Just get on the plane and let us bring you into a week of pure digital analytics. We will shuttle you straight from the airport to the conference venue where your days will be filled with seminars and events that SUPERWEEK has become synonymous with.

GDPR & EPRIVACY: WHERE ARE WE TODAY AND HOW TO MITIGATE DATA PRIVACY RISK

Keynote by Aurélie Pols / Mind Your Privacy - Madrid, Spain

MACHINE LEARNING ANALYTICS: DO TRY THIS AT HOME

Viktor Tarnavsky / Yandex Metrica - Moscow, Russia

Nowadays businesses have too much customer data for a human to handle. Sifting through heaps of it manually trying to find insights is not only gruelling, sometimes it's no longer possible. Solution: let the self-learning algorithms find patterns in data that yield business value. We'll show how a major online retailer, a DIY store and an airline booking service cooked a conversion prediction system out of their various sources of data – CRM, web, and app. This won't hurt, almost anyone can do it – and we'll demostrate how.

PHILOSOPHERS WITH LAPTOPS

Damion Brown / Data Runs Deep - Melbourne, Australia

From Plato to Alain De Boton, philosophers have been the people tasked with finding truth, and bringing it to the masses. To a web analyst, that pursuit of truth should sound familiar: it's the job of web analytics to shine a light in the corners that reveal truth about human behaviour. In this fun and informative talk, Damion looks at how the history of philosophy has parallels with analytics best practice, and along the way ponders on whether the Digital Analytics industry might just have found an incredibly rare thing that nobody thought existed: an actual use for a philosophy graduate.

GETTING A GRIP ON ARTIFICIAL INTELLIGENCE FOR MARKETING

Jim Sterne / Digital Analytics Association - Santa Barbara, USA

AI and Machine Learning will become an integral part of your marketing analytics life so before Matt Gershoff explains how it works, Jim walks you through what it is and how it is being used. From natural language processing and computer vision to chatbots and robots, you'll see how AI is applied to customer interaction. Then, Jim dives into machine learning so you can determine which software services are worth your time, communicate better with the data scientists in your company, decide to become one yourself, and figure out how and where to bring AI and ML into your marketing tool suite.

THE ART OF WAR: GETTING READY FOR GDPR

Prolet Miteva / Autodesk - San Francisco, USA

"In the midst of chaos, there is also opportunity". GDPR is coming and being prepared is not an option, it’s a requirement. Prolet will walk through the challenges presented by the new regulation for all of us, from in-company struggles to winning the battle with, and for the user.

The topic will center around the knowns and the unknowns of GDPR - our fears, worries and opportunities in the journey to compliance.

TURN TO THE SPEAKERS!
MACHINE LEARING AND ARTIFICAL INTELLIGENCE IN 2018

Matt Gershoff (Conductrics - New York, USA)

Jim Sterne (Digital Analytics Association - Santa Barbara, USA)

Viktor Tarnavsky (Yandex Metrica - Moscow, Russia)

Mark Edmondson (IIH Nordic - Copenhagen, Denmark)

Andraž Štalec (Red Orbit - Slovenia)

FIRESIDE CHAT: "ANALYTICS NIGHTMARES"

Julien Coquet / Hub'Scan - Montélimar, France

Join renowned expert Julien Coquet around the fireplace as he takes the role an a digital analytics Gordon Ramsey. Like most digital marketing projects, digital analytics projects suffer from a lack of vision and planning. This leads to poorly executed projects that yield poor results. Sometimes, analytics become a concern after a new manager arrives, only to discover how bad things turned out! Julien will share stories about the worse analytics situations he ever encountered - along with simple yet effective solutions to these problems.

