Hakaru Lab | Published: 2026-08-31 | Updated: 2026-09-03 | Review
Who this is for: Readers who have seen posts on social media like "Start overseas affiliate marketing with AI and earn 500,000 JPY a month in 30 days" and want to verify their authenticity.
Key points (3 lines)
・We verified the promotional claim of "earning 500,000 JPY a month in 30 days by starting overseas affiliate marketing with AI" against official information and public data, and found no independent data showing the reproducibility of achieving both that speed and amount. On the other hand, we found that the underlying mechanism of recurring affiliate commissions does exist.
・This can be used for the judgment of readers who have seen this kind of promotional claim on social media and want to verify its authenticity, or readers who are considering entering overseas affiliate marketing. This is a verification based on research of official information and public data, not hands-on verification by our lab itself.
・The conclusion is based on the scope of official information and public data confirmed as of August 31, 2026. It does not include hands-on verification by our lab itself, nor continuous research on search rankings.
Basis of this review
・Reviewed on: 2026-08-31 (Date of confirming public information. For the reference date of individual sources, refer to each link destination in the text)
・Conflicts of interest: There are no advertising or affiliate links in this article.
There are no advertising or affiliate links in this article.
Posts like "Exceeded a corporate employee's monthly salary in 30 days" or "Earned hundreds of thousands of JPY in the first month after starting overseas affiliate marketing with AI" regularly flow into social media timelines. At Hakaru Lab, we broke down the very methods often used in these posts and verified whether they actually hold true by cross-referencing official information (operators' terms and payment conditions), government announcements, and third-party public data (including self-reports by individuals and aggregations by businesses, each with varying degrees of certainty). To state the conclusion first, within the scope of the materials we checked this time, we could not find independent data showing the reproducibility of the combination of speed and amount such as "500,000 JPY a month in 30 days." While some of the parts that make up the method do exist, we found no evidence to support the overall claims combining them.
We avoid criticizing specific posts or accounts because it deviates from our lab's purpose of objectively verifying the mechanism. Instead, we introduce a "template" that appears to be common across multiple posts our lab has seen so far. Note that the parts labeled as "our lab's organization" in this article are current summaries based on the cases we confirmed, and are subjects for continuous investigation in the future with clarified numbers of cases and periods.
None of the individual elements are novel. However, by connecting them within a short period of "30 days," it creates an impression of reproducibility.
Our lab believes that the reason this template works lies not in the weakness of the readers, but in the structure of the advertising. It is designed to overlap the natural desire many people have, "I want to increase my income without increasing my time," with the expectation that "it might be possible with AI." And what is crucial is the time lag—the hurdle to start is low, but it takes months to verify if it was really effective, and following or purchasing happens before verification. To be honest, our lab does not yet know the root of where this time lag comes from (this is a subject for future verification).
Our lab verified using the following procedure.
The approach is to compare the claims of the promotional text by accumulating only numbers from public information with explicit sources, rather than relying on guesses or impressions.
Actual traffic of new accounts: A record of starting a new Pinterest account reports that over 30 days, 266 pins were posted, resulting in 11 clicks to external sites against 12,030 impressions (self-report by a single poster, poster's report). The conversion rate from impressions to clicks (CTR) is approximately 0.09% (11 ÷ 12,030). A record at the 90-day mark for the same type of account shows 21,969 impressions and 169 clicks; the stated rate on the source page is 0.84%, but recalculation by our lab yields 0.77%. We simply note the discrepancy between the stated figure and the recalculated value on the source page itself (90-day record).
Industry average conversion rate: A survey by Rewardful, an affiliate platform for SaaS, shows an average conversion rate of 0.8% and average commission of $14.10 across 2,847 target programs (Rewardful report). However, the report does not clarify whether these 2,847 programs form a common population for both metrics. Applying these numbers directly to an estimate, calculating a reward of $3,000 to $5,000 a month requires 26,596 to 44,326 affiliate visits per month (calculated by our lab from the report's numbers. This is a simplified assumption that does not consider recurring commissions). While traffic from Pinterest does not necessarily translate directly to SaaS contracts, when viewed as a comparison of the "scale of initial traffic" required, it is off by several orders of magnitude from the reported value of a new Pinterest account (11 clicks a month).
