Skip to main content

We Planned Not to Build Another New Theme Ever Again

Transparency Report #45

Welcome to the 45th edition of the monthly transparency report (for October 2018). In this series, I go through everything that’s been going on in the business – especially the behind-the-scenes stuff – that you might be interested in. Click here to see the previous reports.

Here’s the TOC of what’s to come:

1. A year ago I said we wouldn’t build another new theme. We just did. Here’s why

2. Does page speed really matter for SEO? TL;DR: it does

3. What type of WordPress content to advertise on Facebook?

4. Rethinking transparency reports

1. A year ago I said we wouldn’t build another new theme. We just did. Here’s why

As I was looking through some of the more popular transparency reports on the site, I stumbled upon this. It’s the one where I discussed the future of Gutenberg and tried to figure out what’s the best path for a theme store like ours to follow.

I concluded that we wouldn’t build another new theme, but instead focus on making our existing themes into household names.

It’s roughly a year and a half since that report got published … and we’re just about to release a new theme. What gives?

A couple of reasons:

a) I read the market wrong

To be more precise, I underestimated how fast page builders were taking over, plus expected Gutenberg to have more mainstream appeal quicker.

And okay, Gutenberg is growing kinda nicely, but that’s still growth among the dev crowd and not much concerning the casual user landscape – page builders still reign supreme there.

In that report from 1.5 years ago, I mentioned how far page builders have come in terms of UX and also how much page builder users love lightweight framework themes – like GeneratePress or OceanWP.

At the same time, I still thought that those themes would remain niche, while the other “business themes” (let’s call them that) would continue to dominate the market.

However, looking at the popularity charts today, the framework themes are growing much faster than I expected and will likely take over the top spots sooner than later. Check out GoDaddy’s Hot 100 list, for example. Back in January of 2018 we were at #4 (see below). By May, we dropped to #13.

godaddy top

Initially, I assumed that we would get a stable version of Gutenberg much sooner so that we would be able to pivot Hestia in its direction – capitalize on the theme’s popularity plus its new compatibility with Gutenberg. However, as we all know, things have been moving much slower.

To find a solution here, we needed to either acquire/build a strong partnership with an existing established framework theme (which proved difficult), or build such a theme ourselves.

One way was to adapt Hestia and strip a lot of elements from it, but that wasn’t really possible without breaking existing user sites.

That’s why we decided to build a new theme from scratch.

With that, we also had full freedom as to how we wanted to approach this. So we made it a point to not only focus on page builders like Elementor or Beaver Builder, but also introduce some new concepts that are ready to work with Gutenberg once it gets rolled out to WordPress core.

b) Web development has advanced as well, revealing new possibilities

This comes back to what themes like OceanWP or Astra are doing. They’ve introduced some interesting concepts that have also proven to be popular with users, which has likely strengthen their position overall.

On our end, without a new standard like Gutenberg, we couldn’t just go ahead and drop Bootstrap from Hestia, introduce new CSS solutions, etc. Essentially, we didn’t know what we could expect from Gutenberg and which modification would stand the test of time.

We did a major refactor with Hestia, but the front-end was still hard to migrate without breaking some user sites.

So, a new theme comes into the picture! The aim with Neve is to enable an experience that’s as close to true front-end editing as possible – using Gutenberg.

I’m happy to report that we’ve been able to achieve that, and we’re just about to release this new theme!

Quick sneak peek:


  • Neve works with the official AMP plugin out the box – the AMP theme looks exactly like the non-AMP mobile theme.
  • It also doesn’t use jQuery, Bootstrap or anything that most devs would consider heavy.
  • The “theming” part in Neve comes down to taking care of common HTML elements and blocks – things like inputs, buttons, containers, content width, etc. Everything else can be handled through page builders.
  • On day one, we will also make a pack of starter site designs available – built in Elementor for now, but soon more Gutenberg-made versions will be added as well.
  • Lastly, if there are any Zerif users here, you’ll be happy to know that Neve automatically imports your Zerif front page content and converts it to an Elementor page that looks exactly the same.

So there we have it, a new theme that’s our response to the direction of the WordPress theme market with the big Gutenberg change right around the corner!

2. Does page speed really matter for SEO?

Like, I know that every SEO expert and their dog has been saying the site speed mantra over and over again, but does it actually matter in a way that’s noticeable?

Personally, I’m quite obsessed with page speed lately and am monitoring stuff all the time. Part of that is due to our new product – Optimole (an image-centered site speed-up tool). I’m also moderately obsesses with monitoring our traffic numbers, like most site owners.


Joining these two obsessions has led me to an interesting finding:

At some point, we’ve started having issues with what was our most popular post. We’re updating that one all the time, and pivoting things quite a bit on a regular basis. So in one of the updates, Karol introduced some pretty useful (from a user’s point of view) map charts featuring loading times for hosts from various locations around the globe. There were ~10 of those charts.