THE PATH FROM MICROSOFT EXCEL TO IBM WATSON: EVOLVING HOW WE THINK ABOUT AND USE DATA

Tim Wilson / Search Discovery - Dublin (OH), USA

The data available to marketers -- literally at their fingertips by way of a few mouse clicks -- has exploded over the last decade. Yet, while there is more data -- and it is more accessible -- than it has ever been, the way we think about and use data has hardly evolved at all. With the recent advances in cloud computing and processing power, the industry is abuzz with talk of machine learning and artificial intelligence. How, then, will we get from the world of Microsoft Excel (or Tableau) to a world where "the machines" are automatically and dynamically optimizing all aspects of our marketing?

In this presentation, Tim will make the case that simply "waiting for the magical black box to arrive" is a risky strategy and that, rather, marketers and analysts should dive into (the shallow end of) the data mining and statistical analysis pool to not only get greater value from their data now, but to set themselves up for the machine-driving marketing of future.

CURRENT STATUS OF ID GRAPHS AND THEIR APPLICATIONS IN ANALYTICS

Kristoffer Ewald, Matt Stannard / MetaPeople Group - Switzerland

The talk will answer the core questions of how we can balance a better customer centric understanding of individuals behaviour over time with respect for privacy. As technology changes rapidly the technologies will make it possible to build complete new types of "device" and ID graphs while the respect for privacy tightens this needs to be done to be benefit of consumers. In this talk we'll look at the "why, what and how" and cover:​

• Classes of graphs
• Practical usecases of platform API's
• Designs ​of graph arrays
• Privacy vs. benefit
• Future outlook

MACHINE LEARNING AND HUMAN INTERPRETABILITY: THE COST OF COMPLEXITY

Matt Gershoff / Conductrics - New York, USA

Machine learning can be thought of as sitting at the intersection of computer science and statistics. "Computer Science has focused primarily on how to manually program computers, Machine Learning focuses on the question of how to get computers to program themselves ..." – Tom Mitchel

For many tasks, it makes relatively little difference if these programs are opaque to human introspection. We are almost exclusively concerned with performance for some task ("is this a cat?"; Donkey Kong, etc.). Here, high capacity models, like deep learning, suffer little penalty for marginal increases in representational complexity.

However, for several reasons, marketers tend to be wary about ceding control of their customers’ experiences to black box methods. Firstly, they need assurances that they can trust automation to make reasonable decisions over a large space of possible environments. Secondly, companies often have internal and legal regulatory requirements around the how, and which, data may be used for making marketing decisions. With the EU’s GDPR coming into effect in May 2018, this will be even more of a concern.

We will cover, at a conceptual level, simple linear models, decision trees, and deep learning. I will argue, that while lacking the expressiveness, or capacity, of more complex representations, linear models and decision trees have many appealing properties for the marketing use cases:

Human readable: for any given input, a human can easily determine the output.

Auditable: for regulatory review and approval before use.

Loggable: because each decision node in each decision tree can be uniquely identified, it is possible to register, and retain a history of what policy, or rule, was used for every decision event. This is particularly useful for significant decisions as defined by Article 22 of the GDPR.

MEASUREMENT FOR GROWTH

Krista Seiden / Google Analytics - San Francisco, USA

Growth marketing has been all the rage, but what is it? Is it new age optimizers looking for a fancy title? Marketers spinning their services in a new way? A buzzword for a game of bingo? User lifecycle marketing? In this session, Krista walks you through what Growth Marketing is, why it's important, how it's different, how we can all do it, and most importantly, how measurement works for growth.

LET’S TALK ABOUT ATTRIBUTION

Peter O'Neill / Leap Three Analytics - London, UK

All attribution models are wrong, but some are useful. Except for Data Driven Attribution models of course which just might be right. Correct? Or not? Peter wants to reframe the discussion so it is not about attribution models but about techniques for optimising your marketing spend. He will talk about why the technique of attribution is wrong and where to invest your time/resources instead.

