC was the future of caching and that “he wouldn’t be supporting Opcache” no matter what. One of the buggiest, most mis-managed, and most fraudulent projects in the history of WordPress that has amazingly been the recipient of undending praise and inexplicable adoration, if only because of the vast misunderstanding most webmasters have when it comes to caching and website performance…

  7. We’ve been working with W3TC team over the last years to maintain compatibility with WPML. There are times when response to our messages is faster and there are times when it’s slower, but I can confirm that W3TC looks like a live and kicking project.

    I hope that the team will find a model that allows to cope with the impressive amount of support, while keeping resources for development. I know what it means to support an audience of this size and I can tell you that it’s a very demanding process. I wouldn’t know how to do this without a reliable revenue source, that allows to employ a sufficient team.

    There are different ways to run technical support for a popular product. If this is interesting to folks, I can write a blog post that explains how WPML’s support and compatibility teams work. There’s a chance that this can help other developers, who’s products are becoming popular. It’s a lot more complex than what appears on the outside.

  9. Jeff, re: the lack of feedback from users ~ I could share some things that would really shed some transparency on certain aspects of W3TC’s paid service & the relationship of some of the finer interactions w/ MaxCDN and W3TC, …and I’m sure a lot of other folks could, as well, share stories similar to the ones you’ve linked to on reddit.

    But, since there’s been >”millions” of other users who didn’t need the higher level features, perhaps didn’t use some of W3TC’s features, and never paid a penny, where would that leave those of us if we **did** share our experience?

    Although I respect that it’s not technically “abandoned”, very much like some people’s experiences, I think it’s time to leave it alone and simply move on ;-)

  10. While researching for this article, I requested feedback from thousands of people to learn what recent customers are experiencing with the company but no one responded.

    I’m curious where you solicited this feedback from? Was it only for w3 pro users?

    • I solicited the feedback on the Advanced Facebook WordPress group which has nearly 20k members and I also put out the call on Twitter which is another couple thousand people. Both of which are highly focused around WordPress. Here’s an example of my request: If you’ve subscribed to the W3 Total Cache premium service or purchased a service from them in the last three months, please get in touch I figured I would have gotten a response or two within 24 hours but it wasn’t the case.

      • Funnily enough, I’m a member of AWP and follow you on Twitter. I saw the tweet, but never saw the post on the FB group. I *really* wish there was another forum other than the FB group; seems posts get lost in the shuffle a lot there. Maybe for future requests/announcements you could reach out to an admin and ask them to pin it to the top of the discussion threads in the group.

  11. I get it. Scaling is hard, but if you can’t handle your paying customers you hire more people. If can’t afford to hire more people, you’re not charging enough. There’s a bit more to scaling than that, but it’s not as hard as they make it out to be. Take care of the paying customers first. Don’t ignore the people that put food on your table. Period. Any excuses are just that. Excuses. This has been going on for a long time now. It’s dishonest, and ignoring people who’ve paid you money and entered into a contract with you is simply wrong, and likely illegal. It’s clear they’ve moved on mentally, and that’s fine. But stop taking money from folks who trust you enough to hand over their wallets. Rip out the ads, shut down the payment forms, and go on with your life. Stop stealing from honest people.

  12. I switched my sites and client sites back to Super Cache a while back. When W3TC works, it works great however the config can be bewildering. I also occasionally had really strange errors with it.

    Tech support was indeed slow sometimes. I got the feeling Frederick Townes was a bit disconnected from the support process and also, as mentioned, really did want to provide a quality product.

  13. I, too, had to file a dispute with my credit card company to get a refund on service that I did not wish to renew. Being totally unresponsive to customers is simply not acceptable.

  14. I’m glad to see Frederick wants to do better, but this reply does not reflect the seriousness of the situation. Doing better in the future will surely benefit everyone and I wish him the best, but what about all of the people, myself and peers included, who have paid good money ($99-$200+) for services W3 promoted through the WordPress.org repo and got absolutely nothing? No premium service, not one reply, no refunds, no answers — just headaches and bank disputes.

    After looking through reviews of W3 Total Cache on WordPress.org, it’s clear that W3 has been accepting money for premium services and not delivering since 2014. There has been ample time for this team to grow along with the plugin’s success, and if not, why did they continue taking payments for premium services and ignoring customers? Do we simply write these off as growing pains?

