这是indexloc提供的服务,不要输入任何密码
Populix logo

Populix: Harnessing Google Cloud to transform market research in Indonesia

Google Cloud Results
  • Delivery time for end-to-end research — including survey scripting, fielding, analysis, and reporting — sped up by over 50%

  • Automation reduces QA time by 40%, freeing up the team to focus more on interpretation and strategic consulting

  • Faster research turnaround, streamlining of the initial research stages enables quicker studies, faster insights, and faster responses

  • Automated quality control using Gemini helps identify and eliminate fraudulent survey responses for more reliable and accurate data

  • Automation of tedious tasks, such as verbatim coding and insight generation, allows researchers to focus on more strategic work

Populix transformed its market research operations by migrating to Google Cloud, leveraging its native support for Gemini and cost-effective infrastructure.

The journey to AI-driven insights

We realized that our industry is changing, and we need to change with it because insights need to be delivered faster and deeper. That's why we chose Google Cloud for its native AI support, cost-effectiveness, and reliability.

Jonathan Benhi

Co-founder and Chief Technology Officer, Populix

Before implementing Google Cloud, Populix faced significant challenges in scaling its research infrastructure while maintaining speed and data quality. With credibility being imperative in research, Populix wanted to balance speed and scale with trust and accuracy – while generative AI offers immense efficiency gains, the company wanted to ensure that the insights it helps generate are still grounded in real data and not just plausible-sounding narratives. 

In addition, the company's existing cloud provider was becoming less and less cost-effective, particularly for its servers, leading Populix to seek a better and reliable solution. The company chose Google Cloud principally because of the platform's native support for the Gemini model, which Populix had chosen as its foundational AI model for developing generative AI software.

This seamless integration was not only smooth and stress-free, it was also complemented by the platform's competitive pricing model for the company's specific use cases.

Populix began its gradual migration to Google Cloud in 2019, initially utilizing BigQuery, and significantly expanded its machine learning operations on Google Cloud by 2022. Internally, Populix leverages BigQuery coupled with Looker for its business intelligence. The native integration between these tools allows Populix's BI team to easily dump data into BigQuery and then query it to build dashboards and visualizations. 

This has dramatically saved time and effort, enabling more data-driven decision-making within the company. An example of how this looks in real life is a client using the AI research assistant to quickly draft a questionnaire for a new product, receiving AI-generated insights and summaries from the collected responses within hours, rather than weeks of manual analysis.

Two individuals working on notepads and a laptop

Turbocharging research with generative AI

Google Cloud has significantly improved Populix's operations in two key areas: internal business intelligence and customer-facing Software as a Service (SaaS) features. For its SaaS offerings, Populix has developed an AI research assistant chatbot that helps users create and launch surveys. This assistant, powered by Gemini, drafts survey questions and analyzes respondent responses, providing summarizations that previously required manual effort. This has helped make the survey creation process more user-friendly and deliver deeper insights.

Furthermore, AI has also transformed how Populix's researchers work, from designing studies to analyzing responses and drafting reports. While getting insights faster is exciting, they need to be evaluated with stronger critical thinking to ensure that they remain accurate and actionable.

The time to insight is much faster for us. What previously took weeks now can be done in days, if not hours, because of the AI integration powered by Google Cloud.

Jonathan Benhi

Co-founder and Chief Technology Officer, Populix

For Populix, another significant advantage of AI is its central role in helping researchers shift their mindset from executing every task manually to orchestrating and validating AI-generated outputs. It elevates their role – instead of spending time on repetitive tasks, they can now focus on framing smarter questions, evaluating AI-generated insights, and drawing meaningful conclusions. 

To further enhance this, the company is prioritizing upskilling of researchers in three key areas:

  • Prompt engineering – enabling them to ask better questions and guide the AI effectively throughout the research workflow.
  • AI literacy – giving them a better understanding of the capabilities and boundaries of generative models, especially in terms of bias and hallucination.
  • Human-in-the-loop thinking – helping them become strategic reviewers and insight curators, rather than just data processors.

Ultimately, the goal is to empower researchers to do higher-value work, faster, and with greater impact through AI.

Another example of how Populix is using Google Cloud to enhance research is the NeXa AI Research Assistant. Built using the platform's Vertex AI and Natural Language API, it leverages Gemini to help users articulate research problems, automatically create surveys, and assist in choosing targeted respondents and launching surveys. Populix is now transitioning much of its software into more AI generative software, with Gemini being the long-term foundational model powering all its AI features.

Google Cloud architecture

Shaping the future of research

Building on its implementation of Google Cloud, Populix is now developing a synthetic panel solution to mimic real-world respondent profiles and deliver accurate insights quickly. Designed to be integrated into the company's research automation platform, the synthetic respondents project uses several years' worth of historical survey response data to train a foundational model like Gemini to answer questions. It is envisioned as a complementary solution to real respondent answers, especially for niche audiences that are difficult to obtain through traditional research methods.

Populix is a leading consumer insight platform in Indonesia, connecting businesses with real-time, high-quality insights through its extensive panel of 1M respondents and cutting-edge AI-driven research solutions.

Industry: Technology

Location: Indonesia 

Products: Google Cloud, BigQuery, Gemini, Google Kubernetes Engine (GKE), Looker, Vertex AI

Google Cloud