EXPLORE CLOUDWAYS
Experience lightning-fast loading times and seamless platform
View Demo > A Cloudways Black Friday readiness webinar covering four pillars, speed, security, automation, and monitoring and recovery, with a live platform demo and audience Q&A.
🎙️ Speakers
▸ Miranda — Director Digital Marketing, Cloudways
✨ Key Takeaways
✦ Slow checkouts, bot traffic, outdated plugins, and thin monitoring are the main risks to Black Friday revenue.
✦ Layer your caching across Cloudflare, Varnish and Breeze, and object cache, but never cache cart, checkout, or account pages.
✦ Layer your security too: WAF, rate limiting, browser integrity check, and a deliberate choice on blocking AI crawlers.
✦ Use Site Manager to schedule updates in off-peak windows, and choose Safe Updates for high-stakes sites.
✦ Watch CPU first, then RAM, disk, and bandwidth, and check why CPU is high before scaling up.
✦ Know your backup frequency and retention, and who restores the site if something breaks.
✦ During peak traffic, avoid purge all, batch product edits, and schedule content updates off-peak.
Miranda: Is your website ready for Black Friday for the upcoming season? This question really matters because there are a couple of things that can put a large risk on your revenue and on the outcome of your Black Friday campaign. For example, if your website is loading slow, and especially your checkout, there’s a higher risk that people are dropping off. Another risk is that if you get a lot of bot requests, they are taking a lot of your server resources, and it limits the resources that are left for your legit users. So this is also an important point to get addressed. And then if you have outdated plugins, that makes it easier for your sites to get hacked. And if you monitor, but it’s very thin and you’re not looking into all the details, that can also pose a large risk.
So today we are addressing four pillars: speed, security, automation, and then monitoring and recovery. We first have some information about it, and then we’ll show you in the platform where you can update certain settings.
First of all, speed. A very important factor here is caching. It’s important to check what your current setup for caching is. If you’re on WordPress, it’s important to have Breeze with Varnish enabled. If you have Cloudflare, that will bring extra leverage in terms of speed. You want to have most of your pages cached, which can differ depending on purpose, and that’s okay. But for example, your cart, checkout, and account pages should not be cached, because those are different per user. So it’s good to check those separately so you don’t get any surprises. And if you make any changes to the site, it’s important to know how to purge cache so the visitors actually see those changes.
So what basically happens when a visitor comes to your page? The request goes through a couple of layers that can be cached. The first step is Cloudflare, if you have that enabled. If a page or any resources are already cached there, they get served to the user, and the request is handled on this layer. If it’s not cached, it goes to the next one. Then you have Varnish and Breeze, which show a cached version of your page. If that part is still not cached, it goes to your WordPress and database, where you have, for example, object cache, which is basically a cached version of your database. The purpose is that if anything gets cached on the way, it limits the number of requests that reach your server. That helps make sure your server stays available for all the checkouts and important pages and actions that need to be done during Black Friday.
So we’ll show you where you can find this in your platform. First of all, if we go to the WordPress admin for the Breeze settings, we’ll use this website as a demo. If we go to the plugins, we have Breeze, and under file optimization you can cache your full pages. This one is an interesting one as well, because let’s say you want to cache your full pages, but there’s a certain part you don’t want cached. Maybe you have some dynamic text which needs to be pulled in directly from the database. You can give a class here and say that anything with this class does not need to be cached and needs to be freshly pulled in. So you can control this with the refreshed cache space parts. It’s a very handy feature.
And this is another one, showing a loading overlay. Let’s say you have some text, for example in the color red. Just make sure that the hex color is correct. Then you can also say that anything with this color should not be cached. So these are some interesting features that you can find in the Breeze plugin.
Then, in terms of auto purge and manual purge: if we go back to Breeze, we can see it here as well. Here you can purge all cache. If you make any changes on the website but they’re not visible yet to the user because of cache, you can use this option to manually purge your cache. And here you also have Varnish. If you publish a new blog page or comments, for example, you can auto purge the cache so that it gets updated. These are two important settings to be aware of: the one on Varnish and the other one on Breeze, and then purge all cache.
The next option is on the Cloudways platform, where you also have Varnish settings. If we go to the platform and then to applications, and then to application settings, we have Varnish. Here we can mention any rules that we want to exclude from caching. In Breeze we can say that certain pages, or certain parts of the pages, should not be cached, and we can also do that at the application level. So for example, anything on the WP admin, or certain custom checkout pages, you can add them here as well.