Now, despite the charts being loaded from Google’s own servers (based on Google Chart API), it seems there were so many things to download overall that it caused the crawler to get sort of blocked sometimes.

The page went in and out of Google cache from time to time. Google still ranked it, just on a much lower position – probably relying on other ways of seeing the content of the page (one time even using the AMP version).

After removing those charts, we saw this:

traffic growth

Okay, so what’s the big deal, why was the problem this difficult to identify?

Well, the page load times looked fine in analysis. So it’s not like we could raise an alarm based on that alone. We though that the cause was in the page’s content or something.

The only thing that got us thinking was the lack of the “Cached” label next to the Google listing:

cached label

If you don’t see this thing next to your listings, it basically means that even though the page is in the index, something isn’t right (at least that’s what our conclusion is now).

I recommend using tools like Lighthouse and Google Analytics page speed to have a more granular look at your most important pages.

There are plenty of site monitoring tools like Pingdom that can do much more regarding speed testing and debugging problems like that.

Again, in our case, the problem was not in establishing the initial connection, but rather how long it took to load the page entirely, plus the amount of JavaScript that was executed along the way.

3. What type of WordPress content to advertise on Facebook?

A couple of months ago, we stopped actively working on our Facebook ads and instead just had some automatic ads running. This was mostly stuff handling some classic shop scenarios like retargeting people who viewed our products or attempts to recover abandoned carts.

However, we did want to experiment with a different kind of automatic ads, ones where we took content from our blog RSS feeds and served it to people who visited our sites in the past week.

The goal with that was to set up a maintenance-free configuration (thanks to IncreaseMedia for helping us) that was productive but also not annoying to the users.

We split the articles in two categories:

  • all content ✏️
  • important content ❗

For the second group, even though it’s smaller, we boosted the budget. This gave all blog posts an equal chance.

Facebook being Facebook, it optimized things and started advertising the ones with more engagement more often, and at a lower price.

What did we learn from this?

If you’re like me, you always want to provide your readers with the exact content they’re looking for. So, for instance, when I see that searches like “best hosting companies” get popular, I understand what the person wants to know, and I try to address their questions in our content.

However, I have no idea as to what type of content a user (who has previously visited our blog) will be interested in reading as they’re casually browsing Facebook. They’re not inputting any keywords after all, so I am competing with cat videos and Kardashian news, whether I like it or not.

That is why this whole Facebook ad experiment in serving blog content is so interesting to me. I can finally see what people are eager to read when they’re not set on solving a specific task.

So, as it turns out, the most popular posts from the CodeinWP blog are our free theme collections. What also works well are our “VS” posts ?, especially the ones comparing WordPress to other platforms.

My guess is that even though people might not be googling things like “WooCommerce vs Shopify” all that much, when they see such a post on Facebook, they are still compelled to click on it and see if their favorite platform has come on top.

The most popular post in this regard is this one:

On ThemeIsle, in similar fashion, we have this on top:

After that, the how-to articles look strong as well; with the most popular ones being:

I’m no expert when it comes to Facebook ads, but I would guess that the key with stuff like we’re doing here is to target the user’s curiosity. Maybe BuzzFeed was right all along?!

If you’re wondering, we’re still spending a few thousands of dollars every month on Facebook ads. It’s still hard for me to tell if the investment is actually worth it or not in the grand scheme of things, though…

Ideologically speaking, I’m not too keen on some of the things that Facebook is doing. However, what’s undeniable is that there is a big fight for user attention going on, and if you refuse to participate for whatever reason, you will lose by default.

In our case, with more than 500 articles on various topics in the vault, we want to get them in front of our readers through all channels available.

4. Rethinking transparency reports

If you’ve read through the report this far, you might have noticed that it’s a bit different from last month’s.

I wasn’t entirely happy with last month’s results in terms of traffic and engagement, so I asked for feedback and decided that I probably need to come back to the previous style of reports.

While some people like the personal insights, it’s not the main reason you read these reports. Practical insights and lessons hold much more value, so that’s what I’m going to focus on more.

While looking through the previous reports, it seems that the most popular ones were either built around some drama (like the Zerif suspension) or around a unique insight I shared. So I’ll take this to heart and give you some more of that in the next reports (not drama, insights)!

I would love to hear your opinion about this month’s report!

Okay, that’s all I have for you this month. Thanks for reading and for supporting us! Stay updated and get new reports delivered to you by subscribing here:

All edits and witty rewrites by Karol K.

WordPress Design, WordPress Development, cPanel Hosting, Web Design, Web Development, Graphic Design, Mobile Development, Search Engine Optimization (SEO) and more.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.