Scattered shapes move and separate to form a personalised opportunity map.
Startups and Capitalthrough Where the Cost Fell Away

Turn Programme Reshuffles Into a Personal Opportunity Map

Free services and a paid service priced at A$49 a month already exist abroad. A service could turn programme reshuffles into personalised change alerts, alternative-support recommendations, and duplicate checks for public agencies.

Published 2026. 9. 1.

Nearly half of 472 reviewed programmes are set to change

On August 27, 2026, the Korean government said it would reorganise small-business support programmes across ministries and reflect the changes in its 2027 budget proposal. It is not ending programmes funded by the 2026 budget immediately; it is planning to change their structure and budgets for the following year.

Korea currently has 671 small-business support programmes across 22 ministries, with a budget of about KRW 31 trillion. The government selected 472 programmes across 17 ministries, with KRW 25.5 trillion in budgets, for review.

It plans to adjust 232 programmes. It will merge or end 88, and reduce the budgets of 144, saving about KRW 4 trillion. That is close to one sixth of the budget under review.

For pre-startup support, it plans to combine Challenge K-Startup (도전 K-스타트업), the Pre-Startup Package programme (예비창업패키지), and others into the All Startup Support programme (모두의 창업 지원사업), reducing KRW 76.4 billion including operating costs. It will also combine two policy-finance programmes that had been used heavily by businesses more than 10 years old, reduce them by KRW 267.7 billion, and allocate more to innovation startups.

The savings are not money removed from support altogether. They are funds to be moved elsewhere. The government plans to select 300 businesses each year in new security, pharmaceutical and biotech, climate tech, and K-consumer goods, then provide combined support for research and development, commercialisation, finance, exports, and talent for up to five years. Its 2027 budget proposal includes KRW 238.8 billion for this.

The job is no longer reading notices. It is explaining what changed.

Outside Korea, services now offer personalised recommendations and application-draft features for free or for around KRW 50,000 a month.

Consider the owner of a cosmetics manufacturer in Hwaseong with 12 employees. They search Bizinfo (기업마당), the Korean government’s business-support information portal, and multiple agency websites for export, packaging improvement, hiring, and research-and-development calls. They also save links sent by people they know.

For each call, they copy the application period, region, business-age and revenue requirements, support details, and required documents into an internal work document. They open last year’s application beside this year’s notice to see whether they can reuse the same documents. When the wording is unclear, they call the responsible agency.

When a programme changes its name, saved links and search terms stop working. It becomes difficult to tell whether a merged programme continues the earlier support or has also changed the eligibility rules. People either check again just before the deadline or miss the opportunity.

In a new service, a business enters its region, years in operation, employee count, products, export status, and names of programmes it applied to in the past just once. When an official notice appears, the service compares it with the previous year and labels it as “operating unchanged,” “renamed,” “merged into another programme,” “alternative candidate available,” or “detailed notice pending.”

The alert should not send only a conclusion. It should show the changed sentence—for example, that a limit moved from businesses seven years old or younger to three years old or younger—and link directly to the original notice. Bizinfo provides an official connection that lets other services retrieve notices automatically, so people do not need to re-enter every notice manually from the start.

Some work still belongs to people. They need to judge whether an industry classification fits the actual product, prepare growth plans and evidence of results, and check unclear requirements with the responsible agency. The service is more trustworthy when it does not guarantee selection, but instead reduces uncertainty about what changed and what users need to verify.

There is no need to cover all 671 programmes on day one. Start with one group, such as pre-startups, cosmetics exports, or regional manufacturers, and connect roughly 20 frequently searched programmes. The first goal can be simple: when someone enters a past programme name, help them find its current status and the supporting notice within three minutes.

Services abroad have moved beyond search

The UK government’s Find a Grant is a free service that lets users search central-government grants in one place, save search criteria, and receive new calls by email. During its 2023 pilot, it listed more than 200 programmes and notices worth about KRW 9.9 trillion, and had 93,000 users.

