Whatever you're building, we're here to help you take the first step with confidence.
From Zero to #1: How a Data-Driven GTM Framework Turned an Untapped Category into a Platform's Biggest Revenue Driver
Market: Vietnam · Timeline: 2016–2017 Result: A new property vertical overtook the platform's top-earning category within 12 months — and became its first profitable vertical.
The situation
A high-growth Vietnamese marketplace had a revenue concentration problem. Vehicles dominated earnings because sellers paid for visibility on high-value items. Everything else — including categories with strong organic traffic — wasn't monetising.
Then the data surfaced something unexpected: a real estate category no one had invested in was quietly becoming the platform's second revenue stream. Users were listing properties without any product investment or marketing. Meanwhile, the dominant property portal in Vietnam was proving the market was large, growing, and winnable.
The question wasn't whether to build a dedicated property vertical. It was whether the data could make a credible enough case to justify the bet — and what it would take to win.
What I was asked to do
I was the platform's first data hire. The BI stack was built, trusted, and used. What didn't exist was a framework for evaluating a new market entry. Management had instinct and organic signal. They needed evidence: market size, competitor economics, go-to-market plan, and pricing — before committing to a build.
I proposed a four-week structured analysis, presenting one piece each week. The CEO agreed.
How we built the answer
Week 1 — Market sizing. I built a five-year view of Vietnam's online property advertising market: transaction volumes, urbanisation trends, digital shift. The market was large, growing, and concentrated in one player. That concentration was the opportunity.
Week 2 — Competitor revenue estimation. We needed to know what the leading portal was actually earning. They weren't going to tell us. So we triangulated: SimilarWeb and App Annie for traffic data, manual listing counts across key search terms, and mystery shopping for pricing tiers. Each source was imprecise. Together they produced a number we were prepared to defend.
When we placed that estimate next to the platform's current organic property revenue, the gap was larger than anyone had expected. It reframed the conversation — from "should we do this?" to "how fast can we go?"
Week 3 — Go-to-market plan. Rather than the competitor's model of serving individual agents at scale, we chose a focused in-house sales team targeting established agencies. Lower initial revenue, but manageable at our stage and lower churn risk. On the buyer side, we invested in parallel: marketing to signal inventory depth, and cross-category conversion to bring existing platform users into property.
Week 4 — Pricing. We chose a listing fee plus advertising model over subscription — deliberately. Subscription demands upfront commitment before sellers have seen results. Listing fees let sellers pay for what they use, with advertising as an upsell once leads start coming in. We set an ambitious first-year north star KPI and agreed: if we missed it, we'd reassess.
What reality added
Two complications emerged after launch.
Cash payments were still the norm in Vietnam in 2014. We made the deliberate call to exclude cash at launch, knowing it would constrain early revenue. It was right — it let us focus on building a working product instead of a field collections operation.
Paid features were also delayed by up to six weeks due to engineering scope. Revenue shifted forward. But because we'd built conservatively into the KPI framework and user growth was tracking ahead, we still hit the first-year target.
The biggest funnel insight came later: second-listing drop-off. Sellers who listed once for free frequently resisted paying when the listing expired. The fix was making ROI visible — showing agents exactly how many leads their listings had generated. When sellers saw real demand, they came back and paid. The data made that loop manageable.
The outcome
Within 12 months, the property vertical had overtaken vehicles as the platform's largest revenue category. Within two years, it was the single biggest revenue driver — and the first profitable vertical.
The structural results mattered as much as the revenue: an ambitious KPI hit despite a delayed feature rollout, a sales model that scaled without mass-market complexity, and a pricing structure that produced low churn and predictable growth.
This project also changed how the company used data. Before, analytics reported on what had happened. After, it shaped what to do next. The four-week GTM framework — market sizing, competitor benchmarking, pricing analysis, KPI-setting — became a template for how subsequent strategic bets were evaluated.
The lesson
The competitor revenue estimate was imprecise. But it was directionally correct — and we knew it was, because three independent methods had converged on the same place. Good market intelligence doesn't require perfect data. It requires enough rigour to be defensible and enough honesty to name what you don't know.
