Conclusion First: The SaaS You’re Paying Monthly for is Actually Making Money Off Your Data Twice

Atlassian has significantly expanded its data collection policy in 2025. The terms have been revised to allow extensive use of user behavior data accumulated through tools like Jira, Confluence, and Trello—data on who edits what, when, and how workflows are managed—as training data for AI functionalities.

Tinder has introduced iris scanning for identity verification, while Zoom has announced enhanced security through biometric authentication.

What these three companies have in common is the fact that “they are seeking deeper layers of data from users.” Behavioral logs, biometric information, and business processes. The boundaries of data collection, previously considered necessary for service provision, are clearly shifting.

So, here’s the question.

Does your company know how much its data is worth?

The True Nature of the Data You’re Giving to a “1,500 Yen Monthly Tool”

Many small businesses use SaaS as a “cheap and convenient tool.” Atlassian’s standard plan costs about 1,200 to 1,500 yen per user per month. For a team of ten, that’s 15,000 yen a month. Affordable.

But have you ever thought about what’s happening behind the scenes?

If a team of ten manages a project with Jira for a year, thousands of task data, comments, and workflow transition logs accumulate. This is real training data that reflects “how a company of a certain industry and size operates.”

Atlassian uses this data to enhance its AI feature, “Atlassian Intelligence,” which is sold at a higher monthly rate as a premium offering.

Here’s the structure laid out:

1. You pay a monthly fee to use the tool.
2. Your business data is used for AI training.
3. The AI-equipped premium plan is sold at a higher price.
4. You pay additional fees for the “AI that has become smarter with your data.”

The 1,500 yen monthly fee is for “tool usage” and does not include “data provision fees.” In other words, your data is being given away for free.

Biometric Data is Even More Serious — What Tinder and Zoom’s Iris Scanning Means

The official reason Tinder introduced iris scanning is “to prevent impersonation.” Zoom’s biometric authentication is also touted as a means to “enhance meeting security.” Both sound like they are for the benefit of users at first glance.

However, biometric data is fundamentally different from behavioral logs. Passwords can be changed. Email addresses can be altered. But the iris cannot be changed in a lifetime.

Once leaked, it’s irreversible. And where this data is stored, who can access it, and how long it is retained can only be understood by reading the fine print of the terms.

How many small business owners have read the terms of service for the SaaS they use all the way through? Honestly, probably very few.

Here’s a number you should be aware of. According to IBM’s “Cost of a Data Breach Report 2024,” the average cost per data breach is $4.88 million (approximately 730 million yen). While this is a global average that includes large corporations, it’s not uncommon for small businesses to incur damages in the millions to tens of millions of yen. If biometric data is leaked, the risk of lawsuits skyrockets.

You should calmly calculate the cost of accepting biometric authentication just because it’s “free” or “convenient.”

Let’s Calculate the Price of Data

So, what is the specific “price of data”? Let’s roughly estimate it.

The market price for datasets used for AI training varies greatly depending on quality and quantity. However, structured data on business processes—real logs of “how a company in a certain industry operates”—falls into a fairly high category.

• General text data: a few hundred to a few thousand yen per 10,000 records
• Industry-specific structured data: tens of thousands to hundreds of thousands of yen per 10,000 records
• Biometric data (iris, facial recognition, etc.): a few hundred to a few thousand yen per record (for academic and security purposes)

Assuming a team of ten uses Jira for a year and generates 5,000 task data and workflow logs, conservatively estimating as industry-specific structured data, the market value of that data is in the tens of thousands of yen.

On the other hand, your annual usage fee is about 180,000 yen (1,500 yen × 10 people × 12 months).

In other words, you may be providing data worth tens of thousands of yen for free, equal to your tool usage fee.

This isn’t just a simple story of “losing out.” The problem is that you are entering into this transaction structure without recognizing it.

Three Things Small Businesses Should Do Right Now

Vague suggestions like “let’s create a data strategy” won’t change anything. Let’s narrow it down to three actionable steps you can take starting tomorrow.

1. Read the “Data Use” Section of the Terms for the SaaS You’re Using

You don’t need to read the entire document. Search within the document for “Data,” “AI,” “Machine Learning,” and “Training.” Confirm whether your company’s data will be used for AI training and whether you can opt out of it. This should take less than 30 minutes.

In the case of Atlassian, you can opt out of AI data use from the admin settings. However, the default setting is opt-in (you have agreed). In other words, if you do nothing, your data will be used.

2. Estimate the Cost of “Alternative Tools” Just Once

If you operate an open-source project management tool (e.g., Plane, Taiga) on your own server instead of Atlassian, your data will be completely under your control. The monthly cost may only be server fees, which can be as low as 2,000 to 5,000 yen for about five users.

You might think, “Managing it in-house is a hassle.” However, if you set it up on a cloud VPS with Docker, the configuration can be completed in half a day. In this era, even non-engineers can do it by asking AI for setup instructions.

Choosing whether to keep data in-house or entrust it to others. It’s crucial to make this decision not “just because,” but by “calculating” it.

3. Use the Timing of Contract Renewals to Negotiate the “Data Clause”

When renewing contracts with SaaS vendors, negotiating “to opt out of AI data use in exchange for keeping the price the same” can actually be feasible. This is because data has clear value for the vendor.

Large corporations are already negotiating this. In contracts with thousands to tens of thousands of users, customizing data clauses is not uncommon. Small businesses often think, “We’re too small to negotiate,” but if ten companies band together, the conversation changes. Joint negotiations through industry associations or local business communities is also an option.

Fundamental Question — What Happens to Companies Without Data in the Age of AI?

The structure of what is happening now is simple.

Large platform providers collect user data to strengthen their AI and profit from selling that AI. Those who provide data become increasingly dependent as the AI becomes smarter, and the switching costs rise.

Within this structure, small businesses have two choices.

A. Provide data in exchange for maximizing the benefits of AI (consciously)
B. Keep data in-house and create a system to utilize AI internally

Which is the right choice depends on the industry and scale. However, the most dangerous state is being in a position where you are unconsciously choosing A.

For local small businesses, operational data is the very essence of “how we do our work.” Business processes refined through years of trial and error, customer service know-how, and region-specific business customs. All of this data can be siphoned off, become training material for AI, and be offered to competitors as the same AI functionality.

Your “strengths” are being homogenized through the platform. This is the true cost of giving away data for free.

Conclusion: There’s No Such Thing as Free Data

The policy changes at Atlassian and the introduction of biometric authentication at Tinder and Zoom indicate a single trend: “The scope of data collection will continue to expand in the future.”

What small business owners should do is not chase the cutting edge of technology. They need to understand how much value is placed on the data they generate daily. And confirm whether they are receiving compensation that matches that value.

Are you giving away data worth tens of thousands of yen for a tool that costs 1,500 yen a month?

Today, I urge you to open the admin panel and check the opt-out settings for AI training. It will take less than 30 minutes. That half an hour could be the first step in protecting your company’s data assets.