Figma Spits Out Code, OCR Becomes Free, and E-Commerce Launches in 60 Seconds — A Breakdown of the Crumbling Costs of Outsourced Work
Related Articles
Conclusion First: The “Market Rate” for Outsourcing Has Collapsed
Design outsourcing: 150,000 yen per month. Video production: 200,000 yen per video. OCR data entry: 50,000 yen per month.
This was the “norm” until last year. For small and medium-sized enterprises in rural areas, these outsourcing costs weighed heavily as fixed expenses.
However, by 2025, this market rate has completely collapsed. Figma has released features that allow for seamless design to code generation, Baidu has effectively made OCR free, and AI video generation tools can produce promotional videos in just a few minutes.
We have entered an era where “outsourced work” can be done for less than 50,000 yen per month — in some cases, almost for free.
In this article, we will organize what has changed in terms of specific costs. Business owners who think “this doesn’t concern us” should definitely read on.
—
Figma: “Code Comes Out Without Designers Having to Write It”
The essence of the new features announced by Figma in 2025 is that “design tools have become development environments.”
The previous flow was as follows:
- Designers create designs in Figma.
- The design is handed over to coders.
- Coders convert it into HTML/CSS.
- Every time a revision is needed, back-and-forth communication occurs between designers and coders.
This process of “2→3→4” has quietly consumed costs for small and medium-sized enterprises. Just creating a single landing page (LP) could cost between 100,000 to 150,000 yen. If there are two revisions, an additional 30,000 to 50,000 yen would be incurred. When considered annually, this could amount to hundreds of thousands of yen.
Figma’s new feature, “Coding Layer,” flips this structure on its head. It allows direct editing of code on the design canvas, and AI automatically generates code from designs. Additionally, AI-generated animations and motion graphics have also been included.
In other words, by the time the design is created, a “working product” is nearly complete.
What specifically changes?
- Outsourcing cost for LP production: 150,000 yen → Approximately 2,250 yen per month (Figma Pro) + only in-house labor
- Communication for design revisions: 3-5 days of back-and-forth → Immediate reflection on the spot
- Orders to coders: Increasingly unnecessary
The common belief that “design and coding are separate jobs” is quietly coming to an end.
Of course, developing complex web applications cannot be done solely with this. However, for the LPs, recruitment pages, and campaign sites that small and medium-sized enterprises need on a daily basis, there is almost no reason left to outsource.
I want to ask: How much does your company pay for a single LP? Is that amount really still “reasonable”?
—
Baidu Unlimited OCR: “Mountains of Paper” Turn into Data Instantly, and It’s Free
The impact of Baidu’s unlimited OCR is not about “improved accuracy.” It is a price disruption that is “free and unlimited.”
Let’s look back at the market rates for conventional OCR services:
- Google Cloud Vision API: Approximately 200 yen for every 1,000 requests. If processing 10,000 images per month, it costs 2,000 yen; for 100,000 images, it costs 20,000 yen.
- Domestic OCR services: Monthly fees range from 30,000 to 100,000 yen. This is burdensome for small and medium-sized enterprises.
- Manual data entry outsourcing: 50 to 100 yen per image. For 1,000 images per month, it costs 50,000 to 100,000 yen.
Baidu’s OCR offers this service for “free and unlimited.” It also supports the analysis of long documents and has significantly improved processing speed compared to conventional OCR.
Many small and medium-sized enterprises in rural areas still rely on manual data entry for paper invoices and order forms. In industries like construction, manufacturing, and wholesale, where paper culture is deeply rooted, the impact is substantial.
Tasks that previously cost 50,000 yen per month for part-time workers to input data could potentially become zero.
However, there are precautions. Baidu is a Chinese company. It is necessary to assess data handling policies against your own information security standards. There are risks in directly processing highly confidential documents.
A practical approach would be:
- Low confidentiality documents (catalogs, brochures, public information) → Batch processing with Baidu’s OCR
- High confidentiality documents (contracts, personal information) → Use local OCR tools or Google Cloud Vision in conjunction
By simply differentiating these uses, monthly data entry costs can be reduced to less than half. There is no need to rely on a single tool. “Combining tools appropriately” is a smart strategy for small and medium-sized enterprises.
—
AI Video Production: From 200,000 Yen to Under 2,000 Yen per Video
This is where the numerical impact is the greatest.
Let’s break down the conventional video production costs:
- Planning and scripting: 30,000 to 50,000 yen
- Filming: 50,000 to 100,000 yen
- Editing and narration: 50,000 to 100,000 yen
- Total: 150,000 to 250,000 yen per video
Even if a small or medium-sized enterprise in a rural area wants to engage in “video marketing,” producing just one video per month would cost 200,000 yen. That amounts to 2.4 million yen annually. This is unaffordable. As a result, there was an unspoken understanding that “video is for large companies.”
By 2025, this structure has completely collapsed.