    I would hate to see this issue painted as drama because this is a very serious issue. This is clearly a plugin people still seem to like, but exploiting the WordPress.org repo in this manner is unacceptable and action should be taken, especially given the popularity of this plugin. As a fellow WordPress business owner, it concerns me that this could happen to one of my customers or ~3,000 other people who download W3 each day. It would be a disservice to the patrons of WordPress.org to continue exposing this plugin with upsells in it.

  16. W3TC is a pretty good example of being unprepared for popularity and perhaps not foreseeing the wide array of sites, hosting, and user skill level. Not really hard to see how things got out of hand. Performance with WP can be a little chasing two rabbits. W3TC is an impressive offering, but it requires impressive skill and patuence to configure for it’s greatest returns. Most users have neither attribute.

    Oh, and for those who seem unable to get past it, and use it for their favorite whipping tool, no matter the topic, please consider dropping the damn Emoji complaints. It makes you look exceedingly whiney and simple-minded. It is in core, it is not going to be pulled from core, get over it. Thank you.

    • no matter the topic, please consider dropping the damn Emoji complaints. It makes you look exceedingly whiney and simple-minded. It is in core, it is not going to be pulled from core, get over it. Thank you.

      Who’s making such complaints “no matter the topic”? And would you mind explaining when such complaints are topic-appropriate? I didn’t realize there were rules about such things.

  17. Lot’s of hate in here that is unwarranted.

    Every plugin author has had a dream or currently lusts after getting a tiny sliver of the traction of W3TC has/had. The vast majority have failed.

    Every entrepreneur and business owner makes a huge gamble their product will be adopted, let alone be used so widely, and prays they can monetize that traction somehow to recoup the cost of development and maintenance.

    Support on a free plugin/service/thing is unsustainable, especially at the scale of W3TC and the complexity of the problem it aims to solve.

    The word ‘pivot’ is so cliche because it is such a common thing for a startup/business.. Idea’s fail despite best intentions, so they try a tweak to a paid support model, or a paid SaaS, or a new product line.

    An attempt at a W3TC paid support model may not have proven feasible.

    Fred may have made some mistakes but the arm chair quarterbacking to his motives and the righteous fury from the lazy-boy is laughable.

      • scam! you think it was scam when he provided all the superb features for free in a plugin that the world was praising all these years?

        Clearly you’ve not built a product and provided any support to anyone.

        C’mon! give the guy some breathing space… free loaders!

        • If you pay for something and you don’t get it, it a scam. ¿Do you really think that what the developer did in the past gives him the right to take your money and not give you the service you paid for without any communication?

  18. Absolutely abysmal customer support for paying customers. I had to resort to Twitter, Facebook, and Slack to try to find out if other paying users were receiving support. Ultimately, we had to file a claim against the company with our bank to receive a refund for a defective product that, after multiple emails over the span of multiple weeks, gave us no support at all.

    Good intentions matter not, especially when it comes to people paying money for a service. Delayed responses due to high support volume from paying customers mean that money is coming in, resources should be invested into support staff, and at the very, very least, an automated message be sent out to let paying users know that they are very backed up and when we might expect replies for our issues.

    I can’t support using this plugin, no matter how good it is, anymore due to how scammed my company felt after purchasing premium upgrades and having our emails go into black holes for weeks. It’s the single worst experience I’ve ever had with a WordPress product, or any software product that I can remember.

  19. “Tens of thousands of unit tests” …sorry, I don’t buy it. I’ve been a big proponent of W3TC up until late last year when things just seemed hopeless with the amount of issues increasing consistently.

    I’ve transitioned all my clients to WP Rocket and recommend it to all new clients as well and I couldn’t be happier with it. Super simple UI, excellent performance, solid code and transparency in communication.

    It also works amazingly well with NGINX with a very simple config.

    I don’t see myself ever going back to W3TC.

  20. I definitely prefer WP Rocket to W3TC. Switched multiple sites over to it and haven’t looked back. Easier to setup and use, and customer support actually cares.

  21. I hope Frederick is going through the comments… so this one is for him:

    I understand that you’ve been facing issues and of course, it can’t be easy to provide support to a lot of thankless jerks who somehow believe that it’s their birthright to get the free support. As a business owner I do understand the challenge of this massive scale can’t be easy.