Then on the Cloudflare side, if we go back and open Cloudflare, we can also purge the cache on the Cloudflare end, so we see the fresh version of the page. We’ll come back to this later as well, but here you can also see image optimization, your caching, and if you also want edge page caching. Those are the settings for Cloudflare that you can find in your dashboard.
If we then go to the next pillar, security. Your WAF in Cloudflare blocks malicious requests. In Cloudflare you also have rate limiting. What that does is that if you get more than 200 requests per 60 seconds, that kind of traffic is either blocked or gets challenged, to prevent bots from just firing on your website and eating your server resources. There’s also browser integrity check, which checks for any known or spammy or suspicious requests, which can then be blocked, or you see the Cloudflare challenge. So that’s important to have enabled.
In Cloudflare you can also block AI crawlers. Just be careful with this one. We don’t necessarily want to block AI traffic, so that’s a choice to make: do we want to block all bots, or only specific ones? And whenever you implement these settings, it’s important to check whether your customers and search crawlers can still go through your website, to make sure your content stays indexed in search engines.
What works best with security is having levels of protection. The first level that requests go through is Cloudflare, so there you have what we just discussed: the WAF, the rate limiting, the AI crawler blocking. If anything passes through, it comes to the Cloudways platform, where we also have the firewall, bot protection, and DDoS mitigation, in case you get a lot of requests at one point. And if something still gets through that level, then it comes to your website. So you also want to make sure that your plugins are up to date, so that there are fewer vulnerabilities for anything to get into your website.
Let me show where you can find it. If you go into Cloudflare and then the WAF settings, it’s important to have caching on, and these kinds of items. Then the rate limiting. I was in settings, on the security rules. Here you see the WAF, the web application firewall, so it’s good to have that one up. And here you see the rate limiting, so it’s good to have that on too. It says that if there are more than 200 requests from a single IP within a 60 second interval, those are either blocked or challenged. That’s an important one to have on. And here is the browser integrity check. Those are three switches which are good to have on at any time. The AI crawler one is a choice, because if you want to have your website in large language models, then this is one to consider more carefully.
Then the part on automation. Especially if you’re managing a lot of websites, it can be quite a challenge to keep all of them up to date. A very handy tool for that is Site Manager. It gives one view of all the websites you have and their status. There are also options to do a quick update for your plugins or a Safe Update. We have a separate webinar on it, but I’ll show you the difference when we go into the platform. You can also schedule updates within a peak window that works well for you, and make sure you get alerts if there’s something wrong, so you’re able to act fast.
One thing on the difference between the Quick Update and the Safe Update. If you press quick update, you are directly updating the plugin on your website. It’s very fast, especially if you know that a plugin usually doesn’t have any issues, so it’s a low-risk one. For those cases, a quick update is good. However, if your site is very important and you want to make sure it always stays online with no risk, especially during this time of the year, then you can choose the Safe Update. What it does is first make a copy on a staging domain. It updates the plugin there, and then compares whether the website is still visually the same. If it is, it pushes the plugin live. So with this one you have less risk when you update your plugins in an automated way.
To show that, for Site Manager we go to integrations, and then we see Site Manager. There’s a different way to get there per site as well, but the advantage of going via integrations is that I see the overview for all the websites that I have. If you go to your application, you will also see Site Manager for that particular site. But here I can see, for all the sites that I have on my servers, the number of updates, the performance, and when the next auto update is scheduled. Let’s say we click on this one. Then we can see the two plugins that need to be updated, which version they’re on now, what the new versions are, and also performance. If I click on details, I come to the application-level Site Manager, which is specifically for the application that I’m reviewing.
If I go back to the centralized dashboard and click manage updates, I can say, for example, that I want a quick update, so the updates happen within seconds, but with no compatibility checks or backups. Or the Safe Update, which first creates a backup of your application and checks the compatibility. It takes more time, but it’s safer and reduces the chance of issues.
Then here you have the options for auto updates, which you can set per site, depending on where your audience is based. Here I’m updating my plugins on a weekly basis, on a Saturday night, when hopefully there are fewer visitors on the website. I can also select which plugins I want updated. If there’s a plugin that often causes issues, or I don’t want it updated, I can just unselect it so it’s not part of the updates. And with Safe Updates, I can also say how much difference the before and after is allowed to have. I have it on 0%, but you can also choose 5%, or allow a 10% difference. It then measures how much difference there is, and whether it needs to be pushed live, yes or no.
