KisStartup

For many years, discussions about digital transformation have focused mainly on productivity, efficiency, and economic growth. Yet, through the practical journey of the IDAP – Inclusive Digital Adoption Project, KisStartup has discovered a deeper dimension: digital transformation not only reshapes economies but also preserves and enriches culture – the very soul of communities.

In the highland villages of Lào Cai and Sơn La, technology has not arrived as a foreign wave but has gradually become a bridge that brings local identity to the world. It helps revive cultural heritage in new forms — digitized, shared, and passed on to younger generations.

When Technology Becomes the Storyteller of Identity

Through IDAP’s training programs, many artisans, teachers, and ethnic minority youth have learned how to record, store, and present their cultural heritage using digital tools. Small classes on video production, product photography, and content design have become creative spaces where locals learn to “tell their cultural stories” in their own voice.

In Sơn La, lecturer Lò Thị Ngọc Diệp and her students at Tây Bắc University have worked together to digitize Thai cultural traditions — from performances to instructional videos shared on social media.

Such efforts are turning YouTube into an open classroom where cultural values transform into creative products and services — music classes, cultural tours, ethnic embroidery workshops, and traditional handicraft lessons.

In Lào Cai, Dr. Đặng Thị Oanh and her team of lecturers and students are documenting and promoting the spiritual traditions of the Dao people. Meanwhile, Ms. Vàng Thị Mai has been collecting ancient then songs of the Tày in Bản Liền, inspiring others to join her efforts. Her small class has evolved into a community-run hub, where locals teach, learn, and share their traditions online through YouTube and Facebook.

Digital Technology – The Companion of Heritage

Around the world, technology has become humanity’s extended memory, helping to preserve and revive what once seemed lost.

In Guatemala, Duolingo added the K’iche’ language of the Maya people, reviving a tongue once on the brink of extinction. In Bolivia, the OEI App preserves five indigenous languages with pronunciation guides and e-dictionaries, allowing mountain children to reconnect with their mother tongues.

In Australia, Canada, and New Zealand, technologies such as VR360, 3D scanning, and virtual reality are used to recreate festivals, stilt houses, pottery, and ancient dances — allowing global audiences to walk through indigenous spaces from across the world.

Some projects even experiment with blockchain to protect community intellectual property — from textile patterns and indigo-dyeing recipes to traditional medicines — ensuring that traditional knowledge is shared fairly and safeguarded from commercial exploitation.

Emerging technologies like AI, virtual reality, cloud storage, digital linguistics, and blockchain are helping cultural heritage to live again in interactive, accessible forms. Yet, these tools only gain true meaning when local communities themselves are the storytellers.

Lessons from IDAP: When Communities Lead the Digital Journey

Experiences from Lào Cai and Sơn La show that when digital transformation begins with culture, it creates stronger and more lasting connections than any technical project could.

In Bản Liền, the Tày community not only digitized the production process of ancient tea but also used short videos to tell its history — sharing traditions of leaf-picking, tea-making, and festive tea ceremonies. Their fanpage “Hương Trà Bản Liền” (The Fragrance of Bản Liền Tea) has become more than an online shop — it’s a living cultural classroom.

In Bản Lùn, the Thai community organized folk song classes with livestreams, where children learned to sing, film, and edit videos, adding captions and descriptions. The entire village joined in — each person playing a role — creating a community-built digital archive of ethnic music.

IDAP calls this approach “community-led digital transformation” — where technology is localized and owned by the people, not imposed from outside. This allows culture to evolve naturally, while also generating new economic value without losing authenticity.

Preserve to Develop – Develop to Spread

Technology can preserve, but its ultimate purpose is to bring culture back into modern life.

Music lessons, cultural tours, storytelling videos, and handcrafted souvenirs are ways of transforming heritage into tangible value. Each song, dance, and weaving pattern no longer sits behind glass or in research books but is digitized, shared, and becomes a source of inspiration for new generations.

When a student in Sơn La can teach the tính lute on TikTok, or an artisan in Bản Khảo hosts an online indigo-dyeing class, culture has truly entered the digital age — in the most human, patient, and creative Vietnamese way.                            

Challenges and Vision

Digital transformation in culture poses significant challenges: balancing preservation with commercialization, openness with protection of traditional knowledge. Without proper guidance, technology can distort or exploit heritage. Therefore, the core principle must be that ownership of knowledge and content belongs to the indigenous community.

KisStartup and IDAP’s partners aim for a model of sustainable cultural development, where technology plays a supportive role — helping communities to tell their own stories, share them widely, and create their own value.

The Road Ahead

From small filming classes in Sơn La to cultural tourism products in Lào Cai, digital transformation is rekindling the creative flame within each community.

When technology becomes a tool of culture, not its replacement, we can envision a future where national identity does not fade — but finds new life in the digital world.

“Digital transformation does not blur tradition – it gives tradition a new voice.”

And it is the young people, artisans, teachers, and small enterprises in the mountains — together with KisStartup — who are writing this next chapter: making culture more vibrant, more widespread, and more valuable in the digital era.

© Copyright KisStartup. Content developed under the IDAP Project – Inclusive Digital Adoption Program. Any reproduction, quotation, or reuse must cite KisStartup/IDAP as the source.

Image source: https://steveblank.com/2015/05/06/build-measure-learn-throw-things-again...

When KisStartup began disseminating the Lean Startup methodology in Vietnam in 2015 after the Vietnam-Finland Innovation Partnership Programme - IPP2, hardly anyone was talking about MVP, validated learning, or innovation accounting. Most startup classes at the time revolved around writing business plans, and the numbers founders presented were based on intuition rather than evidence. Ten years have passed, and if we had to summarize that journey, KisStartup would choose two words: "co-learning." We didn't teach startups how to "do Lean," but rather lived Lean with them, experimenting together, failing together, and learning to adapt together. And through that process, KisStartup also became a "Lean organization"—a continuous learning organization, operating leanly, and constantly evolving.

Lean Startup – A Mindset for Learning and Management in Uncertainty

Eric Ries's book, The Lean Startup, was born from the author's own failed startup experiences in Silicon Valley. Ries didn't write about "secrets to success," but about how to reduce risk when you don't know what's right. He laid the foundation for the concept of "management of uncertainty"—a type of management where the goal is not to maintain stability, but to learn quickly to adjust quickly.

The five core principles of Lean Startup—from "Entrepreneurs are everywhere" to "Innovation accounting"—became the compass for the first generation of Vietnamese startups that KisStartup supported. But we quickly realized that Lean only truly makes sense when it is reinterpreted to suit the Vietnamese context, where resources are scarce, data is limited, and perseverance is the most precious asset.