In this session you will learn:

• A method for understanding and explaining different attribution models
• The inherent flaws with the technique of attribution
• Alternative approaches for optimising your marketing spend

YOUR KPIS SUCK – THE FINE ART OF ASKING LESS STUPID QUESTIONS

Steen Rasmussen / IIH Nordic - Copenhagen, Denmark

With the abundance of data available in analytics we are no longer competing on the volume of data we have access to, but the quality of the questions we ask. The challenge is that we have never been taught to ask or taught other people to ask the right questions from analytics. So we spend our lives answering questions like "what is our bounce rate?" or "how much did the last campaign convert?" when we could be asking important and business critical question. This means that we in analytics have to consider putting the KPQ (Key Performance Question) before the KPI (Key Performance Indication) to start asking better questions.

ANALYTICS FOR SEO EXPERIMENTATION

Martijn Scheybeler / Postmates - San Francisco, USA

How you get from 1 keyword to millions and how do you work with all of this analytics data at scale to make sense out of it for your SEO & Analytics strategy. At Postmates, Martijn is heading up the SEO team; he’ll talk about how millions of keywords are helping them make the right decision and how they’re dealing with all of this analytics data and is using it to run more efficient experiments.

DATA-DRIVEN ORCHESTRATION OF COMMUNICATION MECHANICS

Victor Krylov / Exponea - Moscow, Russia

Omni-channel is not a new hype, it's the norm. We have access to omni-channel analytics, omni-channel communication and automation tools. Now we connect that and visualize this connection. Segmentation is needed, not only for people, but for marketing automation.

• Build lifetime customer journey diagram, enabling marketer to orchestrate communication mechanics.
• Orchestrate a trigger-based communications strategy and various communication mechanics.
• Divide customer lifetime into different stages of relations in business, analyse it separately and connect those segments.
• Build customized communications for each stage and systematically push/pull customers toward the next stage.

TURN TO THE SPEAKERS!
2018 - THE GDPR-YEAR

Aurélie Pols (Mind Your Privacy, Madrid, Spain)

Anjali Raghavan (Google Analytics - London, UK)

Lenka Gondová (Exponea - Bratislava, Slovakia)

Prolet Miteva (Autodesk - San Francisco, USA)

mod.: Yehoshua Coren (Analytics Ninja - Jerusalem, Israel)

DANIEL WAISBERG'S BIRTHDAY BASH

LET'S ACTUALLY USE ANALYTICS TO DO SOMETHING

Charles Farina / Analytics Pros - Seattle, USA

There is no shortage of guides and tutorials about how to track everything from videos to the weather (Thanks Simo!). There are very few examples of anyone actually using Google Analytics to take actions and drive change. In this session, I will share my favorite examples from landing page optimization, content ideas, to marrying qualitative with quantitative data. In addition, everyone talks about attribution, but rarely is anyone able to show why it matters for their business. I'll share quick examples to show how to start a meaningful conversation around attribution and show why it matters.

WEB ANALYTICS: THE MISSING MANUAL

Mariia Bocheva / OWOX - Kiev, Ukraine

Websites come in many types, styles, and budgets. Each website, be it a news site, an online store or a personal blog, has its own performance indicators. These may include the number of unique visitors, average order value, conversion rate etc. Web analytics systems are aimed to help track those indicators, analyze them and make decisions. But how you can trust them if the data collection process is screwed and data is scattered. During this speech, you'll get a checklist for your project that covers all the main mistakes in the project settings and analysis. Save your time and learn from the mistakes other teams have already made.

VISUALIZATION PRINCIPLES TO UNDERSTAND & COMMUNICATE DATA

Daniel Waisberg / Google - London, UK

In this presentation, Daniel will talk about data visualization techniques to effectively uncover insights and communicate data. He will talk about charts best practices, types of visualization for different purposes, and some design principles.

WHAT TO DO TO WHEN YOU'RE STUCK IN AN ANALYTICAL DEAD END?

Črt Podlogar / Red Orbit - Slovenia

There are typically two issues that analysts (especially beginner and intermediate level) are faced with, when trying to extract actionable and useful insights out of their data. The first one - data is irrelevant. The second one - data is wrong.

