Using Data to Understand Developers

Who? Where? Why? Using data to understand your audience.

In developer marketing and developer relations, the question we are often trying to answer is,
“Just what is the problem our target developers are trying to solve?”

Until we know this, how can we can demonstrate to our audience that the product, platform or service we offer is their best choice?

We need to know our target developers well in order to understand the problems they and their users experience. So there’s a range of questions we should be asking before we even get on to that key question (yes, it’s questions all the way down!).

The best way to answer questions is with data, of course. If you base your decisions on valid and trustworthy data, and the insights the data offers, you can naturally be more strategic. You can invest in the right tools, technologies and platforms and work towards answering the problem your developers want to solve, which will make you invaluable to them.

Transcript

1. THE ANALYST OF THE DEVELOPER ECONOMY USING DATA TO UNDERSTAND DEVELOPERS ANDREAS CONSTANTINOU CEO & FOUNDER andreas@slashdata.co @andreascon
2. WHO DEVELOPERS ARE WHERE THEY ARE GOING WHAT THEY BUY Developer population sizing Developer segmentation We help the world understand developers Why developers are adopting competitor products – and how you can fix that Emerging platforms – augmented & virtual reality, machine learning We survey 40,000+ developers annually – across web, mobile, IoT, cloud, Machine Learning, AR/VR, games and desktop – to help companies understand who developers are, what they buy and where they are going next.
3. KEY FIGURES 10 DEVELOPER AUDIENCES 40,000+ DEVELOPERS ANNUALLY 2 SURVEY WAVES ANNUALLY 18 GLOBAL WAVES TO DATE 150+ COUNTRIES REACHED 40+ RESEARCH STUDIES PER YEAR 70+ PARTNERS & CHANNELS FOR GLOBAL REACH 9 CUSTOMERS IN TOP-20 GLOBAL BRANDS SLASHDATA RESEARCH Mobile Desktop Industrial IoT Cloud Web Games Augmented & Virtual Reality Machine Learning & Data Science Consumer Electronics Apps/extensions for 3rd party ecosystems
4. 1. How big is the developer community?
5. DEVELOPER PROGRAM BENCHMARKING 1a. HOW IS THE DEVELOPER POPULATION GROWING? Source: Global Developer Population Report 2019 | www.slashdata.co | Copyright SlashData – All rights reserved We estimate the global developer population at 18.9M at the beginning of 2019 Active software developers in millions
6. DEVELOPER PROGRAM BENCHMARKING 1a. HOW IS THE DEVELOPER POPULATION GROWING? Source: Global Developer Population Report 2019 | www.slashdata.co | Copyright SlashData – All rights reserved We estimate the global developer population at 18.9M at the beginning of 2019 Active software developers in millions Growth of the global developer population 14.7 M 15.7 M 16.9 M 18.9 M Q2 2017 Q4 2017 Q2 2018 Q4 2018
7. DEVELOPER PROGRAM BENCHMARKING 1b. HOW IS THE DEVELOPER POPULATION GROWING? Source: Global Developer Population Report 2019 | www.slashdata.co | Copyright SlashData – All rights reserved We estimate the web apps developer population close to 17M Active software developers in millions 16.9 M 13.6 M 13.1 M 12.3 M 12.2 M 9.3 M 8.8 M 5.8 M Web apps Backend services Mobile apps Desktop apps ML, AI, and Data Science Internet of Things Games AR/VR Software sectors and their communities 2.70 AVG # OF SECTORS INVOLVED IN (Q2 2019) 2.65 AVG # OF SECTORS INVOLVED IN (Q4 2018 )
8. DEVELOPER PROGRAM BENCHMARKING 1c. WHICH LANGUAGES DO THEY USE? Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Javascript remains by far the most popular programming language Active software developers in millions Programming language communities 11.4 M 6.9 M 6.8 M 5.5 M 5.2 M 4.0 M 2.1 M 1.9 M 1.6 M 1.1 M 1.0 M .9 M .4 M .3 M Javascript* Java Python C# PHP C/C++ Visual development tools Kotlin Swift Go Ruby Objective C Rust Lua *Javascript including CoffeeScript and TypeScript 2.7 AVG # OF PROGRAMMING LANGUAGES USED (Q2 2019) 3.1 AVG # OF PROGRAMMING LANGUAGES USED (Q4 2018)
9. DEVELOPER PROGRAM BENCHMARKING 1d. LANGUAGE USE IS NOT STATIC Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Developers drop and adopt new languages over time based on existing needs 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Q2 2017 Q4 2017 Q2 2018 Q4 2018 Q2 2019 Javascript* Java Python C# PHP C/C++ Visual Kotlin Swift Go Ruby Objective C Rust Lua *Javascript including CoffeeScript and TypeScript Rank of programming language communities, 2017 - 2019