For KisStartup, "Lean" does not just mean technical efficiency; it is a mindset of economizing and intelligently using all available resources: time, money, knowledge, people, and especially trust. In a country where most startups begin with small personal savings, loans from friends, or family capital, "Lean" must mean "enough to try, not too much to break."

Lean thus became a philosophy of practice, not a slogan. It reminds us that every decision is a hypothesis that needs testing, every step should generate data, and every failure is a lesson cheaper than a big failure later on.

KisStartup's Choice to "Work Together" Instead of Just "Teaching"

When starting out, KisStartup realized that if they only "taught Lean," the Lean Startup would easily become a theory on paper—easy to understand, but difficult to execute. This is because Lean is not a tool, but a habit of action: getting out to meet customers, asking the right questions, running small experiments, and measuring in a tangible way.

Instead of organizing courses that "talk about Lean," KisStartup chose to design real Lean workshops—where startups had to find customer insights themselves, generate hypotheses, create MVPs, and receive real feedback. We read data together with them, analyzed signals, and many times witnessed the "aha moment" when a founder realized: what they thought the customer wanted and what the customer actually needed were two different worlds.

This "co-working" approach is what made KisStartup different. We did not stand in the position of "instructor" but became the second learner, going through each Build–Measure–Learn cycle with the startup. When startups experimented with products, we also experimented with coaching methods. When they measured customer feedback, we measured the effectiveness of our support activities. And when they failed, we learned how to redesign our service model.

From hundreds of workshops, KisStartup gradually formed its initial incubation models—models that also went through their own Lean cycles: small tests, measuring effectiveness, scaling if successful, and "sunsetting" if they failed to generate learning value. It is thanks to this spirit that KisStartup has been able to survive and thrive for 10 years without becoming "bloated" or falling into the bureaucratic spiral often seen in innovation support organizations.

Lessons from Practice – When Lean Becomes a Mirror

After a decade, KisStartup has accompanied hundreds of founders and organizations on their Lean journey. There are success stories we are proud of, but also many incomplete stories that we still cherish as hard-earned lessons. Over 10 years, KisStartup has encountered every scenario: eager startups, discouraged startups, surprisingly successful startups, and seemingly viable models that quickly collapsed.

One tech startup in the tourism sector was passionate about perfecting their app with all features—maps, booking, payment—only to discover that their target customers—homestays in the mountains—had no need for it. It was only after trying an "MVP without an app"—using just a Zalo group and Google Form—that they truly understood the value they could offer. This lesson became a prime example in KisStartup's training program: The MVP is not a technological product, but a learning product.

Conversely, there were also startups that persevered with experimentation and achieved surprising success. A processed agricultural product team used Lean to identify the "flavor and packaging specifications" most favored by consumers. They didn't invest in a large production line from the start but ran small batches, measured feedback, and then scaled up. After two years, their product was sold in many markets—not due to luck, but due to the discipline of learning.

But it is from these differences that we have distilled three core lessons:

1. The MVP is not a technical product – but a learning product.Many startups in Vietnam spent months perfecting features but never asked customers what truly created value. Lean helped them reverse this: test the value before building the feature.
2. No measurement, no learning.Some startup teams "run Lean" but fail to collect quantitative data—all decisions are still based on gut feeling. We learned how to set up minimal "innovation accounting": clearly define the hypothesis – the metric – and the decision threshold before each experiment cycle.
3. Failure is not scary, only the failure to learn is.The models that stopped the earliest often left the most valuable data—because they showed which hypotheses were wrong, thereby opening up new directions.

“There is no failure, only an incomplete loop.” (KisStartup internal note, 2019)

From these successes and failures, KisStartup draws a simple principle: Lean cannot save every startup, but Lean helps every startup know why they failed. And only by understanding the cause can they get back up on the right track.

When Lean Meets Design Thinking and Effectuation – "Lean with Vietnamese Identity"

Throughout the process of practicing Lean, KisStartup realized that there is no single template for innovation. Lean Startup is very strong in the experimentation and measurement phase, but to deeply understand customers and create true value, it needs Design Thinking, and to start in resource-constrained conditions, it also needs Effectuation—the entrepreneurial mindset focused on achieving results from available resources.

Design Thinking: people at the center of leanness

If Lean Startup answers the question "How to learn fastest?", Design Thinking helps us answer "What to learn from people?" Design Thinking begins with empathy—deeply listening to people's difficulties, needs, and motivations—and from there forming ideas, testing solutions, and continuing to learn from feedback.

By combining Lean with Design Thinking, KisStartup helps startups create products that are not only "market-right," but also "people-right." For example, in community tourism projects in Son La and Lao Cai, instead of starting with the question "How to sell tours?", we guided local groups to start with the question "What are visitors truly seeking when they come to our village?" That question opened up a series of observations, conversations, and service experiments—and each subsequent Lean cycle became deeper because every experiment was based on genuine human insights

Design Thinking, therefore, does not oppose Lean, but complements the "emotional" part of the learning loop—so that the product is not only optimized but also meaningful.

Effectuation: leanness from available resources

While Design Thinking starts from the customer, Effectuation—the theory of entrepreneurship by Professor Saras Sarasvathy (Darden School, University of Virginia)—starts from the entrepreneur themselves. Instead of setting a big goal and then figuring out how to mobilize resources, Effectuation teaches us to start with what we already have: knowledge, relationships, small assets, and belief.

When KisStartup applied Effectuation along with Lean, we saw "leanness" reaching a new depth. Founders no longer worried about "lacking capital," but focused on "what do I have in my hands to start the first test cycle?" A founder in the mountainous region started producing herbal tea from her own family garden. Without waiting for fundraising, she tried selling via Facebook, recorded feedback, adjusted the flavor and packaging, and then scaled up. This is Lean originating from Effectuation—learning by doing, within the constraints of real resources, but full of creativity.

For KisStartup, this spirit is especially suited to Vietnam: don't wait until you have enough to start—start to learn and find a way to get enough.

When AI Accelerates Build–Measure–Learn

In 2025, AI is completely changing the rhythm of Lean Startup. If each Build–Measure–Learn cycle previously lasted weeks, AI now helps shorten it to hours or days:

• Build: Create content, mockups, and simulated scenarios using AI and no-code tools.
• Measure: Automatically collect user behavior and analyze real-time feedback.
• Learn: AI suggests pivots, identifying hidden insights in small data.

Thanks to this, startups—and KisStartup itself—can experiment faster, deeper, and more accurately. But even as technology changes, the Lean spirit remains the same: genuine learning, avoiding "vanity metrics," and making evidence-based decisions.