Then, in the unfortunate case that something may happen, it’s very important that you know where you can find things. The important metrics to watch, before it happens so you can take action, are the CPU, the RAM, the disk, and the bandwidth. Those are the important ones to monitor, along with any other application-level metrics. If you have alerts, make sure they go to the right person, and monitor your alerts, so that if something happens you’re able to respond very quickly. Your backup frequency is an important one, and how long you want to keep your backups. And if you want to restore a backup, make sure you know who to contact, so within your team you can push a backup quickly if needed.
So the questions to ask for your application, for your website: who is receiving the notifications? Within your Cloudways environment, you’re also able to add team members. Then the question to ask internally is, if there’s something wrong, who is restoring the website, or who is putting back one of the safe backups? And what is the frequency, so how far back do we need to go? Have that person assigned within your company, or yourself, so you have these questions ready. If something happens, what do I need to do in order to get the website back online?
In terms of the server, the bandwidth, etc., we can find that in the platform if we go to our servers and then to monitoring. Here we see the metrics for each one of those: the RAM, CPU, and disk usage, and bandwidth. For this server, although I have nine websites on it, I’m still quite good in terms of my limits. But especially coming up to the peak season and the peak days with Black Friday, this is an important page to monitor. Some quick things you can also do: get the MCP from Cloudways, and for example set up daily alerts, so that it can pull in the data.
Then in terms of application metrics, we also have monitoring here. We can see, for example, whether certain IPs are sending a lot of traffic, whether I need to block a certain IP, or whether something is suspicious. Here we see the IP requests, and this is masked now, but it’s basically an IP address and how many requests came from it. This looks still quite okay, but if you’re in a situation where you see a lot of spikes, then we can also see what kind of pages they are visiting, and whether there’s something you need to do with this IP address. The same goes for the tabs for bot traffic, what kind of URLs are requested, and the status codes. Here you see that in terms of bots, it’s mainly Semrush bot and Googlebot visiting this website. It’s important to keep an eye on these as well, and you can also review the PHP paths, etc., and the disk usage.
If you experience any issues on the website, it’s good to look into your logs, specifically your access logs but also your error logs. If you find any issues and you’re not sure what they mean, Claude can also be a good friend there, but this is at least the way you can find them in the logs and then the error logs, to resolve an issue you have on your website.
Then in terms of backups, you can see your backups in two places: one on the server, one on the application. I’ll first show you the server one. The backup of the server here is basically every day, and you can change the frequency. Based on your use case, if you want a backup every hour, or you think only once a week is enough, you can change that here, in your server under backups. You can also say what backup retention period you want, for example four weeks, or one week is enough, and what the scheduled time is that you want to take this backup.
Then if I go to my application and then to backups, here it’s called backup and restore. Let’s say something got changed on your website and you’re experiencing an issue and you want to go back to a previous version. You will find that under your application, in this section. If you want to go to your latest backup, you can say restore application. But if you know that an issue got introduced on an earlier date, you can also say you want to go back to, for example, the beginning of this month, and restore the application. Those options you have here. And if you have fixed something and you know it’s now working correctly, you can also take an extra backup so you have a fresh copy after your updates.
If you’re experiencing any issues with your website, make sure you know which area you can work on. If your resources have enough space, so you look at your CPU and your bandwidth and you see you’re not reaching the end, then see what you can do in terms of optimizing your website. Make sure that your caching is on, and check the caching on the different layers: in Cloudflare on the edge, at your server level, and within your website. One thing is switching it on, but also check the settings. Do you want to have all your pages cached? Do you want certain sections of a page not cached? All of that is possible in this area.
If you see any issues with slow pages, check whether it’s applicable to the whole website. Is my whole website slow, or is it only related to certain pages? It could be pages where you’re loading very large images or video or other large file sizes, in which case optimizing that page may also increase performance. And if you look at your CPU or RAM, the metrics we saw, and you see that you’re reaching your limits, like 80 or 90%, that would be a good reason to scale up your server.
To scale up your server, you go to your server, and I will show you as well. You can just switch your slider, and that’s actually it, and then you have more server space. And the good news is that we now have a discount for that. If you want to do vertical scaling, you will get a 20% discount for six months, and you will see that reflected in the platform. Let’s say I go to my server and there’s the option for vertical scaling. I’m on a server with nine websites on 4 GB. If I increase it to 8, you also see here the updated prices already with the 20% discount. So just for you to know.
So in terms of your Black Friday checklist. Make sure you know what cache is currently enabled, and see if you need to make any changes. Check your cart, checkout, and account pages to make sure that those are not cached and are working properly. In terms of security, have the firewall enabled, have the switch for the rate limits enabled, and check the crawler rules, whether you want AI bots to crawl your website during this time or not. In terms of automation, Site Manager is very helpful, especially if you have multiple websites, because you can see all the plugins, or WordPress, that need to be updated across all the websites. It’s very handy to schedule them so they are always up to date.
