Improving SEO for the Smithsonian Learning Lab
Digital Analytics Project, 2022
Project Overview
Smithsonian Learning Lab is an online platform launched in 2016 by the Smithsonian Office of Educational Technology (OET). Its digital museum resources are focused on providing educational materials for teachers, students, and museum educators.
We conducted a formal SEO audit of the site to understand how the current website is optimizing its SEO and provide recommendations to improve its discoverability on search engines.
-
Conducted Technical SEO
Wrote SEO Audit report
Presented to the client
-
On-page, Technical, and Off-page SEO
Google Analytics, Google Search Console, Looker Studio
Moz, Screaming Frog, PageSpeed Insights, Woorank, and Google Mobile-Friendly Test
-
Lillian Yang
Shih Wen Huang
-
Nov 2, 2022 - Dec 14, 2022
Project Goals & Scope
Throughout the Learning Lab's history, the OET team has focused on developing rich new resources aligned with students’ and teachers' needs. However, they have yet to focus as much as they need on ensuring that the Lab's technical infrastructure promotes high discoverability with popular search engines.
Therefore, this project aimed to understand the current performance of SEO by conducting a formal audit of the site and to provide recommendations for improvements.
The project was focused on the Collections pages, which are user-generated aggregations of museum resources and educational content with associated metadata.
Smithsonian Learning Lab Collection page example
Organic traffic has decreased by 71.9% compared to the last year’s.
The Looker Studio Dashboard’s data for this year (Jan 1, 2022 - Dec 31, 2022) showed the need for an SEO audit of Smithsonian Learning Lab’s website because the Organic Traffic Sessions have decreased by 71.9% compared to the last year’s.
The Smithsonian Learning Lab’s Collection page marked 66% of the overall success rate.
We conducted On-page, Technical, and Off-page SEO by assessing 24 items using an SEO Audit Checklist. Among 24 items, 16 were marked as passed (66.6%) while 8 were marked as failed (33.4%).
8 Findings & Recommendations
We found most failures in the On-page SEO category, followed by the Technical SEO category. Off-page category got a great result with only 1 failed checkpoint. From each failed area, we sorted out 8 main findings to suggest improvements.
Issues 1 & 2
Some page titles are too long and meta descriptions are missing
Some page titles were over 60 characters. Not only that, it automatically added the word “Collections with 2 Colons” at the beginning of the title. As a result, the search engine results only displayed a part of the title.
Secondly, proper meta descriptions were missing, so the search result showed the text content in its HTML body part. We grouped these two findings as we thought we could solve them together.
Recommendations 1 & 2
Restrict page title to below 60 characters and add meta descriptions
When the users create a collection and name it, it automatically becomes the collection’s page title. So, on the collection creation page, consider giving them a character constraint to the title so it can be under 60 characters. Also, place all the collection category information or the website information at the end to prioritize the unique collection title.
For any extra information, allow users to put it in a description or a subtitle, and make it serve as a meta description. The meta description will give search engines a good summary of what the page is about.
Issue 3
Several images are missing a descriptive file name and “alt” attribute description
There were several image files that were missing descriptive filenames / or the “alt” text. They are essential as they present alternative text for the image when it cannot be loaded or displayed for any reason.
Recommendation 3
Provide a descriptive filename and alt attribute description for images
To improve this, use short and descriptive names instead of generic or lengthy ones stuffed with keywords. Filling out its alt text helps search engines understand more about the page you're linking to. This will help the site rank in Google Images and drive traffic, as well as improve accessibility as the screen reader can read the image's content.
Issue 4
Some urls are lengthy and non-descriptive
Although URLs of the Learning Lab were doing great in containing descriptive page titles, some Collection URLs had unnecessary special characters and numbers at the end. This might make the URLs look heavy and not credible, especially when sharing the page link with others.
