Which indie SaaS have lost the most revenue lately?

Last updated: 7 September 2026

SUMMARY

Photo AI is the clearest named indie SaaS revenue loser we can verify lately: its publicly disclosed monthly revenue has fallen from $105,000 earlier this year to roughly $80,000, while Chariot AI has suffered the sharper percentage decline at around 41% from its recent MRR high.

The ranking is messier than a simple leaderboard suggests. Public trackers mix recurring SaaS with agencies, marketplaces, one-time products and anonymous businesses, and stale negative-growth cards can make a recovered company look like it is still collapsing.

The strongest comparisons come from businesses where we can reconstruct two useful points in time using payment-verified MRR, recent cash collections, subscriber counts or first-party founder disclosures. That is why named, inspectable businesses matter more here than spectacular anonymous percentage drops.

Photo AI stands out in absolute dollars, but the cause is not obvious. The product still attracts substantial attention, so the decline points more toward weaker conversion, retention, customer spend, traffic quality or some combination of those factors than toward simple disappearance of demand.

Chariot AI tells a different story. Its MRR has fallen from roughly $13,800 to around $8,200 after a fast breakout, which looks less like a mature SaaS slowly decaying and more like a young product that found a burst of demand it has not fully held.

Interior AI is one of the cleaner cases where traffic and revenue are falling together. Its public monthly revenue dropped from roughly $27,000 to $23,000 in a short window while estimated traffic more than halved across two months.

BuildMyAgent.io and FriendFilter + GroupFilter are important because the decline is visible in more than one operating metric. Their falling MRR is accompanied by weaker recent collections or fewer subscribers, making the contraction harder to dismiss as billing noise.

Some apparent losers disappear once the data is refreshed. Aplano looked catastrophic on an older dashboard but fresher readings put it back around $245,000-$249,000 MRR, while ThesisAI appears to have stabilized around six figures after a sharp earlier correction.

The broader indie AI SaaS market is not collapsing. TrustMRR still shows large and growing AI businesses such as GojiberryAI, Chatbase and LLM Gateway, which means the more useful story is increasing dispersion: some products scale extraordinarily fast while others give back meaningful revenue just as quickly.

The recurring vulnerability across many of the declines is concentration. Tiny SaaS businesses can become extremely profitable by relying on one product, one audience, one platform or one acquisition engine, but that same focus can make revenue move much faster when the underlying source of demand weakens.

The practical conclusion is sharper than 'indie SaaS is struggling.' Several successful bootstrapped products are clearly shrinking, Photo AI has the largest named recent dollar decline we can confidently reconstruct, and the public data is now good enough to watch meaningful post-peak contraction in businesses that are still perfectly viable.

Get the biggest database of
profitable internet businesses

We mapped 300+ proven digital businesses so you can skip the blind trial and error. For each one, you get the site, the revenue numbers, the distribution strategy, the repeatable patterns, and ideas to recreate the model in a different niche, channel, or angle.

Get the full database →

Why is ranking the biggest indie SaaS revenue drops so messy?

Ranking the biggest indie SaaS revenue drops is harder than it looks because the public data mixes real recurring software businesses with agencies, marketplaces, one-time products and anonymous companies.

TrustMRR currently tracks more than 4,000 businesses in its SaaS category, but only a small fraction publish enough history for us to compare today's revenue with a meaningful earlier level. Some of the largest declines are also hidden behind names such as “Anonymous Startup.” PulseMRR, for example, has displayed anonymous businesses with six-figure MRR and enormous percentage drops. Without knowing what those companies sell, how large their teams are or whether the revenue is genuinely recurring, putting them in an indie SaaS ranking would be guesswork.

Freshness creates another problem. A negative growth card can stay visible after the underlying business has recovered. We found exactly that with Aplano: an older PulseMRR reading still makes the company look like one of the biggest losers, while newer payment data tells a very different story.

For this analysis, we therefore focus on founder-led or very small-team software businesses where we can compare either payment-verified revenue or a first-party public revenue figure across two useful points in time. We also separate falling cash collected from falling MRR whenever the distinction changes the conclusion.

Which indie SaaS have actually lost the most revenue lately?

Photo AI is the clearest named dollar loser we can verify right now, while Chariot AI has suffered one of the sharpest percentage declines among smaller recurring SaaS products.

The comparison still needs some care because these founders do not all publish the same metric on the same day. Pieter Levels publicly reports rounded monthly revenue for his products. TrustMRR gives us payment-verified MRR and 30-day collections for businesses such as Chariot AI and BuildMyAgent. FriendFilter's useful baseline comes from the revenue level advertised in its own sale listing.