In terms of monitoring, keep a close eye on the graphs for the CPU and the RAM, and especially if you’re nearing your limits, it’s a good time to scale up. And in terms of your backups, see if you have your backups enabled, whether the frequency is right for you, and make sure that if something happens you know who you need to contact in your company to get the backup restored.
Now we switch to the Q&A section. If you have any questions about what we talked about today, specifically about speed, security, or operations, feel free to share them in the chat. We have myself, but we also have our colleagues from the product team in the chat who can help you with that. Another good thing: we gave the overview of what you need to do to make sure your site is ready for Black Friday, and there are a couple of webinars which are more specific to the topics that we discussed. The presentation will be shared with you, and it contains links specifically about Cloudflare Enterprise on Cloudways, Site Manager, how to improve your Core Web Vitals with Varnish caching, and how AI can help diagnose and resolve issues on your website.
Host: Awesome, Miranda. Thank you so much for the demo. I can see there are a few questions in the comment section, so we can take them one by one. The first question is from Abby. Abby is asking, do you recommend having plugins all set to auto update to help with security? Are there any issues to consider before I do that? Previously I was keen on doing more manual updates so I can check easily that nothing breaks.
I think if I start with the answer, we have Site Manager available, and we have Saud as well on the call from the product side. Site Manager is the best way to manage your updates, because it also performs visual regression testing, and it creates the backup first and then performs the updates. So there is a very, very minimal chance to break the site, and you can also schedule the auto updates. By scheduling, you do not need to check each and every plugin manually, and the WordPress core and even the themes manually, to update them. Site Manager will solve this issue.
You can simply use it and schedule the updates, and it will first go with creating a staging site, then create a backup, then perform the visual regression testing, and then perform the updates, and it will also share the results. In the logs you can see which plugins are updated. If there is an issue during the update, you will see the results in the error logs, the reason why this particular plugin was not updated and was rejected by Site Manager. So this is the best way to update the plugins, the core, and the theme. Otherwise, if you go with a manual approach, sometimes you forget to update the plugins or themes. If it’s six months or a year that you’re not updating a plugin, it may break your site, and it also opens the doors to vulnerabilities. So make sure to use the solutions to auto update the plugins. What’s your take on this?
Miranda: Yeah, I would say the same. Especially if you manage a large number of websites, going to every one and doing a manual update just takes a lot of time. If you have Safe Updates enabled, it’s a very safe way, because you keep your website up to date, and at the same time, because of this testing first on the staging website, there’s very little risk that something goes live in production. So yeah, I would go for that option.
Host: Awesome. I think we have another question from Yunus. He is asking, which metric should I watch most closely during the sale? Either CPU, RAM, or something else? If I go first, I must say the CPU usage, this is important, because Cloudways recommends that if your CPU usage is at 80%, you need to upgrade the account. But you also need to see why the CPU usage is high. So make sure to keep an eye on the CPU usage. You can also check in the monitoring tab and see what the reasons are behind this 80% or 100% CPU usage.
So CPU usage is first. If you are utilizing the RAM more, you need to upgrade. If you are using disk space more, you have an option to upscale the disk space as well. So make sure to go with CPU first, then RAM, and then the disk space. This is the pattern you can follow when you are close to the sale season: keep your eye on CPU usage and RAM, and decide whether you need to vertically scale your server or not. Miranda, if you can take this.
Miranda: And also what you mentioned, it’s good to check why the CPU is high. For example, in one case a bot was sending a lot of requests. It was an event site, and they were trying to get to the payment system to check credit cards, for fraudulent actions. In that case I didn’t need to scale up my server, although my CPU was at 90%. What I did was check which IP address it was coming from, or what user agent they were using, and then I blocked it, and the issue was over. So if you see that your CPU is very high, also check whether it’s because of normal traffic, or whether a certain bot is sending a lot of traffic, and whether you need to take any action to block that bot. If it’s high, that gives you a reason to investigate what is going on.
Host: Awesome. The next question is, how do I tell if a slowdown is a server resource problem or an issue with the site itself, like a heavy plugin? As I said, Miranda, you can add your comment here as well. There are certain things we need to check, whether it’s a server issue or a heavy plugin. We have a monitoring option. You need to go to the server, then monitoring, and you can check the summary, what the numbers are on the CPU and the RAM, and you can also check the application-wise details. You need to go to the server, then monitoring, then application-wise details, and identify which website is consuming resources, for example if you have multiple websites on the same server.