    I think – W3TC did what no one else could. It provides features that even best of the caching plugins fail in doing so – there are hardly any commercial ones out there.

    I think its time to take hard calls – look at WPRocket’s model ( preloading cache bot – proof that people are ready to buy stupid stuff) – reduce the features of free version and put them in the pro version. May be add redis cache as a PRO feature.

    Making the github repo public is also a good choice but i’m not sure that how many plugins actually get commits from other developers as much as WP itself gets.

    A big thankyou for doing what you’ve been doing with w3tc all this while! God bless you

  22. I have had many conversations with Fred over the years. I think he is one of the good guys. His heart’s definitely in the right place. And he’s one helluva developer.

    And that’s all I got to say about that.

  23. This is really awful. Someone has given so much of his time and effort, for free, and this is what he gets for it.

    I believe it is time for somebody to help the developers who make WordPress so popular and in doing so enable a few savvy entrepreneurs to make a ton of money.

    I assume that the WordPress people, so many of them, believed that by providing a free program the spirit of that would continue on into the community, and it worked for a while. But obviously, when something gets popular the sharks come out.

    One million users!!!! One million. Unbelievable. 1 crummy dollar from each one of the people who use that plugin and that man would be able to continue his work on an obviously great plugin without having to deal with trying to “monetize” his business.

    Does no one but me realize that monetizing and programming are very different skill sets?

    I think WordPress needs to think about setting up something for programmers like this man, maybe something like a kickstarter website where everyone is guilt-ed or encouraged into kicking in a crummy $1 to these programmers who are really contributing to WordPress and us all. Instead of letting these talented people fall by the wayside because they are pure programmers and are not great at monetizing.

    A little less competition and a bit more kindness and perhaps WordPress will not fall by the wayside too.

  24. W3 is really wonderful but configuring the plugin is really difficult, I am still using it but now thinking to switch over to some other plugin.

  25. I stopped using them years ago. It was a nice plugin indeed, but it shouldn’t take someone a long time to set up the free version.

    I didn’t understand most of the crap on the settings page so I used Google to search and found a blog post written in detail how to edit the plugin and explained in detail what some meant. I felt like I had to go to school just to set up this plugin. I have not paid attention to this plugin in years and didn’t know it was this bad in treating paying customers bad. First of all, paying clients is your number 1 goal to continue to satisfy so they will keep on paying and never mention 1 bad thing about your plugin and even stick up for it.

    Now, it looks like some people basically don’t want to depend on them no more till they see a major plugin update which a easier interface would be nice and suited for newbie bloggers. And a overhaul on support. Support is like the number 1 thing a plugin author needs to focus on besides plugin updates if security is at risk. Support doesn’t have to be perfect because any plugin author or business in general does not have the perfect support system down. Issues to happen when a support staff member maybe deletes your ticket by accident or didn’t like your comments in the support ticket so they moved you to the end of the line.

    Fix the support problems and maybe get some free volunteers to do the support and mention that this job could turn into a paying job based out of performance doing support and making contributions to the plugin as well. Get people who are Eager to work on a plugin they first of all love and maybe these support members will be the one to make that tiny change that we all may want.

  26. Here are my two cents. I have spent hours upon hours working with, and playing support on the forums for this plugin.

    W3TC got to big and there are way too many people who shouldn’t use this plugin. W3TC was not meant for every average joe to use on their blog. That is more in the realm of WordPress Super Cache and there is a sharp learning curve with the plugin.

    Then we get people who write various blog posts regarding the plugin but they all recommend different configurations and the vast majority of them have no clue what they are talking about. This makes the support forums filled with questions that have become repetitive and shows that most people simply shouldn’t be using this plugin.

    You will see average blogs trying to setup varnish, object, database, and every kind of caching module they can think of and while this is great on pen and paper it’s an utter waste in practice. The performance drop off for readers when you are using Disk Enhanced vs (all of the APC, Varnish, etc caching modules) is very minimal. You almost will see no increase but because it’s in the plugin users have developed the mentality that they need it.

    This plugin was marketed as the perfect caching module for every WordPress website and it thus overloaded his work. This plugin isn’t meant for those who don’t know what they are doing. If you are running a basic blog you don’t need all of the caching modules it offers, you would be better off with WordPress Super Cache.