The UK government said the pilot reduced applicants’ application time by 34% and civil servants’ administration time by 72%. By February 2024, it had more than 200,000 registered users, and the grants listed at that time exceeded about KRW 14 trillion. That figure was the value of published notices, not money paid out by the service.

Grantible in Australia helps small businesses, startups, and non-profits enter their details, receive matched support recommendations, and create application drafts. Users review the final content and submit directly through the government’s official application channel.

The service says it tracks 3,336 Australian support programmes. Basic recommendations are free, while its paid version costs A$49 a month, or about KRW 48,000 at the August 2026 exchange rate. This shows how personalised discovery and draft creation can be broken into a smaller purchase than a specialist contract.

Spain’s government began not with company search, but with duplicate checks for public agencies. It built a pilot tool using 1,050,470 national subsidy records to identify signals such as repeated awards, cumulative amounts, and sanction histories, narrowing the cases officials should review first.

In the pilot data, overall classification accuracy was 95%, the rate of identifying actual risk cases was 93%, and 42,152 cases were classified as high risk. The Organisation for Economic Co-operation and Development (OECD) describes this as a stage for testing feasibility. A high-risk flag should therefore not be treated as a confirmed improper claim; the final decision remains with officials investigating the evidence.

Four things to build now

1. Disappeared programme tracker

  • What it does: A user enters the name of a programme they applied to last year. The service shows whether it continues, was renamed, merged, is planned for closure, or has an alternative candidate, along with the supporting notice.
  • Who uses it: Prospective entrepreneurs who prepared for the Pre-Startup Package programme or applied for startup support at least once in the past two years.
  • Why now: From 2027, 232 programmes will be adjusted, creating an information gap for people who saved only past programme names.
  • First screen: A search field for a past programme name, plus three cards: “current status,” “changed conditions,” and “next date to check.”

2. New growth-sector readiness check

  • What it does: It takes a company’s products and current track record, then identifies relevant areas among new security, pharmaceutical and biotech, climate tech, and K-consumer goods, along with missing evidence.
  • Who uses it: Owners of cosmetics or food manufacturers with five to 20 employees who are preparing for overseas sales but do not yet have a dedicated support-programme employee.
  • Why now: The government has announced long-term support from 2027 for 300 businesses each year, with results assessed two years later.
  • First screen: Four sector-selection buttons and fields for “revenue, orders, certifications, and export targets to prove within two years.”

3. Application-material reuse vault

  • What it does: It compares past submitted documents with materials required by a new notice, separating what can be reused, updated, or created from scratch.
  • Who uses it: Business-support staff at regional manufacturers that review two or three or more policy-finance, export-support, and technology-development programmes each year.
  • Why now: Even when programmes are merged, core company materials are repeatedly submitted. Comparing documents and alerting users to expiry dates can reduce preparation time.
  • First screen: A chronological view of recently submitted documents, expiry dates, and documents required again by the current notice.

4. Similar-programme alert for public agencies

  • What it does: An agency enters a proposed support programme. The service flags candidates from other ministries and local governments that overlap in target users, purpose, and support method.
  • Who uses it: Staff at local governments and affiliated support agencies designing new subsidy programmes for regional businesses.
  • Why now: The government reviewed 472 programmes at once, and an overseas example used more than one million records to narrow down cases for review.
  • First screen: A field to paste a programme plan, alongside three similar candidates, overlapping sentences, and items a person should verify.

What to check today

Show one past support-programme notice to three people who applied for support within the last two years. Ask each for 10 minutes how they would find an alternative if its name changed. If two of the three cannot immediately explain the current path and say they would register their past programme name to receive change alerts, the disappeared programme tracker is worth building first.

Why this matters where you are

Support programmes can be difficult to follow whenever names, eligibility rules, or delivery structures change. Check whether people in your market rely on old links, saved search terms, and manually maintained document lists. If they do, begin with a narrow programme category and show users the evidence behind every change rather than only a recommendation.

Sources

7 sources

Every fact in this article came from the pages below. Check them yourself.