MICES 2017

Mix-camp E-commerce Search


14th June 2017
Starting at 8:30 am

Note to participants

If you are registered as a participant but are unable to attend the event, please inform us as soon as possible so that your seat can be given to somebody on the waiting list.

Join us
for a one-day workshop on E-commerce search

In the search community, e-commerce search has received far less attention and is less established than search in other domains, such as enterprise search or general web search. On the other hand, it is a very exciting domain with a lot of challenges. So we think that it is about time to bring together experts of different backgrounds (IT, product managers, UX designers, search managers, information retrieval specialists, search engine vendors, ...) for a one-day event to discuss challenges, ideas, best practices, and case studies in the e-commerce search domain.


The workshop will bring together participants from a variety of backgrounds, all sharing a common interest in e-commerce search. In order to stimulate and facilitate discussion, we will start with scheduled talks in the morning. This will be followed by self-organising sessions: participants are encouraged to initiate discussions about their topics of interest or to give an ad hoc presentation.


The workshop welcomes all topics related to e-commerce search. We have compiled a list of topics that will probably feature at the workshop. This list is not comprehensive. Participants are welcome to discuss further topics at the workshop.

  • Managing e-commerce search: finding the right process and organisation for it, and how does it change over time?
  • How can we measure and improve search quality for e-commerce?
  • What are the best search relevance models for e-commerce search?
  • Personalised e-commerce search
  • Choosing the right search technology and search management tools
  • From keyword search to virtual shopping assistants: designing the e-commerce search user experience
  • Exploring machine learning for e-commerce search
  • How to deal with poor data quality in e-commerce search





09:00 - 10:00

How to make E-Commerce Search great

David Urbansky

Web search has been improving rapidly over the last few years. Google and Bing have invested substantial resources not only in document-driven but also conversational search, with agents such as Google Now and Cortana. The search capabilities of e-commerce shops have comparatively stagnated and are often not even able to understand synonyms or correct simple spelling mistakes. Our talk will shed some light on several of these shortcomings, explain how to overcome them, and give our take on the future of e-commerce search.

David has been working on semantic technologies and search for over eight years. Right after he received his PhD in computer science, he co-founded SEMKNOX - a B2B semantic search solution for e-commerce that takes search to a new level.


10:00 - 10:15


10:15 - 11:20

A Methodology for Investing in Search Relevance

Doug Turnbull
OpenSource Connections

Search can be scary! Do you invest in that expensive product or big project? How do you know it will have returns for your search platform? Would simpler solutions pay more dividends? And how do you experiment with promising ideas safely without inadvertently harming search? At OpenSource Connections, we've developed a methodology based on best practices across hundreds of search consulting engagements. In this talk, we outline a "virtuous" cycle of continuous search improvement. With our radically iterative approach, ideas are proposed and tested both offline and live against business metrics. Ideas "fail fast" or they show promise --indicating how much you should invest further. Either way, our methodology helps you iteratively map out what works best for your search case, while avoiding expensive boondoggles.

Doug is author of Relevant Search and the Chief Relevance Consultant at OpenSource Connections. Doug builds smarter (and more profitable!) search for organizations like O'Reilly Media, Careerbuilder, and the US Patent and Trademark Office. Doug loves helping organizations get more out of search by bridging the technical with the business. He developed a search relevance methodology that guides businesses on making relevance investments. He loves creating relevance tools, including Quepid, Splainer, and the Elasticsearch Learning to Rank Plugin.


Search management workflows

Sebastian Russ

The sheer amount of search requests caused by the variety of users and products can be challenging. Thousands of different search terms make it impossible to get an instant clear picture of the quality of the search results. New trends and products constantly influence search terms and need to be considered when it comes to synonyms and search configuration.

Our talk gives an insight how to continuously improve the quality of search results by establishing an optimization workflow. We cover topics from the initial analysis to managing onsite search over a long period of time. The focus is on clustering and visualizing information to prioritize optimizations and to document progress.

Product search & design specialist Sebastian Russ is well grounded in search technologies as FACT-Finder, Fredhopper, Celebros, epoq and Solr. Working for TUDOCK in Hamburg he frequently analyses search behaviour. Based on his findings he provides advice to merchants about configuring and optimizing their onsite search, aiming for improved frontend usability, better search result quality and enhancing sale-relevant KPIs.


11:20 - 11:45


11:45 - 12:50

Simplifying and Improving Retail Search, Recommendations and Personalization with Lucidworks Fusion

Grant Ingersoll

If you’re in retail, chances are you’re probably getting crushed by Amazon. At Lucidworks, we work with top tier retailers trying to increase conversions and survive in the Age of Amazon. Our Fusion platform enables you to own every aspect of the search and shopping experience using best in class search and machine learning, all while simplifying your operations and keeping your team happy. In this talk with Lucidworks CTO and retail search expert Grant Ingersoll, you’ll learn:

  • How to leverage and own key user interaction data to deliver better search results, personalization, recommendations, and analytics using the Fusion platform
  • How Fusion leverages best in class open source tools like Apache Solr and Spark to reduce time to value and increase operational scalability and stability
  • How to simplify your operations and save significant money on operational costs
  • How to train, test, and deploy your scalable machine learning models for query intent, recommendations, machine learning, and search relevance enhancement without your key data ever leaving your data center.

