Most SMEs Waste Digital Spend—Agritech Shows Why

Rice prices spiked 47% after climate shocks and geopolitical instability, yet Thailand’s rice yields are stuck at roughly 3.1 tons per hectare—far behind Vietnam’s 5.8. That gap isn’t just an agriculture story. It’s a textbook example of what happens when you pour money into “digital transformation” without designing for the people who must use it.

In Thailand, farmers like Somchai Thanakit get flooded with alerts from multiple apps, soil sensors, and satellite dashboards—and still see flat yields. For Singapore startups and SMEs trying to scale across ASEAN, the parallel is uncomfortable: more tools, more dashboards, more campaigns… and not much more revenue.

This post is part of our Singapore Startup Marketing series, focused on how Singapore companies market products regionally. Here’s the stance I’ll defend: Most digital marketing “failure” isn’t about the channel. It’s about misalignment—between what you built and how customers actually behave.

The real problem isn’t tech. It’s misalignment.

If you take one lesson from Asia’s agritech push, let it be this: Digital adoption doesn’t fail because users are “behind.” It fails because solutions are built for a fantasy user.

Thailand’s Smart Farmer efforts show the pattern. Since 2010, ASEAN countries have invested about US$180B in agricultural modernisation. Thailand alone channelled roughly US$12B since 2017 into programmes aligned with “Thailand 4.0”. Yet agricultural productivity growth (TFP) crawled at around 0.8% per year (2015–2023)—well below what’s needed for long-term food security.

The sharpest pain hits smallholders: they represent 80% of Asia’s ~200M farm households but receive <15% of agritech investment. When tools are designed by urban teams for rural realities they don’t live, the outcome is predictable: low sustained usage.

For SMEs, substitute “farmers” with “customers” and “Smart Farmer platform” with “CRM + ads + marketing automation + AI copy tool.” Same movie.

The SME version of “endless digital alerts”

Farmers complain about conflicting app notifications. SME teams have their own version:

• Meta Ads says one thing; Google Ads says another.
• Your agency reports “impressions and clicks”; finance asks for profit.
• Your CRM shows leads; sales says they’re junk.
• WhatsApp enquiries look strong; attribution says “direct.”

Information overload is not insight. It’s just noise at a higher resolution.

When investment doesn’t produce outcomes, measurement is usually the culprit

Thailand’s Smart Farmer programme reportedly registered 1.2M farmers, but active usage stayed below 25%. That’s not a marketing problem; it’s a product and measurement problem. Registration is a vanity metric.

Singapore SMEs fall into the same trap when they celebrate:

• followers
• reach
• website sessions
• “leads” that never pick up

If you want your digital marketing to pay you back, you need a measurement spine that ties spend to business outcomes.

What “outcomes” should mean for Singapore SMEs

Outcomes should be boring and financial. Pick a small set:

1. Contribution margin (or gross profit) per order
2. Cost per qualified lead (CPQL) or cost per sale
3. Sales cycle velocity (days from enquiry to close)
4. Retention / repeat rate (for e-commerce and subscriptions)

Then work backward to the channels.

A practical stance I’ve found works: If you can’t explain your funnel in one whiteboard, you don’t have a funnel—you have a list of tactics.

A simple “efficiency gap” test (steal this)

Vietnam’s yields outperform Thailand’s because tools and practices translate into on-field decisions. SMEs can run an equivalent test:

• Step 1: List your top 5 campaigns.
• Step 2: For each campaign, write the single decision it is meant to influence (e.g., “book a demo,” “request a quote,” “add to cart”).
• Step 3: Check whether the campaign reporting proves that decision happened.

If it doesn’t, you’re funding activity, not performance.

Design for real behaviour, not the user in your pitch deck

The agritech article calls the core issue “anthropological,” and that word matters. Tools built without field understanding become “digital Ferraris” where users need “reliable bicycles.”

In Singapore startup marketing, I see the same mismatch when companies expand to the region:

• They write one set of English ads and expect Malaysia/Indonesia/Thailand to respond the same way.
• They force self-serve checkout when the market expects WhatsApp-first buying.
• They push long webinars when audiences want short demos and price clarity.

The fix is not “localise your copy.” The fix is design around local buying habits.

