Weekly Data News
if your data don't look weird you're not looking hard enough
- Progressively ramping up : improved communication, KPI checklist in Confluence log book, escalation process to IT very much improved.
- Next step : better, clearer daily output message to management team.
- Anyone interested can attend the daily discussion in the Analytics Crew room on HipChat and read the Night Watch Log Book.
- Lots of small tasks
- Analysis of B2B team challenges on Social Ads
- Matching Hub & company pages
- AXA reports
- Initiated discussion about improving our reporting process for sales' customers
- Initial data for trainings search funnel, more to come (from Architects)
- Product line review prep
- Skills dictionary: manual qualification of skills reference labels
- Email target extract after bug on profile consult
- Yellow push bar: Google Analytics tracking. Defined priorities of user pushes.
- Tuning of New Dashboards
- Transactions & revenue: record increase after promo email and price change. Seems so huge that we need to be cautious : figures will be checked further.
- Analysis on premium products
- Campaign reports to follow last emails
- Planned for next week: Emails campaign to Russia & India
Data Management and Machine Learning
Budget Re-ForecastA hell of a lot of data crunching to provide revenue models and activity analysis.
- Manual validation of reference skill labels (currently : 6000 done out of 9000)
- Consensus about the need for a permanent reference lists admin role aka The Documentalist (intern?)
- We need a power tool to manually tag thousands of words with semantic relationships (SKOS) and enrich our data with web data (dbpedia, freebase, etc). In order to assess the feasibility of such a tool we want to organize a UX sketching workshop asap.
KasperOngoing contribution to Kasper implementation.
It is about scalability : as a campaign manager, I want to be able to send the "Who saw my profile" part of Viadeo Inside email to the entire database of members. Obviously we actually do not send the email to the members but rather to a local server built specifically for testing purpose.
A couple of things that have and will keep us busy
- Bandwidth between Amazon VPC and SF datacenter can not be shared nicely by data team and the rest of Viadeo. But solutions are currently being discussed with Ops team both in Paris and SF.
- Access to our internal scalable distributed data stores needs fine tuning (Hue).
- Dependency to Magic Wonder Box algorithm industrialization (SF team). By end of August it should be exposed as a service for us to be able to work on Contact recommendations in future version of Viadeo Inside.
The general feeling of the team is positive! The technology stack we use is getting more and more familiar, it will soon be clicking for some real production action!
- Clémence Desneiges - Manager of Business Intelligence
- Clémentine Blanchon - Business Analyst B2B
- Julie Buire - Business Analyst B2C Growth
- Frédéric Chancholle - Business Analyst B2C Love
- Kristine Romero - Business Analyst Marketing and B2C Money
- Pierre-Emmanuel Servant - Business Analyst Search (San Francisco)
Data Management & Machine Learning
- Eugen Cepoi - Software Architect
- Arnaud de Myttenaere - PhD Candidate
- Julie Séguéla - Data Management Expert
- Julien Bille - Data Platform Engineer
- Jean-Luc Canela - Tech Lead
- Mathieu Chataigner - Data Platform Engineer
- Iñigo Mediavilla - Web Development Engineer