All posts filed under “large-scale data analysis

Intercom’s 1st Data Report! Emoji Trends in Business Messages

Hero_Emoji-Report

Today I published Intercom’s first ever data report with my colleague Sara Yin where we explored emoji trends in business messaging. We analyzed more than two million anonymized conversations that took place between our customers (i.e. businesses) and their end users (i.e. consumers) during a 3-month period, from June to August in 2015 and 2016. Below is a just a couple of the interesting things we found. You can read the full report here and a fun blog post here.

#1 Emojis improve message engagement

We discovered that messages started by a business that contained an emoji were four times more likely to elicit a response from a consumer than those that didn’t!!

#2 The Top 20 Emoji

The top 20 emoji found in messages in 2016 were quite similar to ones you might use in your personal life. Category-wise, over a half (51%) of the top 20 emoji fall under a facial category, followed by object-based emoji (18%). As you might expect in a business setting, money-related emoji also featured (11%).

top20_2016-v21

When we compare the top 20 emoji used by businesses vs. consumers we find some real differences! Consumers used facial emoji 30% more than businesses did (83% vs. 51%); businesses stuck to objects (18%) and money symbols (11%), neither of which show up in the consumer list.

Pie_Chart_Blog

Excited to be joining Intercom as Product Analytics Manager

After 2.5 years in Yahoo Labs, I’ve decided to leave the world of industrial research labs and  join an exciting, mature Irish startup called Intercom as Senior Product Analytics Manager. While I have thoroughly enjoyed my experience and journey in Yahoo and have learned a great deal from my colleagues there, it’s time for a new challenge and the next chapter.

At Intercom my task is to help build up and lead a product analytics team who will work closely with product, design and research teams to generate insights and metrics that will inform the design, development and measurement of Intercom products. The role starts here in San Francisco but is based in Dublin so in a few months time my family and I will be moving back home to Ireland after almost 8 long years away!

Overall I’m super excited about this next venture. In particular the opportunity to work at such an exciting company who cares deeply about user experience, the challenge of leading a product analytics team and shaping a culture around data, the potential to have such a positive impact on Intercom’s products and of course the opportunity to return home to Ireland.

Intercom’s mission is to Make Internet Business Personal. And they are achieving this through an integrated platform of products that enable businesses to acquire, engage, learn from and support their customers.

We’re hiring!!! So if you’re interested in data science, quantitative research and analytics with a passion and interest in product, then please get in touch! Here’s to a very exciting 2016 :-)

Frappé: Paper on arXiv & Context-Aware App Usage Dataset Release

During my last few months in Telefonica Research in 2013 I worked with wonderful colleagues and RecSys gurus Linas Baltrunas and Alexandros Karatzoglou along with scientific director Nuria Oliver on a context-aware mobile app recommendation service called Frappé. Frappé was specifically designed to support novel app discovery experiences. In order to assess it’s effectiveness we deployed Frappé in-the-wild on Google Play and ran a smaller-scale user study with 33 users designed to evaluate user perceptions of using and engaging with an app recommendation service.

Yes, it’s been a while since working on this specific project, however, I have 2 very exciting announcements to share about Frappé.

  • Firstly, a paper describing the Frappé application, our large-scale Google Play deployment and insights from our smaller scale user study has been published on arXiv. In particular we describe actionable lessons learned related to designing, deploying and evaluating mobile context-aware recommender systems in-the-wild with real users. Details and PDF are available here.
  • Secondly, we have released the anonymized Frappé data set!! It can be downloaded from Linas’s website HERE. The dataset contains 96,202 records by 957 users for 4,082 apps. We’re very excited to see what the RecSys and Mobile HCI communities end up doing with this rich dataset, in particular in terms of pushing the envelop in the context-aware recommender systems domain.

If you end up using the data, we ask that you please cite the following paper:

@Article{frappe15,
title={Frappe: Understanding the Usage and Perception of Mobile App Recommendations In-The-Wild},
author = {Linas Baltrunas, Karen Church, Alexandros Karatzoglou, Nuria Oliver},
date={2015},
urldate={2015-05-12},
eprinttype={arxiv},
eprint={arXiv:1505.03014}
}

Happy researching folks!!!

New Focus, New Role – Leading Native & Mobile Ad Analytics Group

As of this quarter in Yahoo Labs, I’m leading a small team of Research Scientists in a new group focused on Native & Mobile Ad Analytics. Native Advertising has become a hot topic in the Web and advertising spaces in recent years. Given it involves creating and placing engaging, relevant content that matches the underlying intent and context of the end-user, there are lots and lots of open research questions in this space. In particular when we think about native advertising in the mobile world!

The ultimate goal of this new group is to conduct research that will enrich our understanding of native and mobile ad experiences and will improve these experiences. We’ll be focusing on large-scale mobile user behavior modeling; understanding and mining mobile-specific signals like mobile device and app use to help enhance mobile ad engagement and conversion; as well as methods and novel studies exploring offline conversion in mobile, i.e. if I show a person an engaging, relevant ad on their device, do they ever act upon that ad in the physical world, e.g. visiting a store, eating at a restaurant, etc. To this end we’ll be analyzing very, very large datasets across a range of Yahoo properties as well as conducting large-scale in-the-wild mobile experiments / studies. Some of the open research questions / challenges in this domain include: Read More