Time until first payout: The time until the reward is actually transferred varies depending on the approval period, payment date, and minimum payout amount for each program. Many systems have a hold period (about 30 to 60 days) for refund handling, and minimum payout amounts vary—roughly speaking, about $5 to $20 for major intermediary platforms (ASP = services connecting advertisers and introducers), and about $30 to $50 for direct-managed programs by businesses (Examples of breakdown: Impact $10 / Awin equivalent to $20 / PartnerStack $5). As an example of an actual payment cycle, systeme.io's official explanation shows an example where payment processing is scheduled to start 56 days after purchase (subject to conditions; not the exact deposit date) (systeme.io payment mechanism), and PartnerStack states that approved commissions are scheduled to become available for withdrawal around the 13th of the following month (there are exceptions such as delays. PartnerStack payment terms). In any case, it is not a system that promises payment within 30 days. This creates a discrepancy with the timeline of "hundreds of thousands of JPY a month starting from the 30th day."
Math confusing LTV and monthly income: In recurring commission programs, the total amount one contract brings over its lifetime (LTV = Life Time Value) is different from the commission actually generated in a given month. For example, assuming you acquire 20 new contracts a month at a recurring commission of $50 each, and assuming zero cancellations, the commission generated in the first month is only for the ones acquired that month, roughly $1,000. An amount often seen like "$10,000 a month" corresponds to the monthly generated commission in the 10th month if contracts continue to accumulate at the same pace, and the cumulative generated commission over those 10 months leading up to it would be roughly $55,000 (our lab's estimate based on a simplified premise of zero cancellations). Although often talked about as single-month results, it is actually the result of accumulation.
The claim that "there are no Japanese competitors": When competing with English content aimed overseas, the competitors are not Japanese speakers, but all publishers in the English-speaking world. Our lab's assessment is that this area is a highly competitive search field dominated by comparison sites like G2, Capterra, and Zapier (this is not the result of continuous research on search rankings, and will be a subject for future verification).
Government warnings: In 2025, the Consumer Affairs Agency announced cases where consumers were actually led to expensive additional support and failed to earn revenue, despite claims of "easy even for beginners," "500,000 JPY a month is the norm," and "money-back guarantee" (Consumer Affairs Agency warning). However, it should be noted that this case is a warning against a single domestic side-hustle support business, and is not a verification of AI, Pinterest, or overseas affiliate software in general.
The verification did not conclude that "everything was a lie." Recurring overseas SaaS affiliate marketing itself does exist. Registration for all of them is free, and the commission rates and minimum payout amounts for direct-managed programs are as follows (LiveChat's 22% is a rate limited to new contracts established after acquiring 5 paid customers).
| Service | Rate | Minimum payout amount |
|---|---|---|
| systeme.io | 60% recurring commission with no application required | $30 |
| AWeber | 30-50% recurring commission (lifetime) | $30 within US, $50 outside US (Payment terms) |
| LiveChat | 20-22% recurring commission with over 40 months average duration | $50 (Fee provisions) |
Affiliate platforms (ASP) like the aforementioned Impact, Awin, and PartnerStack have lower minimum payout amounts, starting from a few dollars to around $20. What we mainly could not find evidence for were the "speed" and "amount" parts, and we believe it is fair to say that the underlying mechanism itself exists.
These are verification items organized by our lab with reference to the aforementioned Consumer Affairs Agency case (which is a case of a single company). We will refine these through continuous research in the future.
If only the amount and period are emphasized without showing these, it is worth pausing to verify.
At Hakaru Lab, when introducing this type of side-hustle method, we have decided to plan with a scenario of 0 JPY in revenue as a conservative estimate (this is not an estimation of the statistical median from surveying a population). This is because we judged that it is a sincere attitude to set a premise of zero, as there is no guarantee that additional income will actually be generated when starting from a state with no foundation such as existing customers, advertising budgets, or followers.
Hands-on verification by our lab itself is yet to come. This article remains a cross-reference of official information and public data, but as a next step, our lab will begin initiatives to verify by actually doing the work ourselves. We will openly disclose the processes and numbers, including results that did not go well. In contrast to the expectations for the "speed" of promotional claims, we will honestly communicate the "slowness" and uncertainty involved in the verification.
Hakaru Lab's research, writing, and verification are carried out through a division of labor by multiple AI models. The decision on policy and final confirmation are handled by humans at the operating entity. The methods and tools introduced are documented only after confirming official information (terms and payment conditions) and public data with clearly identifiable sources. Hands-on verification by our lab itself will proceed sequentially as future initiatives, and we plan to publish them on the methodology page (under construction) along with details on verification procedures and standards.