Lean within KisStartup – The Learning Organization of Co-Learners

When KisStartup helps startups learn Lean, we also apply Lean to ourselves. Every program (such as IDAP – Inclusive Digital Transformation, GEVA – Green Export, or DormLab – Student Laboratory) is built as an organizational MVP: starting small, with a hypothesis, a measurement method, and criteria for adjustment.

If a program brings learning value to both participants and KisStartup, it is scaled up. If not, it is improved or terminated. This approach has allowed KisStartup to maintain the flexibility of a startup throughout 10 years of operation, avoiding operational inertia or dependence on a single model.

It is through this journey that KisStartup understands that leanness is not about reducing scale, but about optimizing meaning—doing less but learning more, doing things right for people, and creating a more lasting impact.

MVP 2025 – Learning from Action, Not Plans

After 10 years of practice, KisStartup has developed the MVP 2025 framework—the "minimal but valuable" version of Lean Startup, suitable for the era of AI and automation.

Today, an MVP not only needs to be "minimal" but must also have a clear learning objective. A good MVP is not the cheapest product, but the product that can generate the strongest signal from the market at the smallest cost.

We often ask founders three questions before they start:

1. Which hypothesis do you want to test first?If you don't know what you're testing, your experiment is meaningless.
2. How will you measure it?No data, no learning.
3. What will you learn if the result is unexpected?Each Lean cycle is only valuable if there is a plan for... failure.

Today, "MVP" is not just "Minimal Viable Product"—it is "Meaningful, Valuable & Practical." The table below is the checklist KisStartup uses when working with startups and designing new services:

Criteria	Verification Question	Practical Example
Core Hypothesis	What are we testing? (need, pricing model, distribution channel?)	“Are customers willing to pay for Product X?”
Learning MVP	Does the experimental version help collect real data?	Selling first via a landing page instead of investing in a website.
Measurement Metrics	Have the pivot/persevere decision metrics been clearly defined?	Number of trial orders > 30 in 2 weeks = continue.
Learning Loop	Is there a plan for improvement after each experiment cycle?	Weekly result review, canvas update.
True Insight	Does the collected data help understand customers deeply, not just "count clicks"?	Analyze feedback to understand why they didn't buy.

When startups can answer these three questions, they have a true MVP in hand. And when they maintain a disciplined Build–Measure–Learn cycle, they are creating a sustainable foundation for learning—not just product development.

The Lean Mindset is Not Outdated, Just Evolving

After ten years, KisStartup realizes that Lean Startup remains one of the most powerful mindsets for dealing with uncertainty. But Lean cannot exist alone. It needs the human element of Design Thinking, the flexibility of Effectuation, and the profound understanding of people in every learning cycle.

Leanness—in its deepest sense—is about living with limits but not being limited, having the ability to learn fast, adapt quickly, and create long-term value even with the smallest resources.

In the AI era of 2025, the Build–Measure–Learn cycle can happen in hours instead of months. But speed is only meaningful when accompanied by depth of learning. That is what KisStartup continues to pursue—not just to help startups succeed, but to build a learning, adaptive, and sustainable ecosystem, where every experiment is directed toward a bigger goal: developing people and businesses in a changing world.

“Lean doesn't mean less—it means learning faster, adapting better to create more sustainable value.” — KisStartup, 10-year reflection

© Copyright belongs to KisStartup. Any form of copying, quoting, or reuse must clearly state the source KisStartup.

Many businesses tend to think of VSS as a “ticket” and ESG as a “scorecard.” In reality, it should be the other way around: ESG is the operational architecture from within — it determines what you produce, how you produce it, how you manage risk, and how you measure performance. Once that internal system operates stably, VSS becomes merely a verification and standardization step — a shared language with buyers. Rigid trade barriers like MRL or EUDR will continue to exist. Therefore, to go far, businesses must turn market requirements into internal capabilities — namely, data, traceability, and SOPs (Standard Operating Procedures).

From ESG to Business Model Design and Renewal

ESG doesn’t make a business “spend more” — if done right, it helps reduce risks, stabilize operations, and enhance credibility. ESG has a profound impact on three core design blocks: value, cost–productivity, and risk–governance.

It forces us to redefine the value proposition — shifting from “cheap and fast” to “stable, transparent, and safe.” By standardizing processes (e.g., saving water, reducing inorganic inputs, segregating and tracing production flows), quality variability decreases, which in turn lowers risk costs — the often-invisible but expensive burden in export. At the same time, ESG builds a “data discipline” that enables businesses to manage technical barriers (for instance, when the default MRL is 0.01 mg/kg if a substance has no specific limit in the EU) and policy risks (as the EUDR requires geolocation of production areas and segregation between compliant and non-traceable goods).

Two Case Studies – From Field to Model

Let’s look at two real-world examples introduced by Dân Việt newspaper to understand how farmers are moving toward sustainable, green production. In practice, ESG is not a PR slogan; it is a set of technical and management decisions — covering soil, water, fertilizer, labor, and data — that lead to stable productivity, lower risk costs, and “audit readiness.”

Case 1 – A Ngum (Bahnar, Gia Lai):
Since 2022, he has eliminated synthetic chemicals, switched to organic–microbial farming, practiced intercropping, and focused on soil ecosystem health. Results: reduced pests, stable yields, over 3.5 tons of coffee beans per hectare, and nearly VND 300 million net income per year. His farm has become a community learning site, showing clear social (S) impact. In terms of the business model, he repositioned his value from “chemical-intensive, yield-driven” to “safe, consistent, ecosystem-based.” With minimal record-keeping and traceability, his natural model aligns well with VSS frameworks that emphasize soil health and farmer welfare.

Case 2 – Nguyễn An Sơn (Đắk Lắk):
Since 2020, he has adopted multi-stem pruning and drip irrigation with a weekly “nutrition menu”, achieving about 40% water savings, 15–30% less inorganic fertilizer, and five-sixths labor savings. Yields reached ≈5.5 tons/ha (about 1.5 times traditional yields), producing 130 tons in 5 years, worth about VND 8.5 billion, with nearly VND 1 billion in annual profit — along with an OCOP 3-star brand. This is ESG through precision farming: saving resources (E), ensuring labor safety (S), and enforcing procedural and data discipline (G). As a result, meeting VSS and technical requirements becomes much easier.

From Practical Cases to the VSS Roadmap

There are no “shortcuts” to VSS compliance. There are two sustainable paths:

• (i) ESG-first – transform technical and management practices to generate standardized data, or
• (ii) Micro-lot-first – start small but compliant to learn fast and minimize “tuition costs.”

Both converge on data–traceability–SOPs.

When ESG practices become routine processes, requirements like MRL, microbial limits, or EUDR become ordinary management indicators. At that point, VSS serves as a “seal of approval” verifying that the system runs effectively — not a “lifebuoy” in crisis.
Certification investment also becomes easier to budget, as actual costs depend on context and scale, including preparation, evaluation, and maintenance — not just the “audit fee.”

Seven Suggested Steps

• Step 1 – Choose “high-impact, low-cost ESG levers”: irrigation, fertilizer, post-harvest hygiene, and labor safety. Identify 2–3 measurable indicators (moisture, fertilizer dosage, PPE work hours).
• Step 2 – Standardize a small pilot process (micro-lot 5–10%): set short SOPs, assign batch codes, separate storage; do it right, fully, and consistently for one crop season.
• Step 3 – Keep disciplined minimal records: digital or paper farming logs, store input receipts, score compliance weekly.
• Step 4 – Measure core technical indicators: test 1–2 lots for MRL, microbiology, heavy metals; refine processes accordingly.
• Step 5 – Make value transparent: use QR/batch codes linked to field photos, geolocation, simplified SOPs; tell the story of “saving water/reducing fertilizer/ensuring food safety.”
• Step 6 – Cross-check with VSS and test negotiation: compare micro-lot results with 15–25 minimum criteria of a target standard; test-tiered pricing with buyers (small contracts, seasonal improvement clauses).
• Step 7 – Tell your story: share authentic, transparent experiences to build a community of practice. Consolidated data will provide a full picture when proof is needed.

These seven steps effectively “package ESG” into business modules — each creating a data asset and a new operational capability — the very elements VSS measures and customers pay for when you can prove them.

Avoid Two Common “Traps” in VSS Implementation

In reality, costs often “inflate” because of mindset traps, not the standards themselves. Businesses tend to avoid action or fall into one of these traps, turning VSS into a burden:

• Trap 1 – Substitution Trap: believing that “having certification = exemption” from technical barriers. Wrong — MRL, microbiological, and EUDR rules are hard barriers. VSS merely helps structure your processes to overcome them consistently.
• Trap 2 – Overextension Trap: adopting multiple standards before having strong data–segregation foundations. The solution is to focus on one core standard aligned with your target segment; build a solid micro-lot before expanding.

To join global value chains and retain value, businesses must turn market requirements into internal capabilities. The shortest path is to redesign the business model around ESG, then use VSS to standardize and demonstrate performance.
From A Ngum (Gia Lai) to Nguyễn An Sơn (Đắk Lắk), both cases prove that ESG is a set of technical and managerial decisions that yield stable productivity, transparent data, lower risk costs — and thereby reduce expenses and increase success probability when pursuing VSS certification.

Note: At the time of writing, the EUDR has been announced by the European Commission to be postponed for another year due to technical reasons, pending approval by the Parliament and member states — but the direction toward geolocation, traceability, and supply segregation remains unchanged.

#GreenExportMindset #GreenExport #ESG #GEVA #KisStartup

© Copyright KisStartup. Content developed under the GEVA Project – Green Export Acceleration through Voluntary Sustainability Standards (VSS). Any reproduction, citation, or reuse must credit KisStartup/GEVA.

References

• EU – MRL (0.01 mg/kg default when no specific MRL): European Commission, EU legislation on MRLs (Food Safety)
• EU – EUDR (traceability, geolocation, compliant/non-compliant segregation): European Commission Green Forum, Traceability and geolocation of commodities subject to EUDR
• EUDR – One-year postponement update: Reuters; Financial Times
• VSS – Market trends and data: ITC, State of Sustainable Markets 2023
• Certification cost structure (context-dependent): Rainforest Alliance, How Much Does Certification Cost?; Fee Catalogue for Certification Bodies
• Field example – Gia Lai (A Ngum, organic–microbial farming): Dân Việt; Báo Gia Lai
• Field example – Đắk Lắk (Nguyễn An Sơn, multi-stem, drip irrigation, OCOP 3-star): Dân Việt; Báo Đắk Lắk

On October 14, 2025, the Government and the Ministry of Science and Technology issued Decree No. 264/2025/ND-CP - an important milestone opening a legal framework for venture investment activities in Vietnam.
For those who have worked for many years in the field of incubation and support for innovative businesses, we see this as a meaningful step forward, contributing to strengthening the confidence of the startup community, creating conditions for technology ideas, business model innovation and sustainable development to be better nurtured.
KisStartup wishes to continue to accompany partners, management agencies and investors to promote the Vietnamese innovation ecosystem to develop strongly, transparently and sustainably.

See details of this important Decree
https://thuvienphapluat.vn/van-ban/Dau-tu/Nghi-dinh-264-2025-ND-CP-Quy-d...

KisStartup JSC
Accompanying Vietnamese innovation
#KisStartup #Innovation #VentureCapital #StartupVietnam #OpenInnovation #VentureCapital #StartupVietnam

 

Digital transformation (DX) has become a matter of survival for Vietnamese enterprises, especially SMEs. However, according to recent surveys, the process faces a series of challenges: high investment costs, lack of digital human resources, limited technological infrastructure, reluctance to change, and the absence of a clear strategy. Only 7.6% of businesses have a well-structured digital transformation plan, while 48.8% have experimented with some solutions but failed to sustain them (Annual Report on Business Digital Transformation, 2022).

In this context, universities – with their combined roles in education, research, and knowledge connection – hold great potential to become strategic partners of enterprises in digital transformation. To support universities wishing to engage more deeply in enterprise digital transformation and to leverage their strengths, KisStartup presents a detailed analysis based on Hồ Tú Bảo’s six-step digital transformation framework. For each stage, corresponding university actions or programs are proposed to highlight their role as knowledge transfer hubs within the digital transformation ecosystem.

1. Awareness and Mindset Change

• Enterprise needs: Most SMEs lack a clear understanding of what digital transformation means or what practical benefits it brings. Many believe it simply means “buying new software.” The biggest barrier lies in management mindset and fear of change. According to a 2025 nationwide survey, 69% of businesses only use email or basic accounting software, without adopting more strategic digital solutions (Ministry of Science and Technology, 2025).
• What universities can do: They can organize awareness workshops, publish research reports on technological trends, or develop Digital Maturity Assessment tools to help SMEs evaluate their readiness. This model is common across Europe and feasible in Vietnam. With academic credibility, universities can better persuade business leaders who often distrust private service providers. International example: European universities have developed Digital Maturity Assessment Tools for SMEs.

