Growing a complex website into a traffic machine. From 300 to 1.2 million organic visits per month
In this case study, I’ll show you how in 2 years, we grew indifferentlanguages.com from nowhere to be found, to be in the top 32k websites in the world and the top 20k websites in the US according to alexa.com (now defunct).
UPDATE: February 2, 2022 – It is now in the top 50k Worldwide, top 40k in the US, and 93 in its category according to similarweb.com
SimilarWeb Rank as of February 2, 2022
There’s a video at the end of this case study with a walkthrough in Google Analytics.
This is how it all started
In June 2014, we were approached by a business owner wanting to take his business to the next level.
This was a smaller startup that had been in “inception mode” for three years and was finally ready to work with an agency that’ll help them grow.
As you may imagine, they didn’t come to us and say:
“Hi, I want a website that gets at least one million visits per month and breaks into the top 42K websites in the world.”
It was a challenge
On one side, their competitors were well-established businesses with 10 to 20 years of online presence.
On the other side, this was going to be focused exclusively on a technical, onsite, and templatized structure. No blogs, no backlinks, none of your “typical” SEO strategies.
1- Studying the Competition and Discovering If This Was a Viable Project.
Even though we were up for the challenge, we needed to do some homework to see if this was even possible…
This not only helps us gather key information about what’s working for competitors but also allows us to understand if your budget aligns with your goals.
After discovering we were in a mid competition market, it was time for THE question:
– Competitor traffic graph
If you know your competition is a $30 million-dollar company and they are investing 3% in digital marketing, that means they have a $900K budget per year. That will give you an idea of what you’re against.
If you only have a $120K budget rather than a $900K budget per year, the question is:
Would that give us enough leverage to compete?
As an example, we’ve had clients competing with companies like Home Depot, which according to a study made by frac.tl had a 2015 online marketing budget of:
- $36 Million for Search.
- $16 Million for Internal Display.
- $8 Million for Online Video.
Obviously, I don’t expect you to have a business that is looking to compete with them in every single category and every single product – unless you are Lowe’s, Costco, etc.
However, if they are one of your competitors, we will need to come up with creative and innovative marketing approaches to get a part of their market share. They have a very mature online presence.
In our current case, after learning the competition’s average budget and the client’s budget, we went for it.
Our promise was to put our best skills together and execute the best strategy possible to turn a real challenge into something that could grow over time.
3- Digging Deeper into the Competition
Now that we were all on board and feeling the excitement that a new project brings, we went into deep research mode.
Some of the tools we used for this project were:
- Ahrefs
- Semrush
- Open Site Explorer (Moz)
- Link Research Tools
- Similarweb
- Buzzsumo
- Screaming Frog
- Knowledge – not a “tool” “per se” but the most important part of the research.
We took the top 5 competitors we had previously analyzed on the surface and looked at what they were doing from an SEO and CRO perspective.
Some of the elements we analyzed were:
The look and feel of the website – design, mobile version vs. desktop version, copy, language, user experience, colors, buttons, trust-building elements, popups, engagement, blog, social sharing…
Structure of the website – categorization, silos, sections, breadcrumbs, internal linking, page size (word count), H tags and descriptions, website speed, technology, and APIs used…
Monetization and shopping experience – mobile experience, price tags, calls to action, coupons, discounts, offers, checkout flow, shipping & handling, refund policies…
Audience – age, sex, marital status, education, income, hobbies.
Marketing – SEO, CRO, paid traffic, remarketing… (we already knew this was a purely organic search project).
After we wrote down what they were doing right and what could be improved, we outlined a specific plan that was created with the intention of helping our client “be the best.”
4- These are some specifics:
Increase the word count per page. The competition didn’t have much content on their pages. Their low description word count could be improved to give more value to search engines and visitors.
Improve the internal linking structure. We built a framework to add link silo structures pointing only to pages within the same categories. A silo structure is basically a structure of contextual links that give a website context for search spiders.
Optimize Titles and H tags properly. Their competitors didn’t have all H tags on point. Some didn’t have H2 tags and some didn’t have them properly optimized.
Add breadcrumb navigation to each page. Competitors had some but not all of them.