    However, if you are using BuddyPress, bbPress, WooCommerce, etc in your website you will need to make use of the other caching modules.

    TO Frederick.
    If what you say is true at least open the GitHub repository and let us see how these changes are coming along. I know there are plenty of us willing to test it. I know I would but there needs to be some kind of evidence at this point for us to actually believe in what you’re saying since you have taken the money of many customers who have never received any support/configuration from your team.

  27. I know Fredrick took the time to do something everyone got for free. However, Fredrick took my money for premium support and did not provide the service I paid for. An ethical business would simply refund the money but Fredrick will not answer emails! As of now, 5 weeks since I paid for his services, he has not refunded my money despite multiple requests! He should be doing the right thing and refunding people who he can’t get to if he’s too busy! SHAME ON HIM! DO NOT SUPPORT UNETHICAL “Businesses” like this.

  28. Well, I feel sorry for all those who paid for support from W3TC and didn’t get it; hopefully you will get some form of reasonable compensation.

    To all others who, like me, have been using this fantastic caching plugin for years without much hassle (except for learning how to configure it), well, shame on you if you don’t do anything else but complain; Townes is not under any obligation to continue to support freeloaders. Also, you got the code; put it on GitHub; continue to develop it, if you wish. After all, you get 100.000 or so lines of code for free. There is little else that you need to do (except the occasional bug fixing; dealing with the occasional exploit or security issue; and fighting obsolescence, as new versions of PHP and WordPress come out regularly scheduled, deprecating lots of functions and changing the way things are programmed).

    More seriously… I believe that Scott Hartley above has hit the nail. W3TC is not for everybody. I have been telling that to pretty much everybody I know — both ‘in the flesh’ but also in many forums and blogs and whatnot. W3TC is way cool in the way it deals with object and database caching (and yes, so far, there is nothing else doing the same; I have seen some ‘widget caching’ plugins which allegedly attempt to do what Townes has done, but at a very small scale and with little success). But you may not really need all those bells and whistles. After all, from the many comments that I’ve read here, with people successfully migrating to page-level caching services (either free or paid for), well, that’s more than proof that none of those websites really, really needed all the extras that W3TC provides. For reasonably static content, after all, any simple page-level caching service is more than enough. Even just activating CloudFlare in front of your server and using Jetpack’s Photon is more than enough — why worry, if you have less than a request per second on your website?

    It’s when you start getting hundreds or thousands or more requests per second, and when most of those requests require juggling with logged-in users and their cookies and tailored JS to handle them — as is the case on the high-end e-Commerce platforms, for example; or a popular BuddyPress installation — then, yes, your puny little page-level caching service will simply die. Because all requests will now be hitting your PHP engine and the database. All of them. Thousands of times per second — or possibly more, because WP (and WP plugins) is keen on doing quite a lot of table joins. Joy.

    It’s at that time that you will be most glad to have an enterprise-class caching module: one that works in those critical conditions. One that is able to isolate the complexity of the WordPress engine and the way each element on a page connects to the database for making requests — when there is no ‘simple’ way to cache a page. That’s where the beauty of W3TC comes into play.

    But for those blogs or websites with 50, 500, even 5000 visitors a day? Forget it. Even if you host quite a lot of websites on the same server — each with low traffic — there will not be a huge performance increase due to W3TC, compared to ‘simple’ page-level caches (such as WP Super Cache, WP Rocket, or WordFence’s own Falcon Engine). W3TC, after all, will be dealing mostly with page caching anyway — perhaps saving the odd database access here or there, but it won’t have a huge edge over other, simpler solutions.

    Consider my own case. After months of struggling to fit all the small websites I host inside the available memory — and utterly failing! — I decided to push all of them to a server with twice the CPUs and four times the RAM. Needless to say, even when unoptimized, the server works like a charm. It neatly purrs with joy at the largesse I have bestowed upon it. Basically, it’s sleeping most of the time — and serving perhaps four or even ten times more requests than before, in a fraction of the time (in some cases… it’s 20-30x faster).