Instead of pretending those numbers are perfectly standardized, we can show exactly what changed.

Indie SaaS Earlier comparable level Current level Approximate drop Metric
Photo AI $105K/month ~$80K/month -$25K, -24% Founder-reported monthly revenue
Chariot AI ~$13.8K MRR ~$8.2K MRR -$5.6K, -41% Payment-verified MRR
Interior AI ~$27K/month ~$23K/month -$4K, -15% Founder-reported monthly revenue
BuildMyAgent.io ~$19.9K MRR ~$17.1K MRR -$2.8K, -14% Payment-verified MRR
FriendFilter + GroupFilter $11.3K MRR ~$8.6K-$8.8K MRR ~-$2.5K, -22% Payment-verified MRR

Get the biggest database of
profitable internet businesses

We mapped 300+ proven digital businesses so you can skip the blind trial and error. For each one, you get the site, the revenue numbers, the distribution strategy, the repeatable patterns, and ideas to recreate the model in a different niche, channel, or angle.

Get the full database →

Is Photo AI still losing revenue right now?

Photo AI is still moving down: Pieter Levels' current public profile puts the AI photography SaaS at roughly $80,000 a month, versus the $105,000 monthly revenue he disclosed in March 2026.

That already represents around $25,000 of monthly revenue disappearing, or close to one quarter of the business. The very recent movement is also negative. A browser snapshot of Levels' public profile from roughly two weeks earlier showed Photo AI at about $89,000 a month. The current figure is around $80,000.

We would not treat rounded figures in an X bio like audited accounts, but the direction is unusually clear. Levels himself supplied the $105,000 figure earlier in the year, then publicly displayed $89,000 and now roughly $80,000.

Photo AI had reached even higher levels before that. Contemporary reporting around its stronger period put revenue above $130,000 a month. Using that peak would make today's decline look much larger, although $105,000 is the cleaner baseline because Levels personally disclosed it this year.

Traffic also complicates the story. Photo AI has continued to attract substantial attention, so the decline cannot be explained simply by the website disappearing from the internet. Something inside the monetization funnel has weakened: conversion, retention, spending per customer, traffic quality, or some combination of them. The public numbers do not let us confidently separate those causes yet.

Did Chariot AI fall back after its $13.8K month?

Chariot AI has fallen hard from its recent high, with payment-verified MRR now around $8,200 after reaching roughly $13,800 earlier in 2026.

SaaS Distribution reconstructed Chariot AI's revenue history from TrustMRR and found that the AI website builder jumped to about $13,800 a month during its breakout period. The latest TrustMRR page shows roughly $8,200 MRR and 373 active subscriptions.

That works out to an MRR decline of around 41% from the high.

Interestingly, Chariot AI collected almost $11,000 over its latest 30-day window, more than its normalized MRR. Annual payments and billing timing can explain part of that gap, which is why the MRR decline gives us the cleaner view of the subscription base.

Chariot is also young. It launched in 2025, sells a $17 monthly website-building product and acquired users through channels including social media and affiliates. A rapid rise followed by a 40% pullback looks very different from an old SaaS slowly losing a few customers each month. Chariot appears to have found a burst of demand that has been difficult to hold at the same level.

Get the biggest database of
profitable internet businesses

We mapped 300+ proven digital businesses so you can skip the blind trial and error. For each one, you get the site, the revenue numbers, the distribution strategy, the repeatable patterns, and ideas to recreate the model in a different niche, channel, or angle.

Get the full database →

Is Interior AI sliding at the same time as Photo AI?

Interior AI is also shrinking lately, with Pieter Levels' public revenue figure falling from roughly $27,000 a month to around $23,000 in about two weeks.

The $4,000 movement looks modest next to Photo AI, but it represents roughly 15% of the business in a very short window. Earlier public estimates had Interior AI closer to the $38,000-$45,000 range, so the longer decline is considerably larger.

The traffic data moves in the same direction. Semrush estimated around 65,000 visits for Interior AI in April 2026, about 43,000 in May and roughly 31,000 in June. That is a traffic reduction of more than half in two months.

Here we have a cleaner explanation than we do for Photo AI. Fewer people are reaching Interior AI, and monthly revenue is falling as well. Competition in AI interior design has also become much denser since the product launched, with image models and general-purpose AI tools increasingly able to perform similar transformations.

Interior AI still makes serious money for an extremely lean product. Its recent direction, though, is clearly down.

Has Nomads.com quietly lost more revenue than people realize?