Three farmer-centred design principles that map cleanly to SME marketing

1. Reduce choices; increase clarity

   • Farmers don’t need five weather apps.
   • Your customers don’t need 12 landing page sections.
   • One offer. One CTA. One next step.

2. Make the recommendation actionable

   • “Humidity is 78%” is data.
   • “Delay spraying by 24 hours” is a decision.
   • In marketing terms: “Our software has AI features” is fluff; “Cut invoice processing from 2 days to 2 hours” is a decision trigger.

3. Meet users where they already are

   • Agritech works better when integrated into cooperatives and extension services.
   • Marketing works better when integrated into existing workflows: WhatsApp, phone calls, distributor networks, retail staff scripts.

The hidden risk: “algorithmic dependency” is real in marketing too

In Thailand, flooding knocked out connectivity and farmers lost access to cloud-based schedules and irrigation controls. Researchers have also observed that long-term reliance on automated systems can reduce manual skill—what economists call algorithmic dependency.

SME teams face an equivalent risk: outsourcing thinking to platforms.

If your marketing strategy is “do what the ad platform suggests,” you’re building a business on incentives you don’t control:

• Platforms optimise for spend and engagement, not your profit.
• Automated bidding can chase cheap clicks that never convert.
• AI-generated creatives can converge into the same bland sameness, driving CPC up over time.

A snippet-worthy rule: Automation is a great servant and a terrible strategist.

Guardrails that keep you in control

• Keep a human-owned positioning document (who it’s for, why you win, proof).
• Maintain a monthly channel scorecard that includes profit-linked metrics.
• Run incrementality tests (geo split, holdout, or time-boxed experiments) at least quarterly.
• Track lead quality with a simple sales feedback loop (e.g., 1–3 rating in the CRM).

What better looks like: fewer platforms, tighter loops, local context

The article points to alternatives like voice-based solutions in local dialects and systems that produce simple recommendations rather than complex dashboards.

For SMEs, “better” usually looks like:

1) One source of truth for leads

If you’re running ads, SEO, and social but enquiries land in five inboxes, you will never fix lead quality. Route everything into one place (CRM or even a structured spreadsheet) with:

• source
• campaign
• first response time
• outcome (won/lost)
• loss reason

This is unglamorous. It’s also where ROAS is born.

2) Regional expansion that respects buying reality

In the Singapore Startup Marketing context, regional growth is often blocked by one assumption: “If it works in Singapore, it’ll work in Bangkok/Jakarta.”

What works instead:

• Use market-specific landing pages (not just translated pages).
• Offer WhatsApp-first flows where appropriate.
• Build local proof fast: 3–5 case studies per market beats global logos nobody recognises.

3) Demand-pull, not tool-push

Agritech fails when it’s technology-push. Marketing fails the same way when you start with tactics:

• “We need TikTok.”
• “We should do influencer marketing.”
• “Let’s run Performance Max.”

Start with the constraint:

• Not enough qualified leads?
• Low conversion from lead to close?
• Long sales cycle?
• Weak repeat purchases?

Then choose the simplest intervention.

A practical checklist for SMEs before you buy another tool

If you’re a Singapore SME planning your 2026 marketing budget, run this quick pre-mortem:

1. Who is the user? (Not “SMEs.” Be specific: role, urgency, objections.)
2. What decision are we trying to change this month?
3. What’s the minimum proof we’ll accept? (e.g., 20 SQLs, 10 paid trials, S$X in gross profit.)
4. What’s the fastest feedback loop? (Sales calls, WhatsApp logs, on-page behaviour.)
5. What gets cut if we don’t hit the proof?

If you can’t answer #5, you’re not running performance marketing—you’re running hope marketing.

The point isn’t to be “more digital.” It’s to be more useful.

Asia’s US$180B agritech push is a warning sign: digital transformation without user-centred design produces impressive demos and disappointing outcomes. Thailand’s experience—high investment, low productivity gains, low active usage—mirrors what happens when SMEs stack tools instead of building tight, local, behaviour-aware funnels.

The teams that win in ASEAN expansion don’t use the most software. They do three things consistently: they pick a clear customer, they design for real behaviour, and they measure what pays the bills.

If your marketing is generating lots of activity but little revenue, the next move probably isn’t a new platform. It’s a better understanding of the person on the other side of the screen.