2. Defining a Digital Transformation Roadmap

• Enterprise needs: SMEs often lack clear strategies or plans. Many initiatives are abandoned midway, causing waste. According to the Ministry of Planning and Investment (2023–2024), micro and small businesses face particular difficulties due to limited capital, human resources, and technical capability.
• What universities can do: Faculties of economics, IT, or management can develop digital readiness assessment frameworks and offer consulting services to build 6-month to 3-year roadmaps tailored to business size. Final-year students can participate as “junior digital consultants,” gaining practical experience while supporting companies.International example: University of Vaasa (Finland) successfully implemented an ecosystem-based digitalization model for local SMEs, yielding mutual benefits.

3. Building Digital Capabilities

• Enterprise needs: The lack of skilled personnel is the most critical barrier. The Enterprise Development Agency (2023) reported that most SMEs lack adequately trained staff to implement digital solutions effectively. External expert services are often unaffordable.
• What universities can do: Design short-term, hands-on training courses using real company data and workflows—for example, training business owners in basic data analytics, marketing staff in digital campaign management, or accountants in data security. International example: IE University (Spain) and Banco Santander launched the “Digitaliza tu negocio” program, providing digital skills training to over 3,000 SMEs. Vietnamese universities can replicate this through short-term certificate programs and online training for broader reach.

4. Identifying Core Technologies

• Enterprise needs: Amid countless ERP, CRM, AI, and IoT solutions, many SMEs struggle to choose the right technology. Wrong decisions lead to wasted investment. Moreover, their infrastructure is often weak and lacks proper devices, software, or cybersecurity systems.
• What universities can do: IT or engineering schools can establish digital technology laboratories where businesses can test solutions before purchasing. Universities can also host technology showcase events featuring multiple vendors, acting as independent technology advisors. International example: The Hartree Centre (UK) partners with universities to let SMEs experiment with AI and supercomputing before making investment decisions.

5. Implementation and Execution

• Enterprise needs: During implementation, challenges arise not only in technology but also in change management: data cleanup, process adaptation, and employee resistance. SMEs often lack mentors to accompany them through the process.
• What universities can do: Deploy research teams, faculty, and students to accompany companies in pilot phases, acting as “light PMOs.” Universities can also establish co-living collaboration models where both sides share costs and co-develop technology applications. International example: Germany’s Mittelstand 4.0 program has proven effective by organizing workshops and coaching SMEs to apply agile and design thinking methods to reduce implementation risks.

6. Business Model Transformation and Operational Adjustment

• Enterprise needs: After applying technology, SMEs must adjust their business models and operations. This is the hardest step, involving organizational culture and long-term strategy. Most Vietnamese SMEs lack experience in using data-driven insights to adapt their models.
• What universities can do: Conduct local SME case studies, organize peer-learning sessions, and support data analysis on customer feedback to help companies refine products, sales channels, and pricing strategies. International example: Utrecht University (Netherlands) collaborated with consulting firms to digitally transform management and learner experience systems—proving that model transformation must be data-driven.

Surveys in Vietnam show that SMEs face major obstacles: high investment costs, shortage of skilled personnel, reluctance to change, weak infrastructure, and lack of clear strategies. In this context, universities have the potential to become strategic pillars—training digital talent, providing applicable knowledge, connecting businesses with technology, and accompanying them throughout the digital transformation journey.

If leveraged effectively, Vietnamese universities can go beyond teaching students to become digital transformation hubs for SMEs, directly contributing to the sustainable development of the national digital economy.

Summary Table

 

Digital Transformation Step (Hồ Tú Bảo)	Vietnamese SME Needs	What Vietnamese Universities Can Do	International Example
1. Awareness & Mindset	Unclear about DX; 69% using only email/accounting tools	Organize mindset-opening workshops; create readiness assessment tools; publish trend reports	European universities’ Digital Maturity Tools
2. Roadmap Definition	Only 7.6% have formal plans; many abandon efforts	Develop readiness frameworks; offer roadmap consulting; involve student consultants	University of Vaasa (Finland) ecosystem-based model
3. Capability Building	Lack of digital workforce; high training costs	Short-term practical courses; “learn by doing” projects; student-SME support bank	IE University (Spain) & Banco Santander – “Digitaliza tu negocio”
4. Core Technology Selection	Hard to choose solutions; weak infrastructure	Build digital labs; host technology showcases; act as neutral advisors	Hartree Centre (UK) AI & HPC testing
5. Implementation	Lack of mentors; employee resistance; messy data	Faculty-student support teams; co-living collaboration models; mentor programs	Mittelstand 4.0 (Germany) agile workshops
6. Business Model Change	Hard to shift culture; lack of data-driven decisions	Conduct local case studies; analyze customer data; foster innovation networks	Utrecht University (Netherlands) management digitalization

 

© Copyright KisStartup. Developed within the IDAP Project – Strengthening an Inclusive Digital Transformation Ecosystem. Any reproduction, quotation, or reuse must cite the source: KisStartup/IDAP.
Source: https://qnu.edu.vn/vi/hoi-nghi-hoi-thao/bai-noi-chuyen-dai-chung-chuyen-...

     

In many discussions with businesses and farming households, the GEVA project has observed a common reality: the concept of Voluntary Sustainability Standards (VSS) often causes confusion. In theory, VSS are designed as voluntary options that businesses can choose to adopt in order to demonstrate their commitment to sustainable development. However, in practice, VSS are increasingly becoming a “soft barrier” that is almost mandatory for companies wishing to enter high-value export markets. In fact, many countries have begun to formalize parts of VSS into legislation to raise production quality standards and facilitate exports.

This shift stems from changes on multiple fronts—governments, large buyers, and most importantly, consumer behavior. Many countries and economic blocs such as the European Union, the United States, Japan, and Canada have integrated sustainability standards, including VSS, into their trade and public procurement policies (ISEAL, 2023). On the private sector side, multinational corporations use VSS as a “common language” to assess and select suppliers. Therefore, even when regulations do not explicitly require a specific certification, commercial practices effectively make VSS an indispensable condition for trade.

It is important to note that achieving a VSS certification does not guarantee customs clearance. Mandatory technical barriers—such as maximum residue levels (MRL) for pesticides, microbiological tests, heavy metal checks, and food safety standards—still apply. For example, in cases where no specific MRL has been established, the EU default level is just 0.01 mg/kg—a very stringent threshold that forces farmers to change their fertilizer and pesticide practices (European Commission, 2023). In other words, VSS help standardize production processes and enhance credibility, but they do not replace mandatory legal requirements.

