New Internet Users (NIU) Initiative

Written By Ben Crabtree

A 3-year strategic growth research initiative that uncovered the human needs of globally underrepresented populations who were/are not yet online.

Problem

In 2018, the topic of New Internet Users (NIUs) was a black box.

  • From 2018 to 2023, 1.2 billion NIUs were forecasted to come online, primarily on smartphones.

  • NIUs being a mystery was a problem that would hinder Google from reaching its growth objectives.

  • Why? These new users faced different constraints from existing users, globally.

  • By demographic comparison, they were identified as semi-literate and lower income, with minimal exposure to digital technology. They also greatly changed the gender and age ratios of the online population.

  • Considering these significant differences, existing product designs would not work well for NIUs.

Challenge

  • At the time, there was limited data on NIUs. Most existing market research was conducted online, which biases towards people who are well-versed in the internet and surveys.

  • NIUs were not yet identifiable in our data logs–behavioral signals were lacking.

  • As the internet’s earliest users, Googlers have/had a hard time relating to people who hadn’t used it before.

My Role

Project & Research Lead. While Google had already developed an intuition around building for the “next billion,” our team saw the need to look ahead and gain an understanding of who the next next billion would be. This is why I developed The New Internet Users Project.

A performance center we visited in-between research sessions in central Brazil.

Goals

  • Understand the variety of motivations, barriers, triggers, and effects of coming online

  • Balance out the demographic information of NIUs in “emerging markets” with behavioral insights

  • Elicit the first-year experience of coming online with mobile-only/-primary internet users in “emerging markets”

  • Identify how product paradigms for core pillars like literacy, identity, and privacy can evolve at the company

  • Identify opportunistic product areas, feature areas, and growth strategies

  • Create new ways of thinking about NIUs (i.e. design guidance); demonstrate industry leadership in these areas

Findings

This study uncovered and defined behavioral distinctions of NIUs, such as:

  • experiencing low digital confidence, paralyzing their usage (beliefs often rooted in having less formal education)

  • infrequently having access to teachers or good-quality help for learning and troubleshooting

  • seeing little value in how internet use connects to their life and aspirations

  • not having the digital skillset to access the internet’s benefits

While most of the insights remain confidential, some were released publicly. Following are five features:

Google I/O 2021

Building Globally Relevant Products. An introductory talk by myself and Product Strategist Neha Malhotra.

The Wall Street Journal, Oct 2020

Google Rethinks App Design for Internet Novices by Katie Deighton

IDEO x Gates Foundation x Google

Designing for Digital Confidence Toolkit

Impact

My hands-on work and team leadership drove 50+ NIU-focused strategic product changes that addressed digital confidence growth, literacy barriers, user education, and cross-product UI consistency.

Impact Examples. NOTE: Details have been scrubbed for confidentiality.

  • New Features: Language-focused

    • Increased daily active users by ##%+

    • Boosted engagement for specific language topics by ##%

    • Raised non-English user language choice uptake by ##%

    • Increased content language diversity led to +$## revenue

  • New Features & Products: Literacy-focused

    • Prominent voice entry points → +##% usage

    • Simplified, less text heavy onboarding → increased comprehension

    • Increasing icon descriptiveness → +## usage

    • More Audio-Described/Read-Out & Visual-forward experiences

  • New Designs & Projects: Digital Literacy & User Education

    • Newly produced how-to videos with real people, and how-to animations → increased user confidence

    • Error prevention designs in utility apps → improved user comprehension

    • Visual and voice-based guides to teach users how to make the most of their phone

  • New Quant Signals: Developed proxy signals to help identify NIUs in data logs

Beam storage, image from fieldwork by me.

Research Process

I believe in leading with a strong vision, and facilitating thoughtful collaboration to bring that vision to form. At each step, and when appropriate, I gathered input by involving relevant parties from my immediate team, the broader team, cross-functional stakeholders, and partner research agencies.

  • Goal: understand existing knowledge, sizing, discussion, and documentation of the population

  • Goal: solidify vernacular before introducing and circulating the term

    • Secondary research led to defining a “New Internet User” as a person who is experiencing the internet for the very first time / has very minimal experience / very recently came online as first-time smartphone owners. Compared to existing users, they will:

    • Reside more in peri-urban and rural areas

    • Balance the online gender ratio (in some countries: more women)

    • Expand the age range of internet users (not majority 18-35)

    • Have limited digital experience or exposure

    • Have less infrastructure access, meaning patchier connectivity

    • Exhibit lower literacy, having spent less time in formal education

    • Be more multilingual

    • Be in a lower socio-economic level

  • Goal: secure stakeholder investment in the project

    Create and circulate pitch deck amongst stakeholders, emphasizing:

    • Market sizing, opportunities, i.e.: From 2018-2023, 1.2 billion people will become “new internet users”

    • Why differences like lower literacy matter

    • Examples of NIU-related design changes with significant impact

    • Why the study is important to meet Google’s global objectives

  • Goal: create a master plan as a source of truth to ensure consistency across locations & local teams

    • Set Research Principles: I strongly believed in making this study as naturalistic as possible, so I revised an existing plan, originally drafted by another researcher, to move away from a typical intervention or usability study that is limiting in its outcomes and brings forth ethical concerns. As a foundational guidepost, I established non-negotiable criteria to ensure rigor and efficiency throughout the study.