10. 2. How do you segment the developer audience?
11. DEVELOPER PROGRAM BENCHMARKING 2a. HOW DO YOU SEGMENT THE DEVELOPER AUDIENCE? 11% 44% 40% 38% 31% 27% 27% 27% 22% 22% 18% 18% 16% 16% 13% We don't differentiate or segment developers Target industry, vertical, or app category Professionals versus hobbyists or students Target audience (consumer, enterprise, SME, internal) Type of development (e.g. web, mobile, games) Geography (region or country) Age, experience, and/or level of education Developer motivations and goals Programming language Devices, operating systems, or platforms targeted (e.g. writing iOS apps) Usage of adjacent technologies or platforms (e.g. using AWS) Business model of the developer Company size (e.g. employees, revenue) Role in the organisation, level of decision power Impact to our business (e.g. money spent wih us) Segmenting Developers Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Target vertical is the most popular developer segmentation variable % of developer program leaders (n=45)
12. DEVELOPER PROGRAM BENCHMARKING 2b. DRAWBACKS OF POPULAR SEGMENTATION PRACTICES Mindshare of Symbian OS and Blackberry among mobile developers between 2012 and 2016 Source: How to segment developers into personas Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Technologies have the bad habit of becoming obsolete % of mobile developers (1Q12 n=1,440, 1Q13 n=5,380, 3Q14 n=10,392, 1Q15 n=7,149, 3Q15 n=7,462, 1Q16 n=11,523) 0% 5% 10% 15% 20% 1Q12 1Q13 3Q13 1Q14 3Q14 1Q15 3Q15 1Q16 Blackberry OS (5/6/7) Blackberry 10 Symbian
13. DEVELOPER PROGRAM BENCHMARKING 2c. DRAWBACKS OF POPULAR SEGMENTATION PRACTICES Developer focus on a single area of involvement Source: How to segment developers into personas Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved The vast majority of developers are active in more than one development area % developers in each sector that are involved in that sector only (n=20,578) 8% 6% 5% 4% 4% 3% 3% 3% 3% 2% Web Games Mobile IoT Desktop Data science Cloud Virtual reality Augmented reality Machine learning
14. DEVELOPER PROGRAM BENCHMARKING 2d. OUR DATA-BASED APPROACH Source: How to segment developers into personas Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved 1. We let the data speak – unsupervised 2. Categories of metrics we use as inputs, among others: Demographics, involvement per area, industry, roles, decision-making status, coding education and experience, cross-sector activity metrics 3. Categories of metrics we exclude: Revenues, technology choices 4. We run clustering analysis 5. We translate clusters into personas
15. DEVELOPER PROGRAM BENCHMARKING 2e. OUR DATA-BASED APPROACH Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Let the data speak – unsupervised. From clusters to our 6 personas YOUNG LEARNERS YOUNG PROFESSIONALS MIDDLE STANDARDS EMERGING EXTENDERS SEASONED DECISION MAKERS INEXPERIENCED LONERS
16. 3. Where are developers going next? A data-driven approach
17. DEVELOPER PROGRAM BENCHMARKING 3a. MAPPING DEVELOPER INVOLVEMENT Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Developers’ involvement across ten development areas % of developers in each sector (n=17,295) Sector % Developers % Professionals % Hobbyists/Students Web apps / SaaS Backend Services Mobile apps Desktop apps Machine Learning / AI / Data science Games Apps/extensions for 3rd party ecosystems Industrial IoT AR/VR Consumer electronics devices 64% 42% 40% 34% 27% 14% 12% 10% 8% 7% 70% 76% 49% 52% 38% 17% 62% 53% 26% 37% 30% 24% 51% 48% 62% 83% 38% 47% 74% 63%
18. DEVELOPER PROGRAM BENCHMARKING 3b. WHAT ARE THE MOST POPULAR CLOUD DEV AREAS? Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Containers is the leading technology in backend development % of backend developers (n=4,096) 46% 44% 38% 28% 28% 25% 24% 23% 22% 21% 19% 19% 9% Containers Database as a Service Cloud Platform as a Service SaaS systems Server app performance management & monitoring Cloud functions or serverless architecture Cloud monitoring services Private cloud software platform Container orchestration tools & management platforms Virtual machines IaaS Legacy enterprise systems Data processing services None of the above Backend technologies used in the past 12 months
