Given our profile as an analytics studio, conferences like MeasureCamp or Data Restart are a must for us, and so even in our busy schedule we do our utmost to be at them. Read what caught our attention and how the whole MeasureCamp went.
Based on recommendations from people in the field and our experience with last year’s MeasureCamp in Brno, where we really started to get into who does what, where and how, we put Bratislava in our calendar too.
MeasureCamp is wonderful in that it still keeps the unconference format in the sense that the programme really is created on the spot (Barcamp originally worked that way too, but dropped this approach). This brings nothing but beautiful moments: seasoned pros arrive with prepared slides and printed handouts, newcomers to MeasureCamp let themselves be tempted into showing something from their drawer, and above all there is a wonderful atmosphere throughout the day.
GULP for analytics deployment – Lukáš Čech, etnetera activate
The first talk I attended was by my namesake from activate, about an approach to deploying code to client websites using Node.js with the GULP module. So I started the day nicely with some hardcore programming. Lukáš explained the history of automation at activate and the reasons why a similar approach may be desirable from the client’s point of view too (security, versioning, multiple environments). As the main advantage of why they create some of their own libraries for conversion measurement for Sklik, Facebook and other services, he explained absolute control over measurement without the need for 3rd-party JavaScript. When we talked afterwards, I understood that this path may await us very soon in situations where Google Tag Manager stops being enough for larger clients, but for that we need a programmer. If anyone is bored, write to us.
Personalizacia cez GTM by Daniel Duris
I continued with a talk by Daniel Duriš from Basta Digital, who showed a nice and working proof of concept for personalising a website (banners, texts and more) through GTM and an associated JavaScript library. In the discussion that followed we agreed that this path can be very fragile and that even in the testing phase they met a number of problems (missing images, incomplete information, etc.). The proof of concept addressed the situation where larger companies, especially multinational ones, are not fast enough at modifying their websites, and a test like this can be a huge benefit precisely for verifying and testing an idea. I was thrilled and am already making plans for a test of our own.
Diffrence in conversion between Facebook, Google Ads and Analytics by Tomas Malik
In this talk, which followed after the coffee break, Tomáš Malík spoke beautifully about the different approaches to perceiving conversions in Google Analytics and in advertising networks; it all of course relates to attribution. The takeaway is that you need to realise that a conversion in Google Analytics, Google Ads, Facebook Ads, Sklik or another advertising network is each time something different and is measured by a different methodology (a different conversion window, a different action, a different way of reporting). So it is often not possible to compare these conversions directly and to resolve their mismatch, very often a time mismatch. We then went over attribution as a refresher and the fact that data-driven attribution is not necessarily a cure-all for everyone. You need to test, analyse, optimise and decide in proportion to your project and your options.
How to Exclude yourself from Web Analytics in GTM by Robert Petkovic
This talk confirmed our approach to excluding yourself from Google Analytics using cookies and URL parameters. Robert and I had a great time exchanging ideas on how the solution can work and how to improve it (visual feedback for the user, filtering reports in Google Analytics and different traffic segments). In principle the solution consists of storing a cookie based on a parameter in the landing URL. The solution of course has its weak points too, but it appears to be the most widely used way at the moment.
Releasing the Power of Google Analytics filters by Carlos Escalera Alonso
Alonso, in his very nicely prepared talk, summed up beautifully how to work with filters in Google Analytics and how to use them correctly. For us a nice summary of procedures already established, and confirmation that we set everything up the best way possible.
Please always create analytics documentation by Marek Cech
After the lunch break I presented our approach to documenting the setup of analytics tools, and together with the participants we grumbled about how time-consuming it is and what it could contain. I highlighted its importance and the reasons to create it. We reached a consensus that documentation must start already with the brief of the given business case and relate everything to it. A very useful discussion and confirmation that our approach is a step in the right direction. It costs you more time now, but in the future it will pay you back several times over.
How to create a perfect dashboard? by Mateusr Muryjas
A wonderful talk on visualisations in Google Data Studio. It contained a multi-step process showing how to think about visualisations in general, followed by practical demonstrations of dashboards. For me, as the product owner of our Data Visualisation service, it was a wonderful talk. I cannot wait to see what we come up with together with its author.
Panel discussion: AI in marketing by Jim Sterne, Peter Jakus, Matus Cimerman
A panel on AI and machine learning followed, which I had been looking forward to since the morning. My expectations were met. With buzzwords like AI I am always afraid of a completely warped approach to the topic. That did not happen here: Jim Sterne, a long-time ambassador of digital analytics, and Matus Cimerman, who opened the Exponea office in London and also worked on AI at Amazon, discussed the topic in a very sophisticated way, and I finally understood what it is about. It was explained very simply when it makes sense to replace people’s work, or dumb rule-based intelligence, with artificial intelligence – it is mainly about the number of attributes a person would have to evaluate to arrive at the result. The main benefit can be more accurate recommendations of products or services on the site or in mailing, offering suitable products after an abandoned cart, segmenting customers on the basis of discovered behaviour patterns, and so on.
Screaming from aneb co to ta zaba umi. (CZ|SK lang) by Martin Zatkovic
At the end of the day I came to listen to a colleague from Marketingová Plzeň, Martin Žatkovič, who spoke very nicely and above all showed the capabilities of the Screaming Frog tool, why to fall in love with it and how it can help not only with SEO.
All in all, MeasureCamp Bratislava 2019 was a wonderful event, which I look forward to again next year. It brought us many interesting contacts and ideas, and the reassurance that the path we have taken is the right one and that there is always something to work on. I was surprised by what interesting companies and projects exist in Slovakia and how well they are doing. As usual I was delighted by the openness of all the colleagues from the field whom I met there. Weekend projects pulled out of drawers, working case studies and proven practices, that is exactly what MeasureCamp is about.
Thanks to the whole organising team and the other participants for a great experience.