    I did some benchmarking. I started from a ‘standard’ WordPress install, with no cache whatsoever. Then I started adding things to it: first, CloudFlare; then a simple page-level cache (I used Falcon Engine, just to save the trouble of installing one extra caching plugin). Then I started tweaking with nginx’s fastcgi_cache, placing everything in memory. By then, few requests ever came to WP itself, but I could do better: get nginx serving the static pages directly, not even hitting the PHP engine. Cool: performance, in terms of page load speed, improved tenfold. So I went on, installing W3TC, and tweaking the configuration so that nginx was serving static content directly from memory — basically, W3TC’s page cache was not even being hit. I still needed W3TC to write static pages on cache, of course, so that nginx could serve them. This wasn’t easy to achieve, and, in some cases, I was not even sure (in spite of all those extra headers) where the actual page was being served from… memory? disk? CloudFlare? Or from WP itself? Sometimes it was not 100% obvious…

    Was there any major improvement? Of course not. From a statistical viewpoint (both the testing server and the server being tested were ‘on the Internet’, although the testing server has practically zero traffic), it was impossible to see a difference in performance between the easy, page-level caching solution, and using W3TC together with a super-tuned nginx/fastcgi_cache/php_fpm solution. On top of that, I started using PHP 7, which includes the nice JIT (yay, 15% speed improvement, just by installing it) — but that didn’t make a huge difference, either, except, of course, that W3TC seems to have many compatibility issues with PHP 7. By the time I’m writing this, who knows, version 1.0 of W3TC might be out, and fix all those issue, but, at this very moment in time, I basically have the choice to keep PHP 7 — which works with everything I’ve got installed on all my websites except W3TC and New Relic — or to go back to PHP5, because, as far as I can see, W3TC isn’t even writing pages to disk at all. Or if it is, it’s hiding them well — so well, in fact, that I don’t know where to point nginx to them :) thus pretty much making the exercise pointless…

    In any case, my long story should just be an example to those who are blaming Townes and grumbling about the lack of fixes and updates and upgrades… be honest with yourselves, and eventually with your customers. Do you really need everything that W3TC offers you? Because if you do, you know how to configure it to make sure you’re getting the performance you need out of W3TC. If you’re scared by the W3TC configuration panel and find it ‘too difficult to use’, then it’s highly likely that neither you nor your client/customer really, really need anything else but a simple page-level caching plugin and a well-configured server to run it. That’s all you need.

    On the flip side of the coin, I can see even here on the comments that there are certainly a few people who do need W3TC for their customers, and they have nothing else but praise for Townes and his excellent work so far. And there are a few curious people like me: willing to learn how to do something new and understand how it works, even if we don’t really, really need all the performance and features of W3TC. But it’s just good old plain fun to learn new things. I still have a few options in W3TC that I haven’t toyed with; and I have only signed up to New Relic today, just to find that it’s not compatible with PHP 7! Bummer. It’s not easy to live on the edge… without getting cut.

  29. I was superfan of W3 Total Cache but… its time to drop this plugin… It shows a lot of errors with PHP7 that needs to be manually fixed. My error_log is also full of php warnings.

    No support, no updates, no news… what should happen to consider a plugin dead? The plugin is abandoned, end of history.

  30. If you want to use a static caching plugin, WP Rocket and Comet Cache are two solid payed options. WP Fastest Cache and Cache Enabler are really solid free options and both are on the WP repo. Have tested WP Super Cache, but it was just okay performance wise.

  31. W3 Total Cache is perfect (if you know properly configure). However, after four months, nothing has changed: no update, no news …

  32. It’s a shame to see this plugin go to the weeds. There are some big challenges to see it grow in the coming years, and without any response from Frederick, I fear it will eventually become too obsolete to recover. Unfortunately, I don’t know of any other tool that can cache in a multi-server setup (Batcache is just as ancient).

    There are a few pressing issues that need to be addressed (blank screen bug, PHP7 compatibility). There are a bunch of fixes in the support forums, but they need to be added to the plugin.

    W3TC was originally positioned as an advanced caching tool for big sites, but became the go-to for every little site, and the support load became unbearable.

  33. After 2 crashes of my pages because of using php 7.0 I decided to change to WP Rocket (after 3 years with W3T). It’s not free but I get support and plugin works great.

  35. I am currently unable to make updates to the homepage of a website that I manage. I finally figured out that this was because of the W3 Total Cache plugin, which I had earlier paid to have professionally configured. I paid $75 for 15 minute email support, but I have not yet received any response to the support request.

You may also like

Newsletter

Subscribe Via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

%d bloggers like this: