Data-Driven Retail in Thailand: How to Measure Promotional ROI and Sell Without Relying on Discounts

Target Audience: Business owners, site managers, store operations managers, and administrative department managers of Japanese-affiliated retail and distribution companies with operations in Thailand, as well as business planning and IT promotion staff at Japan headquarters considering expansion into Thailand.

“We have to cut prices again or nothing will sell” — this sentiment is increasingly common on the front lines of Japanese-affiliated retail and distribution businesses in Thailand. As of 2026, consumer spending in Thailand has grown more cautious, and international organizations including the World Bank have noted a slowdown in Thailand’s economic growth. Competitors are caught in similar discount wars, and many stores have fallen into a vicious cycle: “attract customers with discounts → gross margin erodes → no budget for the next investment.”

However, companies that are protecting their sales and gross margins without relying on discounts share one common trait: they connect data to on-site decision-making. Rather than keeping POS (point-of-sale) data, inventory data, customer purchase history, and promotional effectiveness metrics locked in separate systems, they link these directly to daily ordering, product placement, and promotional decisions.

This article provides a concrete framework for how Japanese-affiliated retailers in Thailand should use data to achieve “selling without relying on discounts,” how to measure promotional ROI (return on investment), and how to connect on-site operations with management reporting. We hope you will read this not as a buzzword discussion about DX, but as practical guidance for changing the numbers on the ground.

1. The Current State of Thailand’s Retail Environment: What Is Happening in 2026

Thailand’s retail market has changed dramatically in recent years. Even as large malls and foreign-affiliated chains continue to open new locations, the spread of e-commerce has transformed consumer purchasing behavior, and foot traffic to physical stores has declined compared to previous years. In addition, Thailand’s economy in 2026 faces heightened external uncertainty, and the situation remains unpredictable for Japanese companies already operating or considering entering the market.

Against this backdrop, the major challenges facing Japanese-affiliated retailers can be broadly summarized as follows.

• Rising consumer price sensitivity: Economic uncertainty has tightened consumer wallets, making reliance on discount sales increasingly common.
• Rising labor costs: Thailand’s minimum wage has been trending upward in recent years, and personnel costs for store operations and logistics are squeezing profitability.
• Talent acquisition and retention: Turnover rates are high in retail and logistics, and when experienced staff leave, tacit knowledge leaves with them.
• Accountability to Japan headquarters: Even when sales are stagnant, Japan headquarters increasingly demands “cost reduction,” “gross margin improvement,” and “explicit investment recovery plans.”
• Price competition with competitors: Being drawn into price wars with local chains and major e-commerce platforms makes it difficult to find a clear axis of differentiation.

Under these conditions, “selling without discounts” is not merely a matter of attitude. It can be achieved by using data to improve operational precision — delivering “the right products, in the right place, at the right time, at the right price.”

2. The Structural Problem with Discount Dependency: Why Gross Margins Keep Getting Squeezed

Promotional strategies that rely on discounts can generate sales in the short term, but over the medium to long term they create multiple problems. First and most significantly, there is the structural compression of gross margins. Once consumers become accustomed to “buying on sale,” their willingness to pay full price diminishes, and over time the proportion of customers who “won’t buy unless there’s a discount” grows.

Next is the problem of inventory management disruption. A cycle of large-quantity orders timed to discount sales → unsold inventory → further markdown disposal tends to emerge easily, leading to a buildup of dead stock. In Thai warehouses, the hot and humid environment compounds this issue, as dead stock also carries the risk of quality deterioration.

Furthermore, unclear promotional effectiveness is a fundamental challenge. “How much did last week’s flyer distribution actually affect foot traffic?” “How much in actual purchases resulted from our SNS advertising spend?” — few Japanese-affiliated retail operations in Thailand can answer these questions. When promotional ROI cannot be measured, the decision to “just try a discount” gets repeated over and over.

To break free from discount dependency, the starting point is using data to understand the root cause of “why we are discounting in the first place.”

3. The Basic Architecture of Data Utilization: What to Connect with What

“Data utilization” is a term used broadly, but the first priority for retail operations is connecting data that already exists. In most stores, POS data, inventory data, ordering data, and accounting data exist in separate systems or ledgers, and are not in a state where cross-functional analysis is possible.

The basic structure for data integration can be organized as follows.

The critical point is not merely to “visualize” this data in a dashboard, but to deliver it as actionable information that allows frontline staff to change their behavior the following day. “Yesterday, there was a period when this product was out of stock. Confirm replenishment before opening today” — data takes root on the floor only when it carries this level of specificity.

4. Measuring Promotional ROI Correctly: Don’t Evaluate on “Sales Increase” Alone

Promotional ROI (Return on Investment) is a metric that shows how much profit (or value) was generated relative to the cost invested in a given promotional activity. Many companies try to measure effectiveness using only the “increase in sales before and after the promotion,” but this is insufficient.

To calculate a more accurate promotional ROI, the following elements must be understood.

• Direct promotional costs: Flyer production costs, SNS advertising spend, in-store POP production costs, discount funding (discount amount × units sold)
• Indirect promotional costs: Staff preparation hours (hours × labor cost), additional ordering and logistics costs
• Incremental gross margin: The increase in sales during the promotional period minus the above costs and the cost of goods sold
• Cannibalization effect: Whether discounting one product is cannibalizing sales of another full-price product
• Impact on repeat purchase rate: Whether customers who visited due to discounts are making repeat purchases at full price, or have become “sale-only shoppers” who only visit during promotions

Measuring these factors quantitatively requires a system capable of cross-analyzing POS data, membership data, and cost data together. In most Japanese-affiliated retail operations in Thailand, these data points are scattered across separate systems and files, requiring enormous manual effort to consolidate. As a result, the decision-making pattern of “just run a discount campaign and judge effectiveness roughly by sales figures” becomes entrenched.

5. Leveraging Demand Forecasting: Breaking Free from “Gut-Feel Ordering”

Another critical pillar in achieving sales without relying on discounts is inventory and ordering management based on demand forecasting. When demand forecasting accuracy is low, the following problems arise.

• Stockouts of best-selling products → lost sales opportunities → leads to the thinking that “it’s faster to move inventory with discounts”
• Excess inventory of slow-moving products → dead stock → disposal and markdown costs accumulate
• Inability to respond to demand peaks and troughs → staff overtime and holiday work → increased labor costs

In an environment like Thailand, demand forecasting must account for seasonal cycles different from Japan (Songkran, Loi Krathong, Chinese New Year, etc.), weather disruptions from climate change, and consumer characteristics that vary by region. Relying solely on the “intuition” of experienced buyers has its limits; quantitative forecasting that combines historical sales data, visitor data, and event calendars has become indispensable.

In recent years, cloud services and ERP systems increasingly incorporate demand forecasting functionality, making it possible to achieve a reasonable level of demand forecasting without hiring specialized data scientists. What matters more than adopting “advanced AI” is first cleaning up and organizing historical sales data, then building an operational framework that allows continuous comparison of forecasts versus actuals.

6. Segment-Based Promotions: Why You Don’t Need to Discount for Everyone

Another perspective for breaking the assumption that “you can’t sell without discounting” is promotional design by customer segment. Offering the same discount to all customers means incurring unnecessary costs even for price-insensitive customers — those who would have purchased at full price anyway.

A rough customer segmentation reveals the following types.

• Loyal customers (high frequency / high basket size): “Special treatment,” “early access to information,” and “membership perks” are more effective than discounts. Offering discounts to this segment tends to represent a loss of gross margin.
• Near-loyal customers (moderate frequency): The goal is to increase purchase frequency. Category-specific point rewards and bundle offers are effective.
• At-risk of lapsing customers: Customers who have not purchased for a defined period. Candidates for retargeting campaigns.
• Price-sensitive customers (sale-only shoppers): Customers who do not purchase at full price. The primary target of discount promotions, but their contribution to gross margin is limited.
• New customers: Aim to convert them into repeat shoppers by delivering a high-quality first experience.

When membership data and POS data are properly organized, this kind of segment analysis can be conducted without major expense. The key is to differentiate “who” you are communicating “what” to and “through which channel,” and to measure the cost-effectiveness of each campaign individually.

7. Visualizing Store Operations: Digitizing Daily Reports and Inspection Records

When the conversation turns to data utilization, it tends to gravitate toward “analytics tools,” “AI,” and “BI (Business Intelligence).” However, if the quality of on-site data is low, no analytics tool will make a difference. Data quality begins with the habit of accurately recording what happens on the floor.

At many Japanese-affiliated retail operations in Thailand, paper-based daily reports, inspection records, and checklists are still in use. Digitizing these creates the following benefits.

• Reduces the man-hours required to create and consolidate daily reports (also reduces the burden on Thai staff of writing reports in Japanese)
• Allows headquarters to grasp in real time the operational status of each store and time slot
• Enables improvement instructions to be issued and tracked as tasks (eliminating “I said it / you didn’t say it” disputes)
• Speeds up response to anomalies (sudden inventory drops, frequent complaints, etc.)
• Streamlines the creation of reports for Japan headquarters

Particularly at Thailand-based operations, many sites struggle with reporting communication (reporting, informing, and consulting) between Japanese managers and Thai staff. Language barriers, differences in communication styles, and information sharing with Japan headquarters across time zones — digitizing on-site data is an extremely effective solution to these challenges.

8. Deciding What to Stop and What to Continue Investing In

In the 2026 business environment, rather than continuing all investments, it is critical to sort out “what to stop and what to advance.” Retail operations in particular tend to have multiple investment projects running simultaneously, and projects with unclear ROI are often continued without review.

When conducting this prioritization, the key mindset is “judge by the numbers from the floor.” Investments that are “being continued out of habit” or “that no one has reviewed since the responsible person changed” can be stopped decisively to concentrate resources where they matter.

9. Leveraging BOI: Covering Data, AI, and Automation Investments with Tax Incentives

For Japanese-affiliated companies operating retail and distribution businesses in Thailand, BOI (Board of Investment) incentives are an element that cannot be ignored in investment decisions. In recent years, BOI has offered broad tax incentives for investments including automation, AI, data analytics, and enterprise management IT (ERP, SCM, etc.).

Investment areas potentially eligible for BOI incentives in the retail and logistics sectors include the following.

• Automated equipment for warehouses and logistics centers (automated storage and sorting systems, etc.)
• IT system investments for data analytics and demand forecasting
• Enterprise management systems such as ERP and SCM
• AI-powered demand forecasting and customer analytics systems

The critical point is to consider BOI applications not after a system implementation decision has been made, but from the investment planning stage. Attempting to apply retroactively may result in failing to meet the requirements. In addition, BOI requirements are periodically revised, so it is strongly recommended to consult official information from the Board of Investment of Thailand for the latest details.

TOMAS TECH works with Japanese-language-capable partners who have experience handling BOI applications, and can provide consultation from the investment planning stage.

10. The 3-Year Payback Calculation: The Logic for Explaining to Japan Headquarters

To gain approval from Japan headquarters for DX and data utilization investments at a Thailand operation, it is not sufficient to say “it will be more convenient” or “we’ll be able to visualize things.” Presenting the investment recovery (ROI) over 3 years in concrete numbers is the most direct path to obtaining approval.

For data utilization and system investment in retail, the 3-year payback calculation framework can broadly be divided into “cost reduction effects” and “recovery of lost opportunity.”