At the same time, new layers of requirements are emerging. A prime example is the EU Deforestation Regulation (EUDR), which obliges exporters of coffee, cocoa, wood, rubber, and other commodities to prove that their products are not linked to deforestation after December 31, 2020. This regulation requires precise geographic coordinates of the production area and a traceability system that can distinguish compliant and non-compliant batches (European Commission, 2023). Although the enforcement deadline has been extended until the end of 2025 for medium and large enterprises and until 2026 for small enterprises, the message is clear: no transparent data, no export.

In this context, many businesses have adopted short-term survival strategies—selling fast and cheap rather than investing in small but certified batches. This “survival choice” reflects three underlying factors:

• High conversion costs – including investment in data systems, staff training, and certification fees, which are a major burden for SMEs (Rainforest Alliance, 2024).
• Behavioral inertia – particularly in agriculture, where changing production habits is much harder than changing techniques.
• Short-term cash flow pressure – forcing businesses to prioritize large-volume, low-margin sales to stay afloat rather than investing in new models that take time to mature.

However, this approach cannot build long-term competitiveness. The State of Sustainable Markets report by ITC, FAO, and IISD shows that VSS-compliant agricultural areas and production volumes continue to grow annually across crops such as coffee, cocoa, and rubber (ITC/FAO/IISD, 2023). This means that early adopters are steadily gaining market trust and competitive advantage, while those focusing solely on low-cost, fast exports risk being excluded from high-quality supply chains.

The key to overcoming this challenge lies in a mindset shift. Instead of passively reacting to buyer demands, businesses should proactively embrace ESG (Environmental – Social – Governance) as a core business goal. ESG should not be viewed as a cost but as a foundation for operational efficiency, risk reduction, brand reputation, and investment opportunities (CRIF Digital, 2024; EdenSeven, 2023). Meanwhile, VSS should be regarded as a measurement and roadmap tool—helping businesses understand their current status, identify improvement areas, and transparently demonstrate progress to customers.

Initial actions do not need to be complex. Businesses and farmers can start by:

• Assessing current status: compare technical requirements of target markets with actual production conditions.
• Building a data handbook: record farming logs, maps of production areas, and participant lists.
• Product segregation: minimize mixing risks through batch codes and separate storage.
• Testing micro-lots: apply strict standards to 5–10% of production as pilot certified lots.
• Studying suitable VSS: understand core requirements before registering for certification.

In short, VSS are no longer an optional choice but have become a crucial tool for accessing export markets. At the same time, ESG must serve as an internal foundation. The shift from a “fast-and-cheap” mindset to a “proactive-and-sustainable” strategy is the only viable path for Vietnamese enterprises and farmers to increase product value and seize opportunities in global supply chains.

To assess your readiness for green transformation, you can use the tool developed by KisStartup under the GEVA project:
https://greenexport.vn/vi/bo-cong-cu-do-luong-muc-do-tuan-thu-tieu-chuan...

© Copyright by KisStartup. This content was developed within the framework of the GEVA project – Incubating and Accelerating Green Exports through Voluntary Sustainability Standards (VSS). Any reproduction, quotation, or reuse must cite KisStartup/GEVA as the source.

References
[0] ITC/FAO/IISD (2023). State of Sustainable Markets.
[2] CRIF Digital (2024). Integrating ESG for sustainable business growth.
[4] Social Value Portal (2023). Social Value and ESG: What’s the difference?
[5] EdenSeven (2023). ESG as a bolt-on vs. strategic integration.
[11] European Commission (2023). EU Deforestation Regulation (EUDR).
[13] European Commission (2023). Maximum Residue Levels (MRLs) for pesticides.
Rainforest Alliance (2024). Certification costs and assurance system.

On October 11, 2025, the event “Introducing the Inclusive Digital Transformation Model – Developing the Digital Ecosystem for Enterprises & Launching EduDX Connect”, organized by KisStartup JSC, took place successfully. The event marked a major milestone — one year of implementing the IDAP - Inclusive Digital Acceleration Program, funded by the Australian Government through the GREAT Project, to strengthen inclusive digital ecosystems for MSMEs in Lao Cai and Son La.

Looking Back on a Year When Digital Transformation Became Tangible

In her opening remarks, Ms. Nguyen Dang Tuan Minh – CEO of KisStartup, reflected on more than a year of developing and refining the Inclusive Digital Transformation Model, based on the approach:

“Enterprise-centered – Market-driven – Co-development among stakeholders.”

The model has helped over 200 local enterprises and organizations enhance their digital capabilities, while connecting universities, technical and human resource service providers, and support organizations to form a sustainable digital transformation ecosystem.

From pilot activities in Lao Cai and Son La, the project has achieved notable results:

• Thai Nguyen University – Lao Cai Campus became the first university in Northern Vietnam to launch a Bachelor’s Program in Digital Economics, marking an important step in developing high-quality digital human resources in mountainous areas.

• Tay Bac University actively collaborated with KisStartup to integrate training, coaching, and enterprise engagement into its programs in Son La, turning digital transformation into a key component of education, research, and community service.

• Local technical and human resource service providers were established for the first time, reducing reliance on external experts.

• Hundreds of SMEs, cooperatives, ethnic minority women, and persons with disabilities (PWDs) stepped out of their comfort zones, adopting digital tools, building online businesses, and confidently connecting with domestic and international markets.

Real Stories – Real People – Real Results

The ceremony also honored 12 “Digital Transformation Stars”, representing hundreds of enterprises that have made remarkable progress throughout the year.

Among them, Mr. Nguyen Huu Hau, owner of Phuc Hau Woodcraft (Son La), a person with severe disability, has refused to let his condition define his limits. Through IDAP’s coaching sessions, he learned to use social media, create photos and videos, and build an online brand. Once a craftsman working quietly behind the scenes, he now runs his own fanpage, shares his production process, and tells his story through authentic visual content. His products have reached customers across Vietnam, and more importantly, he has become a source of inspiration for the disability community, training others to use smartphones and digital tools for online business and independent living.

Meanwhile, Ms. Ma Thi Luyen, founder of Luyen Tho Meat Processing (Lao Cai), exemplifies the power of persistence and learning. Once unfamiliar with computers and digital tools, she learned content planning, photography, design, and online sales management through the project. Despite a slow start, she consistently practiced and even hired local service providers to strengthen her digital presence. Today, her business operates effectively online, and she has become a community leader inspiring other ethnic minority women to embrace digital transformation.