    • Set Core Themes + Research Questions: Involving stakeholders as early as possible is a tactic I always employ. As the project was just getting started, I ran an optional activity across the Next Billion Users organization. During lunch and our weekly “Chai Wednesday” share-out, we brainstormed questions about new internet users. This activity drew over 40 cross-functional teammates and generated more than 100 questions. After reviewing these, I clarified the core themes to focus on.

    • Select Methodology: I designed a longitudinal study structure that took a mixed methods approach, involving in-depth interviews, shadowing, diaries, surveys, and more. Of the 175 NIUs interviewed, 73 became involved in a year-long study documenting their smartphone / internet experiences. After they chose the smartphone they intended to purchase, we offered to pay for it as part of the study, which included in-person visits every few months, as well as phone / messaging check-ins. Visits would involve various activities (journey mapping, usability tests, concept evaluations, etc.).

    • Define Sampling Criteria. (confidential)

    • Determine Recruitment Method. Participants were primarily recruited by:

    • Partner agencies local to our target communities, who recruited door-to-door. Participants self-reported that they intend to use the internet soon, or have come online very recently. I signed off on each participant's profile based on their responses to a screener questionnaire and reverified this information during our first interaction with each participant.

    • Snowballing: through people we met, we learned about family, friends, and work buddies who would also soon obtain their first internet-enabled phone

    • NGOs who work with resource-constrained and/or isolated communities

  • Goal: test out and improve methodology before expanding the study


    • Why pilot in India? I had experience leading 6+ studies in various regions of the country through my work on YouTube Go, in partnership with the agency: D’Well. This familiarity allowed me to concentrate on optimizing the study design, rather than the logistical challenges that inevitably arise when entering a new environment. Additionally, selecting the right team is crucial in fieldwork, and I had previously identified strong, proactive members who were eager to collaborate again.


    • Sample Schedule. To make travel expenses worthwhile to the company, I often design research trips to be two to three weeks in length. I divide us into sub-teams of two or three members to maximize how many sessions can be completed in a day, without bloating the team and overwhelming participants. Each sub-team member has a role: moderator, translator, notetaker/video documentation.

    • (CALENDAR IMAGE?)

  • Goal: demonstrate and drive early impact of the project, gain buy-in for study expansion

    Diligent documentation, material organization, and consistent debriefing in-field allowed me to send a writeup of emerging trends during the trip. This excited stakeholders to see more data. Upon returning to HQ in San Francisco, CA, my partner design strategist and I ran a “synthesis session” where anyone from the team was invited to join to read the transcripts and debrief of a participant, and share what they found interesting during discussion.


    I achieved stakeholder buy-in for study expansion with:


    • Timely insights and quick turnarounds at different fidelities, building anticipation for the final reports or artifacts

    • Proven influence on product roadmaps after running workshops with the Assistant and Identity teams

    • The popularity of the first official report, circulated around the company which was entitled: “Debunking 10 Assumptions Google Makes about New Internet Users”

    • Transparency into process and research rigor (i.e. the involvement of the team in Synthesis Sessions / Discussions, pictured here)

  • Goal: Increase the diversity and robustness of the sample and data, cover top priority markets

    • Select Focus Locations with Local Agencies. Countries were chosen based on company priority. To assess how city density affected experience, we sampled from across urban, peri-urban, and rural locations across five countries, with a skew towards peri-urban and rural to mirror NIU characteristics cited in market research.

    • Brazil: Pernambuco, Minas Gerais [Partner Agency: Mercedes Sanchez]

    • India: Coimbatore, Assam [Partner Agency: D’Well]

    • Indonesia: Banjarmasin, Jogja [Partner Agency: D’Well]

    • Mexico: Chiapas, Atlixco [Partner Agency: INSITUM, now Fjord]

    • Nigeria: Ibadan [Partner Agency: Kantar-Millward Brown] (One location only. Due to safety concerns, we were not allowed to go to a 2nd location. As a result, we increased the sample size in Ibadan.)

    • Developed Study Materials for Over-Time Visits. Here is a visual journey map we created with participants after speaking to them for a year. They recalled events that stood out to them with their first smartphone and internet usage. After filling out the calendar, they reflected on how life had or hadn’t changed with access. All over-time methods can be seen in the Appendix here.

  • Goal: Equip Google to make deeply human-centered decisions for the next next billion coming online

    During and after study expansion to different countries, I drove change by reaching decision makers across 15 product areas through crisp, compelling, and wide-spread insight artifacts, which took the form of:

    Reports
    (decks, anthologies, write-ups, white papers)

    Micro-documentaries
    (narrative videos, demonstrative videos, highlight reels)

    Interactive activities

    (Gorm the Zop, “Are you Worthy of the Internet? Questionnaire,” Intuition Building Quizzes)

    Workshops & Presentations

    (Google I/O, UX University, Leadership Summits, Product Team Workshops / Sprints)

An example research trip schedule at a high level.

Example of an emotion journey map activity done with participants.

Thank you for reviewing this case study. Some details have been redacted to maintain confidentiality.

If you have any questions, please feel free to reach out at tracey@trace-grooves.com

Ben Crabtree
Previous

Taara