Examples of cost reduction effects:

• Reduction in storage and disposal costs through improved inventory turnover
• Reduction in ordering process man-hours (staff time × labor cost rate × annual number of transactions)
• Reduction in administrative man-hours through automation of daily report and document creation
• Elimination of excess staffing (shift optimization)
• Right-sizing promotional expenditures (discontinuing ineffective discount campaigns)

Examples of lost opportunity recovery:

• Recovery of sales opportunities through reduced stockout rates (1% reduction in stockout rate × annual sales × gross margin rate)
• Increase in average transaction value through segment-based promotions
• Improvement in customer lifetime value (LTV) through higher repeat purchase rates

These figures can be estimated as approximations if you have current operational data (stockout rates, disposal rates, labor costs, ordering man-hours, etc.). Presenting the “current state analysis” alongside “projected values after improvement” creates the foundation for an investment business case.

11. Failure Patterns and How to Avoid Them: Common Pitfalls at Thai Operations

There are commonly observed failure patterns in data utilization and DX promotion at Japanese-affiliated retail operations in Thailand. Knowing them in advance significantly increases your ability to avoid them.

Failure Pattern 1: Tools are implemented but usage never takes root
Cases like “we introduced a high-feature POS system, but staff couldn’t master it and ended up managing everything in Excel” are extremely common. The way to avoid this is to gather input from frontline staff before implementation, select a system with an intuitive UI, and include post-implementation training and adoption support in the project plan.

Failure Pattern 2: Only Japanese managers look at the data; it never reaches the floor
“Every morning the Japanese manager checks the dashboard. But nothing has changed for the Thai staff on the floor” — if analysis results are not translated into action on the ground, data utilization produces no results. The way to avoid this is to deliver data to Thai staff in their language and in a format they can use, and to build a system that converts insights into tasks specifying “what to do today.”

Failure Pattern 3: Attempting to implement a large-scale system all at once
Plans like “first integrate POS, inventory, accounting, CRM, and analytics tools all at once” tend to balloon in cost and timeline, and carry a high risk of stalling midway. The way to avoid this is a phased approach: start with “one store,” “one product category,” or “one business process,” measure the results, and then roll out horizontally.

Failure Pattern 4: The system becomes dependent on a single person
A situation where “only one person knows anything about this system” is particularly dangerous in Thailand’s high-turnover environment. The way to avoid this is to create operational manuals in both Thai and Japanese and build a structure in which multiple people can operate the system.

Failure Pattern 5: Continuing without measuring ROI
A situation where “it’s been a year since implementation but we don’t really know what the effect has been — but we can’t stop it” is also dangerous. The way to avoid this is to define “KPIs” and “measurement methods” at the time of implementation and review them on a regular schedule.

12. A Phased Implementation Roadmap: Where to Start

“We want to start using data, but we don’t know where to begin” — this is one of the most common sentiments we hear. The phased approach recommended by TOMAS TECH is as follows.

Step 1: Understand the current state (Months 1–2)
Organize what data currently exists and where. For POS, inventory, ordering, accounting, and daily reports — confirm whether each data source is digital or paper-based, who manages it, and how frequently it is updated.

Step 2: Take the first step (Months 2–4)
Tackle “the single area with the strongest sense of urgency.” For example, if “inventory management accuracy is low and stockouts and excess inventory are ongoing problems,” start with building out the inventory management system. If “it takes two hours every day to consolidate store daily reports,” start with digitizing daily reports. Start small, and measure results.

Step 3: Data integration (Months 4–8)
Using the system established in Step 2 as a foundation, expand integration with adjacent data. If inventory management is in order, try connecting it with POS data to implement automatic reorder point settings, and so on.

Step 4: Application to analysis and decision-making (Month 8 onwards)
Once data is organized and integrated, apply it to analytical work such as demand forecasting, customer analysis, and promotional ROI measurement. This step, if Steps 1–3 have been properly completed, can often be realized without major additional investment.

Step 5: Horizontal rollout (Year 1 onwards)
Once results have been demonstrated in one store, department, or business process, roll out to other stores and operations. At this stage, standardized KPIs and reporting formats become a valuable asset.