The deeper point: a data team that owns the business outcome is a different thing from one that produces reports. Everyone on this project knew that if the vertical failed, the analysis had failed. That accountability changed the questions we asked — not "what does the data say?" but "what does the business need to know, and how do we find out?"
This case study reflects real work led by Hanh Nguyen, founder of Syllog Data, and has been anonymised to protect company confidentiality.
If you're evaluating a new market, a new vertical, or a major strategic bet — and you want a data framework that gives you a defensible answer before you commit — that's exactly what Syllog does.
From Data Chaos to Self-Serve in 6 Months
Client: A leading pre-IPO marketplace operating across three Southeast Asian markets, with thousands of employees and a dominant position in real estate and consumer goods.
Engagement type: Fractional Head of Data & Analytics + Data & Analytics Audit
The situation
When Syllog's engagement began, the company had a 10-person analytics team — and almost no one in the business was getting the data they needed.
The team was entirely focused on building a data product that had low internal adoption. Meanwhile, product and marketing leaders were queueing up outside the CTO's office every week, frustrated that their OKRs were slipping because no one could give them the numbers to make decisions.
The gaps were everywhere:
Multiple tools had been purchased and were either unused or misused
Each team had its own preferred tool, so data lived in silos with no single source of truth
Finance and BI were manually exporting data from systems and emailing spreadsheets
Directors were assigning one person per department each week just to compile reports for management
The same metric — say, renewal rate — showed three different numbers depending on who you asked
The company had been operating for 10 years and had invested in powerful technology. But basic questions about customer behavior on their own platform had never been systematically measured.
What we did
Step 1: Audit first, buy nothing new.
Rather than recommending new tools, we started with a thorough audit of what existed — every tool, every data source, every process. The finding was counterintuitive: the company didn't have the wrong tools. They had the right tools in the wrong hands, used for the wrong purpose. The path forward was optimisation, not replacement.
Step 2: Align on what actually mattered.
We ran structured alignment sessions with the management team, strategy team, and business leads to define L1, L2, and L3 metrics — from company-level north stars down to team-level operational metrics. Getting agreement on definitions was hard. Nobody wants to change their KPI definition mid-year, and nobody wants to admit their numbers might have been wrong. We solved this by running the old and new metrics in parallel until leadership could see for themselves why one version was enough.
Step 3: Build the self-serve layer.
With aligned metrics and a redesigned ETL flow (in partnership with the data engineering team), we built out dashboards in Looker and Looker Studio covering L1 through L3 metrics. We then ran training across the organisation — and critically, built a culture of appropriate self-service: teaching teams not just how to use the tools, but when to use them versus when to bring in an analyst.
For Sales and Finance — teams deeply reliant on Excel and needing to make manual adjustments to data — we designed a hybrid approach: automated pipelines feeding into Excel, with a path to migrate into the BI system over time. Half-baked by design, because forcing a perfect solution would have killed adoption.
Step 4: Free the analysts to do real work.
As self-serve adoption grew, the analytics team shifted from fulfilling repetitive data requests to doing work that actually moved the business — deeper analysis, better data products, stakeholder advisory. The data product the team had been building — which previously had stalled without business buy-in — finally got onto the product roadmap, improved iteratively based on real stakeholder input, and launched across all markets.
The results
In 6 months:
70% of business data needs are now met through self-service (measured by ticket volume)
Manual reporting work reduced by 70% across the organisation — Finance, BI, and operational teams no longer spend weekly cycles on data compilation
Analyst-to-stakeholder ratio improved from 1:2 to 1:3–4, without any headcount reduction — even through two company-wide restructuring rounds
Management team moved from working only with finance numbers to having a holistic, real-time view of the business via automated daily dashboards
Product team could ship features on schedule, unblocked by data delays
The CTO's door was no longer a complaint queue
This is the kind of work Syllog does under its Fractional Head of Data engagement — embedded leadership that combines audit, strategy, execution, and change management. If your data team is busy but your business is still running blind, let’s talk.