Beyond the "2025 Digital Cliff": Where Legacy Modernization Stands and the Next Move

In September 2018, Japan's Ministry of Economy, Trade and Industry (METI) introduced the phrase "the 2025 Digital Cliff" in its DX Report. The warning: if companies failed to modernize their aging core systems, Japan could face economic losses of up to 12 trillion yen per year from 2025 onward. Now that 2025 has passed, where do we actually stand?

What the "2025 Cliff" Really Was

The "2025 Cliff" flagged the risk of leaving complex, black-boxed legacy systems untouched. METI based the warning on three key projections:

A shortage of roughly 430,000 IT workers by 2025
An estimated 60% of large-enterprise core systems running for 21 years or more
A warning of up to 12 trillion yen in annual economic loss if left unaddressed

Old systems impose a triple burden: soaring maintenance costs, a vanishing pool of engineers who understand them, and missed business opportunities. This was never just an IT-department issue — it was a matter of corporate survival.

Where We Stand in 2026: Was the Cliff Crossed?

In May 2025, METI and the IPA published the "Legacy System Modernization Committee Summary Report." Its figures show both progress and stagnation.

About 61% of companies still run legacy systems. Among large enterprises, roughly 74% remain dependent on outdated technology. That is down from about 80% in 2018 — but the pace of renewal remains slow.

In other words, the cliff has not been fully crossed. And the talent problem is getting worse: METI now projects a shortage of up to 790,000 IT workers by 2030. As veteran engineers who understand legacy systems age out and retire, knowledge becomes "person-dependent" and the technical information needed for upgrades is quietly being lost.

The "3 + 2" Barriers to Modernization

The summary report groups the obstacles into three technical barriers and two business barriers.

Three technical barriers

Obsolescence: outdated tech stacks that are hard to maintain
Bloat and complexity: systems grown massive through repeated patching
Black-boxing: missing documentation, so no one grasps the whole picture

Two business barriers

Change-averse culture: organizational inertia favoring the status quo
Underinvestment in IT: treating modernization as a cost rather than strategic investment

The report also names a structural cause: in Japan, only about 30% of engineers work inside user companies (versus roughly 70% in the U.S.). Lacking in-house technical capability robs firms of ownership over their own modernization.

Five Modernization Approaches — and How to Choose

"Modernization" covers very different paths. Here are the five common approaches, framed by goal, timeline, and risk.

Approach	What it does	Best when
Rehost	Move the same program to a new platform (e.g. cloud) without changing the code	You want quick cost savings and fast wins
Refactor	Keep the language and features, but improve the internal code structure	You want better maintainability without a full rebuild
Rewrite	Rebuild the code from the ground up in a modern language	You need a step-change in performance, scalability, maintainability
Replace	Swap the system for a new product or SaaS	Standard features are good enough for that domain
Redocument	Rebuild documentation to dissolve the black box	You need to understand the current state first

The key is not "everything at once," but combining approaches by business importance: push standard processes to SaaS, and carefully rewrite the systems that are sources of competitive advantage.

METI's Four Recommended Measures

The report distills company action into four moves.

Shift leadership mindset: commit budget and fold DX into long-term business planning
Make systems visible: inventory IT assets and set modernization priorities
Standardize: separate standard functions from differentiating ones
Secure IT talent: grow in-house engineers and bring capability back inside

The Next Step for Japanese and SME Companies

More important than whether the cliff was "crossed" is what you do tomorrow. Companies that succeed at modernization almost always start small.

Start with an inventory: list every system and identify the ones that would halt the business if they failed
Prioritize: use an impact-versus-age matrix to decide what to tackle first
Match the method: don't treat every system the same — mix rehost and rewrite as needed
Move with a partner: where in-house talent is short, co-create with a trusted development partner

In the age of AI, legacy systems are becoming the single biggest barrier to "connecting your own data with AI." Modernization is both paying down past debt and investing to unlock future growth. The cliff was never the finish line — it was the start of a new climb.

Accelerate your DX with Be A Racer

From cloud migration and AI adoption to full-stack development — we deliver the fastest digital transformation, end to end. Let's talk.

Book a free consultation