TOMAS TECH’s Perspective

TOMAS TECH is headquartered in Bangkok, Thailand, and provides IT system implementation and operational support to Japanese-affiliated manufacturers, logistics companies, and retailers throughout Thailand and the broader ASEAN region. Here is a concise overview of the value TOMAS TECH can offer to address the challenge described in this article — “selling through data, rather than discounts.”

Inventory Management System PEGASUS: PEGASUS, TOMAS TECH’s inventory management system, provides real-time visibility into warehouse and store inventory, and supports reorder point management, inbound/outbound inventory management, and inventory turnover analysis. It enables a shift away from person-dependent inventory management practices — “ordering by gut feel,” “not knowing what’s where” — to reduce both stockouts and excess inventory. It is also used in retail settings for back-of-store inventory and multi-location inventory management.

Paperless App i-Reporter: i-Reporter is an application for digitizing paper-based daily reports, inspection records, and checklists. On-site information entered by Thai staff via smartphone or tablet can be delivered to managers in Japanese. It supports the diverse documents generated at retail sites — store daily reports, hygiene inspection records, product display checklists, and more — and contributes to reducing the man-hours spent on paper consolidation while improving on-site data quality.

Operations Management System: A system for visualizing staff work status, working hours, and productivity. In retail settings, it can be used to optimize staffing relative to time-slot visitor volume and to analyze labor productivity. It supports improved shift planning precision and right-sizing of labor costs.

Smartwatch System: A system through which frontline staff receive real-time instructions via smartwatch and can report task completion. It increases the speed of floor-level communication for tasks such as replenishment instructions, anomaly reporting, and cleaning tasks.

TOMAS TECH’s approach is a phased method: start with a small unit — “one warehouse,” “one store,” “one form” — establish it on the ground, and then roll out horizontally. Including post-implementation Japanese-language support, we provide continuous assistance with the practical operational challenges of running a Thailand-based operation. If you are interested, please feel free to contact us through the TOMAS TECH contact page.

Summary

For Thai retailers to protect sales and gross margins without relying on discounts, it is essential to return to the fundamental principle of “connecting data to on-site decision-making.”

Here is a summary of the key points covered in this article.

• Recognize the structural problems of discount dependency (margin compression, inventory disruption, unclear promotional effectiveness) and use data to understand the root causes
• Build a basic architecture that “connects” POS, inventory, ordering, customer, promotional expenditure, and operations data
• Measure promotional ROI to include not just “sales increase,” but also discount funding, indirect costs, cannibalization, and repeat purchase rates
• Use demand forecasting-based inventory and ordering management to reduce both stockouts and excess inventory
• Through segment-based promotional design, eliminate the need to discount for every customer
• Improve the quality of on-site data by digitizing store operations (daily reports, inspection records)
• Prioritize “investments to stop” and “investments to continue,” and concentrate limited resources accordingly
• Consider BOI incentives from the investment planning stage to maximize tax benefits
• Use “3-year payback” figures when presenting to Japan headquarters, demonstrating both cost reductions and opportunity loss recovery
• Understand failure patterns in advance and increase adoption rates by starting with a phased, small-unit implementation approach

Thailand’s economic environment in 2026 has entered a phase where relying solely on “increasing sales” is no longer viable. However, by reducing losses in inventory, operations, and promotions, and embedding data-driven decision-making on the front lines, it is possible to build a profit structure that is resilient to economic fluctuations. Not DX as a buzzword, but DX that changes the numbers on the ground — with that perspective in mind, please begin with just one small improvement.

References

• World Bank Thailand — Latest information and reports on Thailand’s economy
• Thailand BOI (Board of Investment) — Incentives for automation, AI, and enterprise management IT investments
• JETRO Thailand — Thailand market information and business environment reports
• S&P Global PMI — Thailand manufacturing and services sector sentiment indices
• METI Manufacturing White Paper 2025 — Digitalization trends in manufacturing and distribution