SMX 2019 - Recap - eviom
post-template-default,single,single-post,postid-58827,single-format-standard,qode-core-1.2,ajax_fade,page_not_loaded,,pitch child-child-ver-1.0.0,pitch-ver-2.2, vertical_menu_with_scroll,smooth_scroll,grid_1300,blog_installed,wpb-js-composer js-comp-ver-5.4.7,vc_responsive

SMX 2019 – Recap

It is already over again, the SMX 2019, which took place on 2. and 3. April in Munich. The SMX counts to the most important German SEO and SEA conferences.

Two days were filled with knowledge, recommendations for action, tips, inspirations and new food for thought, which were conveyed by online marketing greats such as Rand Fishkin or John Müller in exciting lectures.

For those who couldn’t be there themselves, we’ve put together a recap with our top 5 sessions:


Current developments and challenges in online marketing and how they can be mastered

It all started with Rand Fishkin, one of the most respected SEO experts. In his keynote, he spoke about current developments and challenges in online marketing and about strategies, how these can be mastered in order to survive in the future market.


Key Learnings:


1. The organic traffic via Google decreases continuously

  • the reason for this is that the search engine increasingly delivers the answers itself.
  • users therefore no longer need to visit a website to get an answer to their search query.
  • compared to 2016, organic clicks fell by 30%.


2. Organic traffic via social media platforms also decreases

  • social networks such as Facebook, Instagram and Co. are interested in keeping users with them as long as possible and not redirecting them to other websites.


3.  The return on investment of search engine advertising decreases

  • many companies focus more on growth than on profit.
  • the competition for the top positions becomes tougher and higher bids are due to remain visible with the ads.


4. Influencer marketing loses importance

  • influencers rarely need to provide numbers about engagement, etc. to customers.
  • this makes is difficult to determine whether influencers are really successful for a company.


In order to remain successful despite these challenges, companies should focus on building a strong brand and addressing customers again through tried and tested and classic online marketing measures such as email marketing.


Automation holds many changes but also opportunities for marketers

Jim Sterne, co-founder of the digital Analytics Association, strongly believes that machine learning will become increasingly important in online marketing and change the professions. Digitisation will help marketers to make their work even better. However, workers are NOT completely replaceable, since machines are very good at identifying correlations and patterns, but evaluations, interpretations, results and creativity can still be implemented by humans. Nevertheless, a lot will change in the near future, especially in the era of marketing.


Key Learnings:


1. Optimize content not only for web search, but also for voice search, augmented reality, facial recognition.


2. in the future, customers will deal with content in other environments, such as voice control in cars.


3. something will change in the handling of your own data: According to Sterne, in the future each person will have a central ,container’ in which all personal data is stored and can be accessed by various services such as Amazon, Facebook & Co.


4. in contrast to companies, central data storage is secure and trustworthy: It enables every consumer to have a personal assistant who knows everything about the person in question and worked for them.


Ecommerce Websites: The Importance of the User Experience

André Goldmann presented six approaches to UX optimization in his lecture. According to him, the most important areas are user interface design, information infrastructure, interactive design, usability, user research and content strategy.


Key Learnings:


1. create trust – but not only on the start page.


2. pack categories into the visible area.

  • subcategories are often overlooked by users.
  • if the user can not find what he is looking for, he leaves the site and continues searching elsewhere.


3. enable product related notifications through pre order feature.


4. represent products across devices.


5. make filters more user-friendly and align them to the needs of the user.


6. emotions count: Generate, sexy ‘content’.


In the cases many US shops were mentioned in the cases, including B&H. The German online shop, is also often shown as best practice for individual elements. The e-commerce giant Amazon may not be a gem, but it can shine with a 100% customer orientation.


Predictive SEO

How can SEOs make data-driven decisions? Sebastian Erlhofer dealt with this in his presentation. For him, predictive SEO means using data to predict optimization success in SEO. Everyone who operates search engine optimization is regularly confronted with three questions:


1. what should be optimized?

2. when should the optimization take place?

3. how intensive and complete must be optimization be in order to achieve optimal ROI?


The last point is critical because it is a budget issue. The costs invested in SEO must ultimately pay off. Calculating the potential on the basis of the status quo alone is not enough. Sebastian Erlhofer demonstrated in a case when and how everyone can use SEO Machine Learning. In this scenario, the objective was to make a CTR prediction for the keyword “buy shoes” when the keyword is optimized from position 8 to 3:


1. current CTR data is retrieved daily via API from the Google Search Console. For this test only data with more than 10 clicks and position 1-10 will be considered.

2. via node.js further information is taken from Google, which significantly influences the CTR: Are there many ads, a picture block, a video box or other featured snippets? Enriched results cause the organic hit in the SERPs to slide down. Therefore the height of these blocks is evaluated and considered with pixels via node.js.

3. his data from the SERPs is stored in JSON.

4. now the data must be processed. For this purpose, the data can be loaded into a machine learning program, which makes a prediction based on different algorithms.


This scenario is simplified. Since Google, for example, changes and tests many aspects of the SERP representation, the data must be refined several times until valid data can be predicted at the end. The data from the SERP compilation can be used in a variety of ways for predictive SEO.


Online Ads: New formats, updates and more


Brad Geddes and Navah Hopkins talked about the differences between Responsive Search Ads (RSA) and Extended Text Ads (ETA).

Studies show that RSA ads have a better click rate and should be used when the target is traffic or awareness.

ETA ads, on the other hand, have a much lower click rate, but a much better conversion rate. So it the campaigns have a conversion focus you should mainly use this kind of ads.



Also the topic headlines was topic of the lecture. If you look at the titles in the Google search results, they all seem pretty boring. CTAs like „buy now“ or „download here“, have had their day. User benefits and special features or USPs should be built into the headlines. The cases brought along by the speakers showed a significant increase in the click rate and conversion rate.


More tips for better click rates

In Germany, questions in the headline can also lead to better click rates. W-questions in ads are recommended especially in the B2B area.



The SMX Munich 2019 was again a great success and definitely worth a visit. We were able to take a lot with us during the various sessions, which took place at a consistently high level. Especially the detailed answering of open questions and the constructive exchange with the audience, as well as the active networking stood out for us. Now it‘s time to incorporate the latest developments and trends into our daily work. Until next time, SMX!