Context: So, we picked up an interesting client at the end of February. The client had a decent website and were already ranking well for their respective 3 cities where they had offices. However, their number of sign-ups compared to the traffic share was less, and they were not bringing in the revenue their top 5 competitors were bringing in every country.
All in all, they had a total of 131 pages on the website dedicated to 3 cities and our job was to help them bring the numbers up.
Starting up with the optimization
As always, the biggest problem in SEO is what to optimize instead of starting to optimize every page you encounter. So, we looked at their total pages, and ended up realizing:
-
The first city had 39 pages created for it, and 7 of it were bringing in 87% of the entire traffic directed towards the city. This included the 5 service pages and 2 blog pages
-
The second city was their largest in the USA, and had 65 pages. Out of these 55, 14 were bringing in 91% of the traffic, and included 4 service pages, and 10 blog pages with almost every blog page leading to a click to at least 1 more page
-
The third city had 27 pages, and 6 pages were contributing 84% of their entire traffic, out of which 3 were service pages and 3 were blog pages
Now, we were down to fixing 27 pages instead of 131, and we were looking to upgrade 88% of the entire traffic. This meant 104 pages were contributing to 12% of the entire traffic, or on an average 0.11% traffic per page, which didn’t make any sense. We convinced them to purge 10 pages per month that were bloating their site for the first 6 months, while we created 6 pages to off-set a significant loss as well, and that was one loose-end being tied up, having 24 pages down from the initial number.
Next up was looking at city-wide data and optimizing pages on an individual level.
City 1 and 3 belonged to the same state, and had similar kind of keywords that people were looking for. City 2 was of a nearby state, but the target keywords were a little different, especially the long tails. What’s more, we realized that the competitor’s landing pages were extremely different in terms of their design, which meant that we had to spend some time in redesign and especially getting a sign-up form on some of the pages (initially this was present as a sign-up button in the middle of the page)
SEO Optimization part
-
Lastly came the actual work on the pages. For content optimization, it was an extremely simple task: look at the keywords that they were missing compared to their competitors, and optimize for them. The biggest takeaway here was that Questions and Answers were a big part of the content that we wanted to include. So, in came the content in the form of featured snippets. We made sure that all the money-making keywords were present in the headings across the pages which were mostly in the form of questions, and their answers were as descriptive as possible.
-
The next thing we definitely wanted to include as part of content optimization were client testimonials. This was going to be tricky, as not every testimonial would have worked. We had to first ensure that the clients are comfortable in sharing the case, and also they aren’t too long and verbose either. Lastly, we had to ensure that the clients provide their permissions that the testimonials can be edited to an extent. We were finally able to get at least 1 video per service page, and the pages that added at least 2 client testimonials added nearly 35% to their traffic (we added the transcripts for each video). Whatever added text reviews we were able to get were a bonus, but there was a concentrated effort on our part to push the team to get more reviews and testimonials that were to be added to the website.
-
Next up were the backlinks and citations. We ensured that every service page added 2 localized link and 2 generalized link each, with 4 pages across 3 cities getting a total of 43 links in total based on the keywords that they were targeting, and the competitors that were ranking. The NAP citations were built up to an extent including GMB, Apple, and Bing, but we had to ensure we were covering all bases. So, we built them up for top-50 directories including law directories.
End Results:
Across the 3 cities, we were able to get the following results:
-
Among the 27 pages, 17 saw an increment of at least 20% with the highest being 57%, 4 pages were between -5 to 5%, and 6 pages saw a drop ranging between 9 to 17%
-
Among the pages that saw a drop, the most consistent part was the least number of reviews/testimonials/backlinks that were built up
-
Overall, we were able to add 41.3% revenue compared to the previous year during this quarter when the initial expectation was in the range of 20%
-
Overall, we were able to grow the traffic by 32.9% compared to the previous year during this quarter with the initial expectation was around 25%
[link] [comments]
from Search Engine Optimization: The Latest SEO News https://www.reddit.com/r/SEO/comments/nee9tx/lessons_from_updating_a_website_locally_in_the/>
No comments:
Post a Comment