Updated AI Model to Improve Accessibility Suggested Solutions

Release Date: July 29th 2024

We’ve updated the AI model we use for our AI-suggested solutions for accessibility from OpenAI GPT-3.5 Turbo 16k to OpenAI GPT-4o Mini, to improve the quality of responses.

As always, these suggestions should be considered as guidance, and you should look carefully at the suggestion to make sure it’s the right solution for your site. You can also give us feedback on the suggestions within the app. Find out more about our AI-Suggested Solutions for Accessibility.

Block AI Setting

Release Date: July 23rd 2024

While Lumar does not send any sensitive crawl data to generative Large Language Models (LLMs), some of our AI features may send other crawl summary data to facilitate the suggested solutions or content. If required, you can now disable our AI features to ensure that no information is sent to any third-party LLM models.

You can do this in the account settings by unchecking the box labeled ‘Enable AI Functionality.’

Screenshots for Color Contrast

Release Date: July 23rd 2024

We’ve added a screenshot for color contrast to the issue details page to save time. This avoids having to navigate to the page and inspect the element in the browser.

In color contrast reports, where a value exists in the Background Avg Color column, you can see the screenshot by clicking on the Issue ID and scrolling down.

Shared Links Include Visible Columns & Sort

Release Date: July 15th 2024

When you generate a shared link, the recipient will now see the specific set of columns and sort order that were visible when the shared link was generated instead of the report’s default columns.

We already remember the columns you have selected for a report across all projects, which is saved in your browser’s local storage. When viewing a report with customized columns via a shared link, the columns will not be maintained after you leave the report and will return to your own personalized columns.

Robots.txt Settings Cleanup

Release Date: July 15th 2024

Lumar has multiple settings which can control the behavior for robots.txt during a crawl. These have now all be merged into a single settings group called ‘Robots.txt’ in the advanced project settings, in step 4 of the crawl setup.

Note: Because all pages impact the overall user experience and accessibility of a site, Site Speed (Lighthouse) and Accessibility crawls are set to ignore robots.txt by default. If required, you can change these settings here. 

Unavailable After Reports

Release Date: July 12th 2024

The unavailable_after robots directive is used to specify a date when a page will become unavailable and should be dropped from a search engine’s index.

We now capture the unavailable after value, and calculate the number of days difference from the date specified, and the time the page was crawled. A negative number indicates the unavailable after date is in the past.

We created 2 new reports that show you pages that have already become unavailable (unavailable_after Non-Indexable Pages), and pages that will become unavailable in the future (unavailable_after Scheduled Pages).

You can find these reports in the Non-Indexable subcategory (under Indexability) or by using the search reports function.

Invalid URL Lengths Increased

Release Date: July 12th 2024

We increased the maximum URL length before they are reported as malformed to 2048 characters in the URL path, and 10240 characters in the URL search parameters.

Pages with an Unavailable_after Robots Directive Reported as Non-indexable

Release Date: July 12th

Pages with an unavailable_after robots directive where the date specified is in the past when a crawl was run will be reported as non-indexable.

Canonical Tags with Invalid Attributes in the Static HTML Will Be Ignored

Release Date: July 12th 2024

Canonical Tags with invalid type, lang, media or hreflang attributes in the static HTML will now be ignored. This was already applied to canonical tags in the rendered HTML, however the ‘rendered canonical mismatch’ report could show an inconsistency where the canonical tag in the rendered HTML was ignored but in the static HTML it was not ignored.

Not in SERP Reports Updated to Include Pages with Null Impressions

Release Date: July 4th 2024

Pages returned by the Search Console API with null impressions were not included in either ‘Not in SERP’ or ‘in SERP’ reports. We have updated the following report queries to include these pages so you may see a small increase.

• Primary Pages not in SERPs (primary_pages_not_in_serp)
• Primary Pages Not in SERPs with Low Deeprank (primary_pages_not_in_serp_with_low_deeprank)
• Primary Pages Not in SERPs without Followed Links (primary_pages_not_in_serp_without_followed_links)
• Primary Pages Not in SERPs Not in Sitemaps (primary_pages_not_in_serp_not_in_sitemaps
• Primary Pages Not in SERPs with No Backlinks (primary_pages_not_in_serp_with_no_backlinks)

The pages will still be marked as ‘Found in Google Search Console’.

Feedback

As always, we’re keen to hear your feedback to help us improve the Lumar platform. You can do this very easily by clicking on the smiley face in the bottom left-hand corner of any of our apps.