In the monitoring, as Miranda already showed us in the demo, we have analytics, we have PHP, and we have MySQL. Here we need to identify the slow PHP scripts and slow database queries. These are the most important things for identifying whether the issue is in a plugin or in the server resources. If the server resources are healthy but specific pages are slow, for example your product page or your description page is slow, then you need to investigate the application and the plugin, because the server is showing healthy metrics. You need to check the CPU, the RAM, even the PHP processes, and one more important thing is the response time and the HTTP errors. You can see the HTTP errors and see how the response time is behaving inside the platform. In this way you can identify the pattern, whether the issue is in the server resources or with a heavy plugin or theme.
So check all the resources, check the error logs, check the monitoring tab, and then make a decision about whether the issue is with the server or with the plugin. One more thing you can do is connect with the support team. You can inform them that the CPU numbers are good, and if you can identify which plugin is causing issues, you can also get input from the support team. I think this is the best way. Miranda, what’s your take on it?
Miranda: Yeah, I think it’s a good part of analyzing exactly what is causing the issue, because then you know what to update. Maybe also good to know that in the platform we have Copilot, so you can ask questions there to get more insights into what is causing the issue and what you need to update. That might be worth exploring as well.
Host: Awesome. Let’s move on to the next question, and meanwhile, if anyone has any other question, feel free to ask. There was a question related to Cloudflare, and Saud already answered this question, so I think we are good. We have a question related to Breeze, and I just got the answer from the product manager. Let me read it out, and I can also share the answer in the chat.
The answer from Mansour is that when you update a single product or page, Breeze does not clear the whole website on Varnish and Cloudflare. It clears the page and related URLs for a product. This can also include, for example, shop and category pages linked to it. So these pages may need to rebuild their cache once, which can cause some short load, but it should not cold start the entire site. To keep performance stable, avoid purge all during peak traffic. Prefer updating only the products or pages you need, so you do not need to purge the entire website at once. Make sure to purge the cache on specific pages.
Don’t turn off Breeze, Varnish, or Cloudflare during the sale, because we are not here to change the settings, we are here to optimize them for the website. Make sure to not turn off the caching layers, because it will cause a problem. If cache warm-up is available, keep it enabled so important pages refill quickly after a purge. As we discussed, you can also cache certain parts of the page as well. And batch product edits when possible, instead of many separate saves in a short time. For example, if you have product updates, do them in batches, because whenever you update a product page it creates the cache and also creates requests to cache that page.
And the last point we shared is to schedule non-urgent content updates outside peak hours. For example, if your product page is very important and you are observing a footfall on the website, with many people visiting right now, you do not need to schedule the content update at that time. Make sure to do it outside the peak hours. In this way you will have more bandwidth on the HTTP request side and on the caching side, so your website will not break, and you will see the clean version of your website. If you make any update, you will see the update on the website. So this is the best answer, and I am also copy pasting the answer here.
I think, Miranda, we are good on the Q&A side as well, and the good part is we are also on time. I would like to thank everyone for joining the call, and thanks to you for taking the charge and showing the presentation and guiding us on how we can prepare our websites for the peak season. I think it is a good start for everyone to implement all those pointers Miranda discussed and make your website ready for Black Friday.
The last thing is, we will share the recording and the slides with you, and the slides have all the webinar links. For example, if you want to learn about Cloudflare, about Site Manager in detail, about AI agents, you can check out the links. We also attach the benchmarking results: what we achieved if you enable Cloudflare, if you enable Breeze. The Breeze plugin link is also there. You can connect with the support team, and the support team can connect you with Oweis Alam. Oweis is the person behind the Breeze plugin, so you can also take help from him.
And the last thing is that we are offering 20% off for six months for existing customers. You just need to go to the vertical scaling option and scale the server, and you can avail the 20% off. If you are a new customer, we are offering 80% off for two months. So you can avail that offer as well. I will share the post-event email with you with all the links and all the details, and once again, thanks to everyone for joining this webinar. We will continue this webinar series and come again with another interesting topic. So that’s it from my side, Miranda. If you have any closing remarks, then we can conclude the seminar.
Miranda: Yeah, thank you all for joining, and hopefully you find it useful. It’s also good to check out the links and to see the benchmarks and why it would be good, for example, to enable caching. Thank you.
Host: Awesome. Thank you so much, everyone, for joining the call. Take care. Have a nice rest of the week. We are ending the call.
Answer a few questions, and we'll present you with a personalized tour of the Cloudways platform based on your answers.
Black Friday

Days
Hours
Min
Sec
*Up to $200 off per month