Recommendation 4
Remove unnecessary special characters and numbers
Remove unreadable special letters and numbers and only leave keywords that are relevant to the site’s content. This will create easier, friendlier URLs that are short and descriptive. Keep using a directory structure and the page title that helps visitors understand where they are at on the site.
Issue 5
Page load speeds are slow
Page speed is one of the essential element that impacts SEO performance. The longer it takes a website to load its pages, the lower the pages get ranked on the search results. According to Google PageSpeed Insights tool, both desktop and mobile’s site speed was assessed as fail.
Recommendation 5
Optimize Largest Contentful Paint
Our recommendation is to first improve Largest Contentful Paint, which is the primary metric that measures the time of the largest content to be visible relative to when the page first starts to load.
Using images in next-gen formats, reducing unused codes, and deferring offscreen images will improve LCP.
Issue 6
Robots.txt is lacking sitemap
Robot.txt helps search engines to crawl web pages. It indicates instructions that allows or disallows the web crawler’s behavior. Smithsonian Learning Lab had a robots.txt file, however, is wasn’t referencing to its XML sitemap, which also already existed. The robots.txt was allowing the search engines to access every content on the website but wasn’t providing additional information for search engines to understand the website’s structure.
Recommendation 6
Add sitemap location to robots.txt file
Since Smithsonian Learning Lab already had both robots.txt file and XML sitemap, adding the sitemap location to the robots.txt is the only thing to be done. Comparatively simple task to do, but this will greatly improve the search result of the site.
Issue 7
Search result data is not structured
Schema.org is a markup language that structures data in variety of ways. Web pages will appear more prominently on search results. However, we discovered that Smithsonian Learning Lab was not using Schema markups.
For instance, FAQ markup shows the content of the web page in a format of expandable accordions with questions and answers about the searched queries.
Even though we searched branded keywords like Smithsonian Learning Lab, or non-branded top searched queries like Hispanic Heritage Month, we couldn’t discover any content from the website.
Recommendation 7
Implement Schema markups
Therefore, we recommend to start implementing some of the Schema mark ups such as FAQ markup for better discoverability and engagement of the content of Collection pages. Through these markups, the search results will show rich snippets of information and will possibly improve the click-through rate as well.
Issue 8
Knowledge panel is missing
Google Knowledge Panel is a knowledge base from which Google serves relevant information in an infobox beside its search results. This allows the user to see the answer in a glance. However, Smithsonian Learning Lab has not yet created a knowledge panel which missed the opportunity to increase its visibility and attain more potential users.
Recommendation 8
Create a knowledge panel
We recommend to create a knowledge panel. First step is to open a Google Business account and verify site with Google Search Console. Then add structured data markup and ensure high-authority domain. By doing so will expedite the process for claim a knowledge graph from Google
Next Steps
Smithsonian Learning Lab had already implemented a survey to receive users’ feedback; however, the location was not discoverable as it was hidden in the footer section. If we had more time, we would try using a modal design or sending out subscription emails so that it could receive more feedback and tailor questions related to SEO performance.
We also would want to conduct benchmarking analysis on potential competitors’ (museum, public education resource, and digital collection) websites to see who Smithsonian Learning Lab is up against online and what current strengths and weaknesses are.
Finally, after implementing 8 SEO recommendations, we would like to continue using the dashboard to monitor the website's organic channel traffic to sustain a consistent and accessible user experience.
We presented our findings and recommendations to the Director and the Product Owner of the OET team, and they were delighted with the project result. They especially liked the spectrum of recommendations, including both immediately actionable items and long-term initiatives. We delivered a report and Looker Studio dashboard for the team to implement our recommendations and optimize the Learning Lab’s SEO performance.
Through this project, we learned that conducting SEO audit and digital analysis on organic traffic has excellent synergy. The data on organic traffic helped us understand current SEO performance at the beginning, so we could set a more precise goal. But not only that, it will be continuously beneficial as the organic traffic data could measure the SEO improvement recommendations’ effectiveness.