Guerrilla Analytics

Data analytics projects are extremely dynamic, fast paced and, if not managed correctly, quite chaotic. You have a team of mixed skill sets and your clients and end users rarely understand what can be achieved and its complexities. Analytics requirements change often and tooling is often restricted or limited. Data arrives piece meal with replacements, supersets, additions and flaws. Your team is producing multiple deliverables per day, any of which may become project critical and have to be reproduced or audited. Guerrilla Analytics addresses the need for techniques to manage agile data analytics projects based on his experience in research and in consulting. Enda Ridge, PhD introduces simple principles that cover all the basics of managing analytics projects. You will learn simple processes and workflows for setting up the data analytics environment, data provenance and deliverables tracking, data manipulation patterns and designing analytics for test and reproducibility. This book lets managers and data analysts get an analytics project off the ground fast, responding to dynamic project requirements without compromising coordination, reproducibility, traceability or quality. You will learn: * Reproducible, Robust analytics: how to design analytics work products so that they are reproducible, testable and stand up to audit and external scrutiny* Setting up an analytics project: how to set-up your analytics environment in simple terms of tooling, structure, skill sets and conventions* Tracking work: how to track data and work products easily with minimum overhead* Analytics patterns: common 'patterns' of analytics development from project experience, the equivalent of which are already well established in software engineering