EduDX Connect – Bridging Knowledge and Business

Within the same event, EduDX Connect was officially launched — an initiative by KisStartup to connect universities, enterprises, professional organizations, and technology providers for promoting innovation, digital transformation, and commercialization of research outcomes.

Built on the foundation of the IDAP Project, EduDX Connect inherits the proven “Triple Helix” collaboration model piloted in Lao Cai and Son La, aiming to:

• Develop a qualified digital workforce,

• Share best practices, and

• Foster long-term partnerships among academia, industry, and support institutions.

The network was founded by 16 core members representing three key sectors:

• Higher education institutions, including universities and colleges from Hanoi, Hai Phong, Lao Cai, and other key northern regions;

• Professional organizations, including national entrepreneurship support centers and professional associations;

• Technology and digital training enterprises, spanning e-commerce platforms, software solution providers, and online education companies.

Together, they form a bridge between knowledge, technology, and the market, advancing an inclusive and innovative digital transformation ecosystem.

 

Spreading the Spirit of “More Inclusive – Better Connected – Strongly Disseminated”

Voices from the Ground

Ms. Vang Thi Moi, founder of Moi Design in Lao Cai, shared her journey “from hesitant to confident” through the program. She learned how to work with clients professionally and expanded her service toward inclusive design for persons with disabilities, contributing creative products for local businesses.
Ms. Nguyen Hong Giang, expert supporting PWD groups, shared her experience working with over 20 individuals in Lao Cai and Son La, helping them gain confidence and access economic opportunities through digital tools. She emphasized that empathy, patience, and trust are the keys to ensuring genuine inclusion in digital transformation.

Commitments for Future Action

• Thai Nguyen University – Lao Cai Campus reaffirmed its cooperation with KisStartup in five key areas: digitalizing enterprise processes, developing digital marketing, and maintaining the lecturer–student–enterprise collaboration model.

• Tay Bac University committed to integrating business-support activities into student training programs.

• Technology companies in the EduDX Connect network — such as UNICA, FINAN (Sobanhang), and Onelog — pledged long-term engagement, providing tools, internships, and opportunities for students to apply digital solutions in local contexts.

The event concluded with heartfelt gratitude to all partners who have accompanied the IDAP journey — from businesses, lecturers, and students to local organizations and donors.
The message echoed throughout the program:

“Digital transformation is not just about technology — it is about unlocking the potential of people and communities.”

From mountain villages and small cooperatives to universities and tech companies, everyone is working together to build an inclusive digital ecosystem — a place where everyone has the opportunity to learn, connect, and grow sustainably.

 

In the digital age, AI has become a familiar tool for businesses in planning, customer care, market research, and content creation.However, behind every AI command is a data center that consumes electricity, water, and emits CO₂. Without mindful usage, the environmental cost can quickly exceed expectations.

KisStartup – with experience supporting thousands of businesses on their innovation and digital transformation journeys – has compiled this guide to help companies use AI responsibly, efficiently, and in an environmentally friendly way.We call it Green Standard Prompting: boosting productivity while reducing emissions.

Why do we need "Green Standard Prompting"?
Every AI command consumes energy and water:
- Gemini (Google): approx. 0.24 Wh, emits 0.03 gCO₂, and uses 0.26 ml of water per average text prompt.
- ChatGPT (GPT-4o): approx. 0.3 Wh per prompt.

1 million prompts can consume around 300 KWh, equal to a household’s electricity use in one month.
So, every time you revise a prompt repeatedly, you're multiplying the power and water usage. That’s why carefully crafting your prompt is not only time- and cost-effective, but also a clear ESG (Environmental, Social, Governance) action.
Principles of Green Standard Prompting:

1. Be clear about your goal: what you want, for whom, and in what format.
2. Provide enough context: product, data, constraints.
3. Limit output length: specify word count or number of bullets.
4. Choose the right model: simple tasks → lightweight models.
5. Ask AI to request more info if needed, instead of guessing.​
6. Save and reuse good prompts to avoid repetition.

Example Green Standard Prompts
1. Planning Content Marketing
System (Role): You are a Sustainable Content Marketing expert.
User Prompt:

• Goal: Plan 2 weeks of content for a {industry} fanpage.
• Audience: {target customers}
• Context: product {...}, USP {...}, budget {...}
• Output (≤200 words):
  - 14-day content calendar
  - Captions ≤30 words
  - Hashtags ≤5 per post​
• Constraints: prioritize repurposing existing content, ask up to 3 follow-up questions if data is missing.

2. Developing a Green Export Plan
System (Role): You are a Green Export & ESG expert.
User Prompt:

• Goal: Create a 6-month export plan for {product} to {market}.
• Context: certifications, production capacity, current partners
• Output (≤250 words):
  1. 5 green requirements/VSS (Voluntary Sustainability Standards) for the marke.
  2. 3 current capability gaps
  3. 3 priority actions for the first 90 days
  4. 2 long-term opportunities​
• Constraints: include a checklist for executives, ask up to 5 follow-up questions if data is missing, cite sources.

​Steps to Build a “Green Standard” AI Assistant for Your Business

1. Define the assistant's role (e.g., Content Coach, Export Advisor).
2. Standardize the system prompt (role + green principles).
3. Create a sample prompt library (like examples above).
4. Train with real data (products, certifications, customer info).
5. Test & refine to minimize prompt iterations.
6. Integrate into workflows (chatbot, CRM, Notion/Slack).
7. Monitor & report on resource savings (tokens, kWh, CO₂, water).

Checklist Green AI Prompting
Before you type a command:

• Is the goal, target audience, and output format clear?
• Is the data sufficient so AI doesn’t have to guess?
• Have you set an output length limit?

When choosing a model:

• Do you really need a large model?
• Are you asking for images/slides unnecessarily when text suffices?

During execution:

• Does the AI ask follow-up questions when data is missing?
• Can this prompt be reused?

After completion:

• Is the output immediately usable, or does it require re-running?
• ​Can the prompt be shared with teammates?

Digital Transformation (DT) is no longer just about applying technology; it is increasingly seen as a journey of growth that enhances the self-learning ability of business owners. For small and medium-sized enterprises (SMEs), the capacity for continuous and independent learning is a key factor to adapt to a volatile market and to drive strategic decision-making (McKinsey, 2023). Drawing on experience from projects implemented by KisStartup in Vietnam, this article highlights how DT – especially through Artificial Intelligence (AI) – empowers entrepreneurs to strengthen their self-learning skills, which in turn directly reshapes their strategic choices.