Črt will share an analytics approach that solved the problem of irrelevant data through context. Sometimes pure numbers and volumes of aggregated data about users, cohorts, distributions and behavior will simply not tell you anything actionable that you can work with. When that happens you can attach context to your data with a few specific content changes and a few short and simple questions that you ask your website users. The other problem that many are faced is having plain and simply wrong data. This can happen for a multitude of reasons - poor understanding of theory and analytical concepts, mistakes in implementation or misconception or wrong understanding of the user behavior on the website. Črt will share cases and examples, where we made mistakes in our analytical approach, how we found out where the issue was and how we went about fixing it.

TURN TO THE SPEAKERS!
DATA INTEGRATION

Prolet Miteva (Autodesk - San Francisco, USA)

Tim Wilson (Search Discovery - Dublin (OH), USA)

David McBride (IBM - San Francisco, USA)

mod.: Michael Helbling (Search Discovery - Atlanta, USA)

GOLDEN PUNCHCARD PRIZE

Demonstrate any Digital Analytics solutions or method of your own that is way beyond the defaults. Who decides who's gonna win? The audience. Send your nomination to 2018@superweek.hu!

FIRESIDE CHAT: THE DIGITAL ANALYTICS POWER HOUR...LIVE!

The top-rated explicit digital analytics podcast returns to Superweek to do another live recording. Michael Helbling and Tim Wilson have been joined by Moe Kiss since last they were in Hungary, and all three will take center stage with toasty embers on their posteriors and their largest set of guests ever in front of them: the Superweek attendees! The evening will be part mid-conference recap, part diving deeper on a few of the discussions they've had with attendees, and a whole lot of fun... with drinks in one hand and microphones in the other.

WHY YOU SHOULD FIRE YOUR MEDIA AGENCY

Charles Farina / Analytics Pros, Seattle + Martin von Wenckstern / Toronto, Canada

If you work with a media agency (or are one) the first question to ask them is how many data scientists do you have? Do you prefer Amazon Web Services, Microsoft Azure, or the Google Cloud Platform? Come see examples from one of Canada's largest retailers of advertising spending that is wasted from poor targeting, access issues, and lack of big data understanding. We will also dive into examples of broken implementations of Analytics that cause even more issues. If you are not in-sourcing the core components of your Media and Analytics you are almost certainly at risk or already suffering from many of these problems. In this session, Martin and Charles Farina will show you what you need to find the right partner, but more importantly what you also have to provide.

DATA WORKFLOWS USING APACHE AIRFLOW

Tahir Fayyaz / Google Cloud Platform Team specialising in Data & Machine Learning, BigQuery expert - Google, UK

Building and maintaining data pipelines and workflows can be very complicated, difficult to manage and error prone. AirBnb created and open-sourced Apache Airflow in 2015 to solve a lot of the problems. If you have ever needed to automate, schedule, handle errors or send out alerts for your data workflows then Airflow will be the perfect new tool in your data toolbox.

Starting with the basics and then diving into some complex examples for Google Cloud tools like BigQuery and Dataproc, this session will help you get started with Airflow and show how you can start using in your own workflows.

RETURN ON ANALYTICS IN A SHORT TIME-FRAME

Peter Jakuš / Exponea - Bratislava, Slovakia

Several companies see the need to be data-competent, however, there is a struggle in the implementation of analytics strategy due to various factors, including bureaucracy and the lack of organizational agility.

Peter will talk about successful cases that the team has experienced while dealing with the implementation of marketing and data strategy into everything from startups to corporate dinosaurs - covering verticals from telco, fashion, finance, travel and other e-commerce. You will learn how to "hack" company politics, find allies and facilitate the necessary mindset with stakeholders in order to bring the company to the next level of digital analytics maturity.