19. DEVELOPER PROGRAM BENCHMARKING 3c. HOW FAR AHEAD IS KUBERNETES? Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Kubernetes mindshare is almost twice as high as the first runner-up % of backend developers who use orchestration tools (n=899) Container orchestration tools & management platforms platforms usage and awareness 60% 19% 5% 6% 19% 10% Kubernetes AWS ECS/EKS Docker Swarm Google Container Engine Microsoft Azure Container Service Red Hat OpenShift Microsoft Azure Service Fabric Pivotal Container Service HashiCorp Nomad Mesosphere DC/OS We develop this in-house / self-hosted Currently using Stopped using Evaluated but rejected Aware but never tried Unaware 1.7M KUBERNETES DEVELOPERS
20. 4. Where to find your audience?
21. DEVELOPER PROGRAM BENCHMARKING 4a. GLOBAL REACH Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Geo distribution of the Developer Economics 17th survey respondents % of developers (n=17,298) 22% NORTH AMERICA CANADA USA MEXICO BRAZIL 6% SOUTH AMERICA 11% E. EUROPE & RUSSIA FINLAND 21% W.EUROPE & ISRAEL UK GERMANY FRANCE ITALY SPAIN SOUTH AFRICA 9% AFRICA& MIDDLE EAST 11% SOUTH ASIA 18% EAST ASIA RUSSIA PAKISTAN INDIA 2% OCEANIA VIETNAM JAPAN CHINA AUSTRALIA
22. DEVELOPER PROGRAM BENCHMARKING 4b. REGION IS A CLEAR DIFFERENTIATOR Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved However, half of all developer programs have no regional differentiation % of developer program leaders (n=48) 50% 8% 10% 10% 19% Our program is global, without geographic differentiation We focus on our home market, global reach is a bonus We localise content, but don't geographically differentiate our services or outreach Our team is centralised, but we differentiate our services by region We have regional teams in place Regional strategy of developer programs
23. 5. How to support developers? A data-driven approach
24. DEVELOPER PROGRAM BENCHMARKING 3d. DISCOVERING WHAT MATTERS TO DEVELOPERS Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Documentation, tutorials, tools, and community engagement are the core of developer programs % of developers (n=16,087) Most important features that companies should offer to support developers Core Systematically in the top 4, of key importance to professionals and experienced developers Extended core Usually found just below core, and one level above all other attributes 61% 47% 45% 33% Documentation & sample code Development tools, integrations & libraries Tutorials & how-to videos Answers in public forums (e.g. Stack Overflow) Training courses & hands-on labs Technical support Official forums
25. DEVELOPER PROGRAM BENCHMARKING 3e. BENCHMARKING DEVELOPER PROGRAMS (2/3) Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Google, Mozilla, and Microsoft lead the pack % of developers (n=13,975) Google Mozilla Unity Microsoft Unreal / Epic Games Digital Ocean Red Hat Qt Cisco Amazon Salesforce / Heroku Facebook incl. React, Oculus Intel NVIDIA IBM ARM AMD Oracle Qualcomm Samsung SAP Alibaba Tencent Twitter / MoPub 15% 25% 35% 45% 55% 65% -20 -15 -10 -5 0 5 10 15 20 25 Engagement (% of program users that use resources at least weekly) Developer Satisfaction (weighted score of program features, -100 to 100) Bubble size: Developer adoption (% of developers using each program, log)
26. DEVELOPER PROGRAM BENCHMARKING 3e. BENCHMARKING DEVELOPER PROGRAMS (3/3) Source: Developer Economics Q2 2019 | www.slashdata.co | Copyright SlashData – All rights reserved Tracking satisfaction for certain attributes of developer programs by company % of developers (n=13,975) Satisfaction score based on a 5* rating system. Scores are calculated as the % of 5* responses minus the % of 1* and 2* responses times 100. Development tools & libraries Professional certification Documentation & sample code Information via newsletters, blogs, and social media Conferences & trade shows Microsoft Google Amazon Intel
27. THE ANALYST OF THE DEVELOPER ECONOMY WANT ANY OF THESE DATA POINTS FOR YOUR PRESENTATIONS? EMAIL ME: andreas@slashdata.co