atomix Team Insights: reflecting on 2019

Saturday, December 21, 2019 | By atomix
Australian GDPR Compliance

The final weeks of the year are a great time to reflect on all that we’ve learned during the past 12 months.

We’ve really enjoyed sharing our team learnings with you – and we hope you’ve enjoyed reading them! This month, we’ve been delving deeper into our reporting, discovering more about our friend BERT, and poring over Google’s freshly updated Quality Rater’s Guidelines.

Let’s dig in.

BERT goes global

On Monday 9 December, Google confirmed via their Twitter account that they had started to roll out BERT worldwide to 72 languages. ICYMI: BERT is Google’s natural language processing algorithm, and it helps Google to better understand conversational search queries.

Previously, Google had noted that BERT in U.S English had an effect on 10% of search queries – so we’re excited to learn how BERT affects searches in other languages.

Read more on BERT’s worldwide release


Google’s Updated Search Quality Guidelines emphasise diversity

In early December, Google’s search quality evaluator guidelines were updated, with a focus on diversity, impartiality and also a noticeable change in how the search giant refers to its users.

The biggest update was a new introductory section which explains why and how people – notably not “users” – conduct web searches. It delves into how searches should have results that reflect not only the diversity of the world we live in but reflect the diversity of the people that use search.

Read the full summary here


Google Search Console performance reporting

Google Search Console has started to send monthly Search Performance reports via email to the property owner. The reports summarise clicks and impressions from organic search, top search landing pages by growth and overall performance, top-performing search queries and demographic breakdowns.

Learn more about these reports


Local Search Algorithm Update: neural matching

In early December, Google confirmed an update to local search, where they added neural matching to local searches. This means that Google can better understand the intent of the search, and how words are used to change the meaning of a search query. Google Local Search’s three pillars of relevance, prominence and distance haven’t changed: what has changed is Google’s understanding of relevance with neural matching.

Read more about the local search algorithm update


Search results page user study

There’s no doubt that featured snippets and ‘position zero’ are hot spots to grab on the search engine results page (SERP). A recent study of the search engine results page aimed to discover how these new features have changed searchers’ behaviour. The value of the absolute first spot on the results page is based on the assumption that users read the results from top to bottom. The study found that, due to the new featured snippets and larger content areas displayed on the SERP, users are literally ‘bouncing’ between different elements as they see them, not necessarily in a linear fashion. UX leader Nielsen Norman Group has named this “the pinball pattern”. This shows us a clear reason why we need to optimise toward search features: to increase brand visibility.

Check out the results report here


Have you noticed any changes since BERT has been rolled out?

What about your local search results performance – have you seen any changes?

We hope you have a fantastic holiday break, and we look forward to working with you again in 2020!

The atomix growth team

Josh, Sarah, Georgie, Chloe, Liz and Adam