Developer Relations is full of imprecise notions like increasing goodwill in the community and changing developer perceptions. But it also awash with data like unique visitors, twitter followers, monthly active users and GitHub stars. How is someone in Developer Relations supposed to know what metrics are important to their product or business? This session will take a look at some of the more popular metrics and provide practical advice about how to compute and use them.
13. 2009 2010 2011 2012 2013 2014 20162015
“Active customer” is defined as a non-Amazon customer with AWS account usage activity in the past month, including the free tier
Over 2 Million Active AWS Customers
21. Spring Community Population (Open Source)
User base: 2,500,000
Key Ratios
Core:Expert 20:1
Forum:Registered 10:1
Registered:Global 20:1
Global Population
Registered
Forum
Experts
Core
Active: 12,000
Committers: 20
Population
20x
30x
20x
Subscribers: 125,000
Answers: 400
10x
User base: 2,500,000
Started in technology at BEA Systems doing some evangelism and eventually running their Dev2Dev program.
Joined SpringSource to help Adrian Colyer and Rod Johnson build a powerful open source community of Java developers
Acquired by VMware and worked with the Groovy/Grails and RabbitMQ communities
Built the Developer Relations team for Cloud Foundry while at VMware and Pivotal
Joined AWS in 2013 to run technical evangelism and community building
Peter Ferdinand Drucker (/ˈdrʌkər/; German: [ˈdʀʊkɐ]; November 19, 1909 – November 11, 2005) was an Austrian-born American management consultant, educator, and author, whose writings contributed to the philosophical and practical foundations of the modern business corporation. He was also a leader in the development of management education, he invented the concept known as management by objectives and self-control,[1]and he has been described as "the founder of modern management".[2]
William Thomson, 1st Baron Kelvin, OM, GCVO, PC, FRS, FRSE (/ˈkɛlvɪn/; 26 June 1824 – 17 December 1907) was a Scots-Irish[1][2]mathematical physicist and engineer who was born in Belfast in 1824. At the University of Glasgow he did important work in the mathematical analysis of electricity and formulation of the first and second laws of thermodynamics, and did much to unify the emerging discipline of physics in its modern form. He worked closely with mathematics professor Hugh Blackburn in his work. He also had a career as an electric telegraph engineer and inventor, which propelled him into the public eye and ensured his wealth, fame and honour. For his work on the transatlantic telegraph project he was knighted in 1866 by Queen Victoria, becoming Sir William Thomson. He had extensive maritime interests and was most noted for his work on the mariner's compass, which had previously been limited in reliability.
Phil @ischi (Philipp Fehre) Follow
More
Replying to @DevRelChap @matthewrevell @mewzherder
Still smiling at DAU everytime ... In german IT it is commonly used as an abbreviated stupidest possible user (dümmster anzunehmender user)
“Active customer” is defined as a non-Amazon customer with AWS account usage activity in the past month, including the free tier
Silicon Valley Season 4, Episode 1: Daily Active Users
Richard Hendricks: “Piper Chat has 120 thousand daily active users, growing organically at 18% week over week”
VC: “Hit to a million daily actives users while sustaining the same kind of growth, then everyone in town will be trying to kidnap you.”
DAU: https://engineering.helpshift.com/2014/active-users/
Active user is the cardinality of the set of user ids where an id is mentioned in the log of system events
DAU: https://engineering.helpshift.com/2014/active-users/
Linear Counting
LogLog Counting
HyperLog Counting
https://amplitude.com/blog/2016/01/14/measuring-active-users/
Externalities: Other things (press, events, external blogs)
Masking Churn: Increasing adoption/trail then abandonment is not visible
Wrong Event Source: Are the events you are collecting ids from the right ones
Impressions or reach?
At AWS Technical Evangelism, we consider 1 hour of in-person interaction as the standard unit.
Engagement Metric = sum of attendance at all engagement activities (conference talk, webinar, Twitch broadcast) multiplied by a weight for each activity
Sign up? Giving you a payment instrument? First bill? First time exceeding the free tier?
Channel
Definition
Paid Search (SEM) - Customers coming from paid ads on the search engine results page. Paid search keywords are classified as “Brand” or “Non-Brand”.
Natural Search (SEO) - Customers coming from search engine through organic search results.
Social Media -Marketing campaign ad on social sites or referrals from social site pages (both adobe and AWS identified social sites). May be designated “Paid” when the click is the result of advertising on social media.
Banners ads - Banner Advertisements created as part of a campaign which are running on external sites
External links - Customers referred from external link campaign managed by us.
Email - Customers referred by emails sent by us
Internal - Customers referred from our internal sites
Referring Domains - Customers referred by external domains but not through any of our campaigns, including links in news articles and blog posts.
Direct - Customer either typed in web address directly or came through a bookmark for our site
Text Ads - External ad placements consisting of text only. These are uncommon.
Unknown - Referring channel for the click could not be established.
𝑝 is probability of conversion
𝛽 𝑖𝑗 probability when the 𝑖 th touch is through channel 𝑗
𝑓 𝑖𝑗 is an indicator for whether channel 𝑗 is the 𝑖 th touch
We estimate the impact of different marketing channels using a logistic regression model for the probability of signup:
log 𝑝 1−𝑝 = 𝑖=1 𝑘 𝑗=1 𝑐 𝛽 𝑖𝑗 × 𝑓 𝑖𝑗 ,
In the radio series and the first novel, a group of hyper-intelligent pan-dimensional beings demand to learn the Answer to the Ultimate Question of Life, The Universe, and Everything from the supercomputer, Deep Thought, specially built for this purpose. It takes Deep Thought 7½ million years to compute and check the answer, which turns out to be 42. Deep Thought points out that the answer seems meaningless because the beings who instructed it never actually knew what the Question was.[1]