 

Self-directed learning has long been considered an important ability for organizations to adapt (Knowles, 1975). In the digital age, the integration of AI tools turns learning activities from passive intake into an active process of exploration and strategic reflection. Instead of asking “What is this concept?”, business owners gradually learn to search by themselves, experiment, and test strategic options through digital platforms.

Recent studies confirm that AI not only automates processes but also creates personalized learning environments for managers and employees. Platforms such as LinkedIn Learning, Coursera for Business, and Udemy Business have applied AI algorithms to analyze learning progress, detect knowledge gaps, and suggest suitable skill paths (MISA, 2023). This shortens the time to acquire new capabilities and at the same time promotes the adjustment of business strategies.

AI and Business Effectiveness

Adoption rate and learning effectiveness
Global surveys show that the rate of AI adoption in business management has increased rapidly. According to IBM, Forbes, and McKinsey, the percentage of businesses applying AI rose from 33% in 2022 to 72% in 2024 (SkillsBridge, 2023). Another study of 7,500 companies showed that 35% had integrated AI into their processes, while 42% were experimenting (IBM, 2023).

The effect on learning is very clear. AI-based training systems can detect skill gaps, forecast progress, and adjust the curriculum accordingly (AMIS, 2024). Thanks to this, business owners develop a habit of guided self-learning, which is both repetitive and evidence-based.

Performance and strategic innovation
AI-based automation brings significant improvements in performance:

• Businesses report a 20–30% increase in labor productivity thanks to data analysis and decision support (McKinsey Global Institute, 2023).

• In customer service, AI helps increase productivity 1.71 times while reducing staff from 600 to 350 people (Nhân Dân, 2024).

• The application of AI in work management saves an average of 5.4% of weekly working time (~2.2 hours per employee) (Louis, 2024).

These numbers show that AI not only optimizes processes but also creates conditions for employees to focus more on strategic activities, giving business owners more space for critical thinking and strategic innovation.

Cost optimization and human resource development
AI also helps reduce operating costs by up to 25% (Gartner, 2023). This saving allows SMEs to reinvest in training and innovation. When AI is integrated into human resource management and development, entrepreneurs themselves become active learners, ready to test different pricing scenarios, market strategies, and partnership models.

KisStartup’s Approach: Stronger Businesses through Smarter Entrepreneurs

KisStartup’s projects show that digital transformation is not about “doing things for businesses,” but about empowering them to do it themselves. Businesses are encouraged to directly use AI tools, analyze results, reflect, and adjust their own strategies.

In digital transformation accelerator programs in the Northern mountainous region of Vietnam, small homestay owners applied AI tools to design their own marketing campaigns. Export-oriented SMEs used data analysis to adjust product prices and find new markets. Although at first they still made mistakes—such as not providing non-sensitive data to the tools—it was precisely these experiences that helped them understand that openness and transparency are conditions for AI to maximize effectiveness.

We emphasize that digital transformation is a process of nurturing lifelong learning capacity for business owners. The goal is not only short-term productivity, but also forming the habit of self-learning, experimenting, and continuously adapting—qualities that are essential for strategic innovation in an uncertain environment.

Theoretical and Practical Implications

• Resource-Based View (RBV): AI-supported self-learning helps businesses reconfigure resources into competitive advantages.

• Dynamic Capabilities Framework (DCF): Continuous self-learning strengthens the ability to “sense, seize, and transform” which is necessary for strategic flexibility (Teece, 2018).

• Scaling implications: When businesses build internal learning capacity with AI, scaling becomes more efficient, reducing marginal costs and improving operational performance.

International data matches KisStartup’s observations: SMEs applying AI not only improve productivity but also shift their strategic mindset from reactive to proactive. Therefore, digital transformation is not only a technological change but also a transformation of awareness and organization.

Conclusion

Digital transformation, especially with AI, should be understood as a catalyst for the self-learning capacity and strategic innovation of entrepreneurs. Evidence shows that AI increases productivity, reduces costs, and creates personalized learning environments. The greatest value lies in business owners being able to self-learn, self-reflect, and shape their own strategies.

KisStartup’s approach emphasizes this factor: equipping businesses with the ability to explore and ask questions, so that digital transformation becomes the path toward adaptability and long-term competitiveness. In the context of globalization, successful businesses are not necessarily those with the most advanced technology, but those whose leaders know how to learn and continuously adapt.

 

References

• AMIS. (2024). AI-based learning progress analysis. Retrieved from https://amis.misa.vn/245267/phan-tich-tien-do-hoc-tap-bang-ai/
• Gartner. (2023). AI adoption and cost optimization in SMEs. Gartner Research Brief.
• IBM. (2023). AI adoption survey of 7,500 enterprises worldwide. IBM Global AI Index.
• Louis. (2024). AI application in work: saving 5.4% of working time. CloudGo. Retrieved from https://cloudgo.vn/ung-dung-ai-trong-cong-viec
• McKinsey & Company. (2023). The State of AI in 2023. McKinsey Global Institute.
• Nhân Dân. (2024). Improving business management efficiency with AI and smart data. Retrieved from https://nhandan.vn/nang-cao-hieu-suat-quan-tri-doanh-nghiep-voi-ai-va-du-lieu-thong-minh-post887975.html
• SkillsBridge. (2023). AI boosting efficiency: Performance optimization strategies. Retrieved from https://www.skillsbridge.vn/blogs/ai-nang-cao-hieu-suat/chien-thuat-toi-uu-hieu-suat
• Teece, D. J. (2018). Business models and dynamic capabilities. Long Range Planning, 51(1), 40–49.

• For an average text command:
  • Gemini (Google): approximately 0.24 Wh of electricity, emits 0.03 gCO₂, uses 0.26 ml of water.
  • ChatGPT (GPT-4o): estimated at around 0.3 Wh of electricity.
• These numbers may seem small, but for 1 million commands → approximately 300 kWh, which is the electricity consumption of a household in one month.
• Additionally, each 0.3 Wh of electricity could be equivalent to 0.03–0.21 gCO₂ depending on the "cleanliness" of the energy source.

• Vague prompt → AI gives incorrect responses → need to run again.
• No length limit → AI generates unnecessarily long text → consumes tokens, uses more electricity.
• Choosing an overly powerful model for a simple task → like using a truck to carry a bag of vegetables.

1. Clear goal: Specify exactly what you need, for whom, and in what format.
2. Provide sufficient context: Give data, conditions, and constraints upfront.
3. Limit output: Request specific word count or number of bullet points.
4. Choose the right model: Simple tasks → small models. Complex tasks → large models.
5. Avoid multimedia waste: Only ask for images/slides when absolutely necessary.
6. Save good prompts: Reuse them, don’t "reinvent the wheel."