Using current AI video generation tools (such as Runway Gen-3, Pika, Kling, Sora, etc.), promotional videos can be generated from text prompts in just a few minutes. With a subscription fee of 2,000 to 5,000 yen per month, several dozen videos can be produced monthly.
When converted to cost per video, it amounts to a few hundred yen to 2,000 yen. This is 1/100 of the previous cost.
Of course, the quality cannot be compared to that of television commercials. However, for social media ads, YouTube shorts, or company introduction videos on recruitment pages, the quality is more than sufficient.
Consider this:
If a small or medium-sized enterprise in a rural area could post ten videos on social media each month, what would happen? They could stand on the same “playing field” as the marketing departments of large companies. And the cost would be under 5,000 yen per month. This is not about “beating large companies”; it is about “standing on the same ground.” For small and medium-sized enterprises, simply being able to stand on the same ground is revolutionary.
—
E-Commerce Site Construction: The Era Where Stores Open in 60 Seconds
It has been a few years since Shopify and BASE have been touted for their ability to “easily create e-commerce sites.” However, the AI integration in 2025 takes it to a whole new level.
Using Shopify’s AI assistant “Sidekick” or BASE’s AI features, you can upload product photos and answer a few questions, and it will automatically generate product descriptions, suggest pricing, and apply design templates. Literally, an e-commerce site can be set up in 60 seconds.
Previously, outsourcing the construction of an e-commerce site would cost between 300,000 to 1 million yen. Monthly operational costs would also range from 30,000 to 50,000 yen. Now, this can be done with just a few thousand yen in platform usage fees.
When manufacturers or farmers in rural areas want to “sell directly,” the “cost of building an e-commerce site” is no longer a barrier.
—
Breakdown of Outsourcing Cost Collapse
| Service | Previous Outsourcing Cost (Monthly) | 2025 Cost (Monthly) | Reduction Rate | Main Tools |
|---|---|---|---|---|
| LP/Web Page Production | 100,000 – 150,000 yen | 2,250 yen – (Figma Pro) | Approximately 95% | Figma + AI Code Generation |
| Data Entry/OCR | 50,000 – 100,000 yen | 0 – a few thousand yen | Approximately 90 – 100% | Baidu OCR, Google Vision |
| Video Production (for SNS) | 150,000 – 250,000 yen/video | 2,000 – 5,000 yen/month | Approximately 97% | Runway, Pika, Kling |
| E-Commerce Site Construction | 300,000 – 1,000,000 yen (initial) | 3,000 – 5,000 yen/month | Approximately 95% | Shopify, BASE |
| Translation/Multilingual Support | 10 – 15 yen per character | Almost 0 yen | Approximately 99% | DeepL, Claude, GPT |
Looking at this table, can you still say, “We can still rely on outsourcing”?
—
So, What Should We Do?
“I understand it’s amazing, but our employees aren’t IT-savvy” — this is the most common reaction. And it’s the most dangerous one.
The commonality among these tools is that they are designed to be used “without specialized knowledge.” Figma’s Coding Layer was created for those who cannot write code. AI video tools were made for people without experience in video production. OCR is as simple as pressing a button.
It’s not that “you can’t use it because you’re not IT-savvy”; it’s that “you can’t use it because you haven’t tried it.”
Here are three specific actions to take:
1. Try Just One First
Don’t try to implement everything at once. Choose the area where your company incurs the most outsourcing costs and try the corresponding tool for just one week. Most have free plans available.
2. Take Inventory of Your “Outsourcing Costs”
List all outsourcing costs from the past 12 months. Design, printing, video, data entry, translation, web production — the total will likely be a staggering amount. Mark those that can be “replaced by AI tools.”
3. Decide in Advance How to Use the Money Saved
Cost reduction is a means, not an end. If you save 100,000 yen per month, what will you do with that money? Develop new products, raise employee salaries, or invest in advertising? Deciding in advance how to use the “saved money” will change your motivation for implementation.
—
What’s Truly Scary Is Not the “Decrease in Costs”
Finally, let’s discuss the structural aspect.
The fact that outsourcing costs have decreased means that they have also decreased for your competitors. Your peers in the neighboring town can use the same tools for 5,000 yen per month.
In other words, this is not a story of “gaining an advantage by using it” but rather “being left behind if you don’t use it.”
Design, video, e-commerce, data entry — what was once something “only companies that could afford it could do” has transformed into something “anyone can do.” The moment this change occurs, the rules of competition change.
The era where “spending money on outsourcing” provided a competitive advantage is over. From now on, the competition will be about “what to create” and “who to deliver it to.”
And for small and medium-sized enterprises in rural areas, this is actually an opportunity. They can use the same tools without the abundant budgets of large companies. Being familiar with the field allows them to create content that resonates. Their quick decision-making enables them to act first.
The only question is whether they will act.
I encourage you to try just one thing today while reviewing your outsourcing invoices.
JA
EN