ANALYSIS OF COMPETING HYPOTHESES

Moe Kiss / The Iconic - Sydney, Australia

Moe shares the most valuable analytical technique she learned from her time in government – ACH. Often, confirmation bias (the tendency to interpret new evidence as confirmation of one’s existing beliefs) railroads an analyst’s findings. ACH draws on scientific methodology, cognitive psychology and decision analysis to avoid this analytical trap. The strength of ACH is that it is difficult for an analyst’s natural biases to influence the outcome of their analysis. Moe teaches how to build your own hypotheses matrix and set up to disprove hypotheses using a case study. And yes, she’s applying ACH to her work at THE ICONIC, Australia & New Zealand’s largest online retailer.

MACHINE LEARNING, AI AND PREDICTIVE ANALYTICS IN E-COMMERCE

Andraž Štalec / Red Orbit - Slovenia

The increasing complexity of digital landscape on one side and huge business expectations on the other side are the driving force of change in e-commerce. Fueled by tons of data machine learning and artificial intelligence are slowly becoming the norm. But algorithms themselves won't be able to change the companies and deliver success. Entire companies need to change as well. How to embrace this change? Where to start and what to expect? How to organize yourselves? We'll deep-dive into data-driven digital marketing framework, followed by insights and case studies from clients and finish up with a stack of tools and takeaways you can use to produce some quick wins.

Key takeaways:

• Insiders look into roadmap to AI driven e-commerce
• List of organizational prerequirements
• Key concepts of data-driven organization
• Stack of processes and tools that enable and drive AI driven e-commerce
• Insights from our past projects that you can implement tomorrow

USING A MOBILE PHONE AS A THERAPIST

Peter Meyer / Nordea Bank - Copenhagen, Denmark

Can your mobile apps help you learn about yourself?

A lot of apps track what you do, where you do it, and how much time you spend doing it. You can quite easily access datasets already provided for your own exploration, for instance from activity trackers and location sharing apps. Maybe that information can learn you something, if you are willing to explore it a bit.

In this presentation, Peter will "take his own medicine", and have a look at R-generated illustrations of personal datasets, originating from apps installed on his own mobile phone.

GOOGLE ANALYTICS IN BANKING ORGANIZATIONS

Gerasimos Nikolopoulos / Growth - Athens, Greece

Google Analytics certainly provides us the opportunity to track everything on a website. But how can we take advantage of this opportunity, within such an enterprise, to drive change and improve the quality of the services it provides? Both online and offline. In this session we will go through the four major pylons of such an endeavor:

• Web & App optimization
• Data driven Design
• Integrated data from all sources

PROGRAMMATIC CAMPAIGN OPTIMIZATION FROM A DATA SCIENTIST'S PERSPECTIVE

Gergely Kalmar / Webrepublic - Zurich, Switzerland

The European digital marketing industry is inevitably shifting towards programmatic advertising, which poses difficult challenges for all involved parties – publishers, marketeers and analysts alike. As the technical infrastructure is getting more complex, it is increasingly difficult to properly collect, integrate and analyse the appropriate customer behavioral data sets, while those players who succeed first get a strong competitive advantage on the market. Luckily, it is not required anymore to employ a swarm of data engineers and data scientists to get started, as many of the critical business questions can be answered with relatively little effort using off-the-shelf tools.

In this session we will review the data landscape related to programmatic advertising, while also learning about the related analysis techniques that any company should be able to deploy from small businesses to international corporations. We will specifically focus on mastering audience clustering as well as on models that can be used for multi-channel marketing mix optimization initiatives.

TURN TO THE SPEAKERS!
MEASUREMENT TECHNOLOGIES, TOOLS, MARKETING CLOUDS AND SYSTEMS

Moe Kiss (The Iconic - Sydney, Australia)

Charles Farina (Analytics Pros - Seattle, USA)

Tahir Fayyaz (Google - London, UK)

Kristoffer Ewald (MetaPeople - Zürich, Switzerland)

Michael Helbling (Search Discovery - Atlanta, USA)

DO YOU NEED AN ATTRIBUTION STRATEGY?