Optimize images. Competitors weren’t really taking advantage of this. Optimizing images can send additional traffic from the image section on Search Engines.
It doesn’t sound too complicated, right?
Other things we initially explored but didn’t deploy:
- Using voice APIs to add pronunciation for each word.
- Using Image APIs to create unique images for each word.
First Phase
2121 traffic data – from similarweb.com
It was time to put all the pieces together and execute.
The system that we designed would pull information from a database to automatically create fully optimized pages with:
* The right titles were optimized and written with the correct keywords. They also utilized psychological triggers to “help” people searching online to click on the client’s website instead of the competition – compelling titles with the right keywords on them.
* The right H tags were also optimized with the right keywords, synonyms, LSI terms (Latent Semantic Indexing), and variations that would cover a wide number of keywords within the group of keywords we wanted to rank for.
* The right amount of content with optimized keywords, synonyms, and LSI terms. The pages would have at least 20% more content than the competitors with a higher word count across the board.
* Optimized images that would contain keyword-rich file names, metadata, and relevant alt tags that would add even more keywords relevant to what we wanted to add.
* The right internal linking structure to build a semantically relevant structure and categorization to the website as a whole. These links had exact match anchor text, synonyms, and LSI variations.
* Breadcrumbs: They add an extra layer of internal linking, help with indexation and crawlability (we had a lot of pages to build), and create a better user experience.
We also did some other things like adding multiple sitemaps.xml (50K pages per sitemap max) for pages and images, setting up everything in webmaster tools, fletching and crawling the website, some light schema, etc.
First Results
About 6 months after the first phase was implemented, the site started to take off.
We were really pleased, excited, and proud of how we had turned an initial challenge, with no guarantees of becoming successful, into a project that was getting good traction at 465,190 visits per month.
But something happened.
On February 2016, after 7 months of growth, the traffic became stuck and plateaued.
Second Phase
Even though the traffic had plateaued, we knew we had created something very valuable: a proven framework that brought us excellent results.
After hitting that plateau, it was the perfect time to push our framework to the next level: creating a new big segment of pages and categories.
If we were right, this would be another growth stage.
If we were wrong, this had the potential to turn into a big spoof.
So, what happened?
We deployed the second phase and we got yet another home run. We were back in growth mode.
With up to 808,043 visits per month, we felt invincible.
As you may imagine, we were in such a rush looking at how the traffic had been increasing before the plateau that we just had to keep pushing it. We thought we had a proven and invincible framework.
We were crushing some competition.
So, the third phase was born.
Third Phase – Epic Fail?
In the third phase, we did something slightly different. We decided to translate the website into two additional languages in order to grow it three-fold.
This was going to be the epitome of the project.
We were expecting to create something impressive.
However, not everything went as expected.
Check the image below to see why:
No bueno.
In 3 months, we went from 968K visits per month to 639K visits per month – losing around ⅓ of the traffic.
This looked like we were having an epic crash!
We analyzed every single bit of information we could possibly gather to try and figure out what was causing this monumental traffic loss (and of course revenue loss), but we couldn’t find anything major that was causing this problem.
Did we experience a brief golden era and this was the end?
We came to the conclusion that Google was a little bit confused because of the two extra languages launch. In fact, the traffic loss started close to the traffic gain for the other two languages.
There was nothing wrong with our approach so we had to trust the process and the system that we had put in place.
In the worst-case scenario, we would lose all traffic and have to develop another framework. At least we knew we had a framework that was successful up to this point.
So, we all just tried to relax and trust what we’d done.
And then the magic happened…
BOOM!
We were once again experiencing rapid growth.
To the point that we achieved a major milestone crossing the 1 million visits per month for the English version only (you can watch the video to see how other languages did). It doesn’t seem this growth is going to stop anytime soon.
This has been one of our most epic and challenging case studies to date.
Nothing was “average” and there were a lot of ups and downs, but we worked with our client through the good times and through the bad.
We found a way to make things work and formed potential solutions in the event of these undesirable worst-case scenarios.
All in all, a rewarding experience that will continue to grow in the years to come.
Here’s the video where I log into Google Analytics and Search Console for you to see it “live.”