Grant is the CTO and co-founder of Lucidworks, co-author of Taming Text, co-founder of Apache Mahout and a long-standing committer on the Apache Lucene and Solr open source projects. Grant’s experience includes engineering a variety of search, question answering, and natural language processing applications for a variety of domains and languages. He earned his B.S. from Amherst College in Math and Computer Science and his M.S. in Computer Science from Syracuse University.


From FACT-Finder to Solr - How we became search professionals within 12 months

Alexandre Burgard

In 2015 the MYTOYS GROUP developed its own shop platform, which is based on Solr for search and navigation. We will take you on a short journey to learn more about how we tried to beat FACT-Finder with Solr within three and a half months. You will get an impression of our ups and downs and why we now have set up a whole search team. Additionally, we will explain how we learned the difference between enterprise search and e-commerce search. Now our search team is becoming one of the best search teams of the Otto Group.

Alexandre works as team leader ‘product management’ at MYTOYS GROUP.


12:50 - 14:00


14:00 - 15:05

You got that search engine running. Now what?

Peter Rieger

Installing a fast and reliable search engine for your online shop is essential today. Just like a store clerk, it helps your customers navigate through your product range and converting them into happy buyers. But unlike a real person, your search engine does not memorize the things a customer did not find or whether he asked the wrong questions for his or her search. And it won't recognize the shelf full of things that never got asked for.

We will talk about why and how your search engine is important for your customer comprehension and we'll have a look at the treasure of data that your search engine is sitting on and how you can benefit from its power to gain valuable customer insights.

Peter works as managing director at ESEMOS.


Beyond precision and recall – considerations for better search experience

Andreas Brückner
Fredhopper, part of ATTRAQT Group plc

Getting on-site search right is often reduced to an algorithmic challenge. However, this can miss conversational elements as users struggle to articulate their needs. Only by combining the two, search can truly deliver. In his talk Andreas will discuss examples of how large online retailers have made their search more intuitive and effective for users.

Andreas is the subject matter expert for search and merchandising at Fredhopper, the on-site search technology provider trusted by many of Europe’s largest online shops. Andreas provides thought leadership and best practice advice to industry heavy-weights such as ASOS, Otto.de, Bonprix and Selfridges


15:30 - 18:00

Stop stupid fuzzy searches (breakout session 'Implementing Search')

Andreas Wagner
CXP commerce experts

Fuzzy logic is a must-have for almost any search application and especially in the case of e-commerce. Most search tools include fuzzy logic to handle plurals, misspellings, term decomposition and other near-matches. The goal of using fuzzy logic is to increase recall for a given search and ultimately improve results. But in reality we are often reducing precision and produce inconsistent results for queries with the same meaning.

At MICES 2017 CXP introduces a search tool independent solution which aims to make FUZZY SEARCH as fast, easy and accurate as EXACT SEARCH by introducing machine learning and query rewrites. In our talk, we present our language-independent approach to tackle the fuzzy search problem at scale. Furthermore, we open-source our baseline SmartQuery fuzzyLib during the MICES 2017 event.

Andreas has been building and optimizing search engines for almost 10 years. He worked for FACT-Finder and Fredhopper in the past and is now heading R&D for SmartQuery at CXP.


Performance optimizations for e-commerce search powered by Apache Solr (breakout session 'Implementing Search')

Tomasz Sobczak

There are some specific criteria in e-commerce search which all engineers need to meet: handling traffic peaks, keeping response times very short, managing continuous updates of products (as popularity, margins, availability in stock are changing all the time). This talk presents steps that can be taken to increase the performance of search applications. It will deal with issues connected with index segments, caches, creating data models and running queries. Although it will be illustrated with Solr examples, Elasticsearch users can still find something interesting - both are based on Apache Lucene and pretty similar on low level details.

Tomasz is a consultant providing search-driven and data-driven solutions. He helps organizations to store, manage and find information, to build search strategies and manage findability investments. He also works as a trainer and conducts workshops focused on software (Apache Solr and Elasticsearch) and on the business side of search projects (e.g. how to improve users' search experiences in organizations).



Isabel Drost-Fromm

Isabel works as Open Source Strategist at Europace AG. She is co-founder of Apache Mahout and mentored several incubating Apache projects. Isabel is interested in all things search and text mining with a decent machine learning background. True to the nature of people living in Berlin she loves having friends fly in for a brief visit - as a result she co-founded and is still one of the creative heads behind Berlin Buzzwords, a tech conference on all things search, scale and storage.

René Kriegler

René has worked as a search consultant for clients in Germany and abroad for more than ten years. While he is interested in nearly all aspects of search and NLP, his technological focus is on Solr/Lucene. René maintains the Querqy open source library.

Paul Maria Bartusch

Paul is a freelance consultant in the e-commerce domain. He supports his client as product manager, Scrum product owner and business development consultant, focussing on search (onsite, SEO/SEA), mobile and responsive application development, and agile transitioning.

Registration is closed

Unfortunately, all 70 seats at MICES 2017 are taken.

If you are already registered as a participant but are unable to attend the event, please inform us as soon as possible so that your seat can be given to somebody on the waiting list.


  • myToys Group
  • OpenSource Connections

Location and Venue

Hosted by myToys


myToys Group

Potsdamer Straße 192

10783 Berlin


Nearby Public Transport

Underground Train Stop (U-Bahn)

U7 Kleistpark

Overground Train Stop (S-Bahn)

S1, S2, S25 Yorkstraße

Bus Stop

106, 187, 204, M48, M85 Kleistpark

See www.bvg.de/en to plan your journey.