Jeff Sauer / Jeffalytics - Planet Earth

Marketing attribution is a buzzword that has everyone the paid advertising community talking. But what does it mean for you? Is it really a magic bullet, or is attribution just a bunch of bullshit? This talk goes through the need for attribution, shares practical applications and technologies, and ends with a framework for you to get started.

OMNICHANNEL PERSONALISATION AND HOW DIGITAL DATA IS NOTHING ON IT'S OWN

Rob McLaughlin / SKY - London, UK

Digital channels offer organisations the ability to recognise customers and personalise experiences. What are the opportunities in personalisation? What are the risks associated with failing to deliver personalisation?

Rob will expose the journey he has taken the Sky business on to date, the wins, learnings and distractions which have made this area the most challenging and rewarding in his career to date. Many organisations talk about personalisation, this is the story of how a long established organisation embraced relevance and customer centricity to drive business objectives. Without spoiling the surprise; this is about making change happen at speed & scale, not running a complex multifaceted programme. Major themes will cover business design, outcome and objective setting, the role of technology and data, evaluation and performance.

Key takeaways:

• Failure to act of customer recognition carrys significant risks, don’t cause offence by ignoring your most treasured assets
• Be true to your objectives, use relevance to drive your business, personalisation is not a trick or a gimmick
• Identify key technologies and data sets in your organisation and bring them together with the right mix of people, change will follow quickly

GETTING PERSONAL - CUSTOMER PROFILING WITH MACHINE LEARNING IN BIGQUERY

Mark Edmondson & Casper Radil / IIH Nordic - Copenhagen, Denmark

With BigQuery business and organisations have a unique chance of taking there analytics data and start the transformation towards a data lake. By combining customer, analytics, marketing and CRM data here we not only get a repository where can have room to add or work with data as we see fit, we also open up for the opportunity to use machine learning to actually sift through our data to help determine the causality and relationship between the individual data points. This way we use the full power of data to define our segments and profiles based on their actual behavior and not our prejudice.

WHAT IS THE ROI OF NOT MAKING BAD DECISIONS

Russell McAthy / Cubed - London, UK

ROI isn’t a static line in the sand. Analytics guides our constant strive for improvements. How can we change the conversation about the ROI of analytics for it to become a foundation for decision making.

GOOGLE ATTRIBUTION AND ATTRIBUTION360: HOW THE NEW SOLUTIONS WILL CHANGE ATTRIBUTION CONVERSATION IN 2018

Stanislaw Magierski / Google - Warsaw, Poland

TURN TO THE SPEAKERS!
WHERE ARE WE AS AN INDUSTRY IN 2018

Doug Hall (ConversionWorks - London, UK)

Jim Sterne (Digital Analytics Association - Santa Barbara, USA)

Russell McAthy (Cubed - London, UK)

Jeff Sauer (Jeffalytics - Somewhere Around The World)

CLOSING KEYNOTE

Doug Hall / ConversionWorks - London, UK

This is my fifth SUPERWEEK! My first was back in 2014 and having attended in 2015, '16, '17 and now '18 this is an event I just can't miss. SPWK is recognised as the event that starts the year off for the analytics industry with a massive bang. We all love the fireside chats, golden punchcard prize, live podcasts, sharing a drink with superstars of the industry and of course, the famous night time bonfire and mulled wine!

But, of course, the main reason for coming to this amazing venue is the programme of speakers and the incredible value we get from the talks and demonstrations. Since my first visit we've seen attendees and speakers from over 30 countries across 5 continents covering an amazing variety of subjects in extraordinary detail. 2017 was a particularly rich affair that I found hugely insightful. I learnt a ton of new stuff. I was inspired and totally motivated.

I want to describe the highlights from 2017 and demonstrate how ConversionWorks took the new techniques and ideas and delivered value and innovation for our clients. This isn't a self congratulatory back slapping exercise though, because I want to show you how SUPERWEEK will change you, your business, your clients and your total outlook on your career for the better. I'll show you how the big ideas changed our thinking, how the things that raised the bar elevated standards and how the moon shots...worked!