Nomads.com is currently only around $15,000 a month in Levels' public figures, far below several previously disclosed revenue levels for the long-running digital-nomad business.

The very recent movement is small. A snapshot from roughly two weeks earlier showed about $16,000 a month, so the latest change is around $1,000.

Zoom out and the decline becomes much more striking. Public figures put Nomads.com around $38,000 a month during 2025, while older peaks were higher again. Moving from roughly $38,000 to $15,000 means around 60% of that monthly revenue has disappeared.

We should be careful with the word SaaS here. Nomads.com has changed its pricing model over the years and currently promotes inexpensive one-time memberships rather than behaving like a conventional monthly B2B subscription product. That makes its revenue less comparable with something such as BuildMyAgent.

Still, Nomads.com is useful for understanding the wider pattern. A famous internet business can remain active, have tens of thousands of members and continue making meaningful money while becoming much smaller than it once was.

Get the biggest database of
profitable internet businesses

We mapped 300+ proven digital businesses so you can skip the blind trial and error. For each one, you get the site, the revenue numbers, the distribution strategy, the repeatable patterns, and ideas to recreate the model in a different niche, channel, or angle.

Get the full database →

Is BuildMyAgent.io actually shrinking?

BuildMyAgent.io is genuinely shrinking now because both its recurring revenue and its recent cash collections have moved down.

An earlier public snapshot put the AI-agent platform around $19,900 MRR with approximately $19,600 collected over 30 days. TrustMRR's newer payment data shows roughly $17,100 MRR, 353 active subscriptions and about $14,600 collected over the latest 30 days.

The normalized recurring base has therefore fallen about 14%, while cash collected over 30 days is down roughly 25% across those snapshots.

Those two measures moving together make BuildMyAgent a cleaner decline than businesses where only cash collection suddenly drops.

The company has still generated more than $700,000 in verified lifetime revenue, so we are looking at contraction inside an established small SaaS rather than a project going to zero. Its founder has also said that paid acquisition remains largely untapped. BuildMyAgent could recover, but the revenue deterioration visible today is real.

Has FriendFilter lost customers as well as MRR?

FriendFilter + GroupFilter has lost both recurring revenue and subscribers, making it one of the cleaner examples of an older indie SaaS moving backward.

The business has been advertised at $11,300 MRR from roughly 390 subscribers. Current TrustMRR readings put FriendFilter around $8,600-$8,800 MRR, with its latest detailed page showing about 355 active subscriptions.

That puts MRR roughly 22% below the advertised level and the subscription base about 9% lower.

FriendFilter is especially interesting because the product launched in 2019 and solves a straightforward Facebook-management problem. There is no obvious generative-AI boom whose temporary revenue is simply evaporating.

Its dependency is Facebook itself. FriendFilter helps users identify and remove inactive connections, while GroupFilter serves people managing Facebook groups. Any change in Facebook usage, APIs, platform rules or the importance creators place on Facebook can affect the value of the product.

For FriendFilter, the decline looks like real subscriber attrition rather than a strange billing month.

Get the biggest database of
profitable internet businesses

We mapped 300+ proven digital businesses so you can skip the blind trial and error. For each one, you get the site, the revenue numbers, the distribution strategy, the repeatable patterns, and ideas to recreate the model in a different niche, channel, or angle.

Get the full database →

Did Aplano actually lose 44% of its recurring revenue?

Aplano should no longer be described as one of the biggest current indie SaaS losers because fresher payment data puts its recurring revenue back around $249,000 MRR.

An older PulseMRR dashboard still displays Aplano around $249,000 MRR with a -44.5% growth figure. That looks dramatic until we inspect how old the underlying dashboard snapshot is and compare it with newer readings.

SaaS Distribution's more recent TrustMRR-derived crawl puts Aplano at roughly $249,000 MRR. A recent TrustMRR founder page also showed approximately $245,000 MRR and around $210,000 collected over 30 days.

Whatever caused the earlier negative reading, the claim that Aplano has currently lost almost half of its recurring business does not survive a fresh check.

Aplano looked like our biggest named loser when we first inspected the older snapshot. The newer numbers are strong enough that we should remove it from that ranking altogether.

Did ThesisAI's 26% MRR drop keep getting worse?

ThesisAI suffered a large revenue correction earlier this year, but the latest available data puts the academic-writing SaaS around $120,000 a month rather than showing an ongoing collapse.

The earlier payment-derived snapshot was ugly: roughly $125,700 MRR, about $119,100 collected over 30 days and a -25.9% month-on-month growth reading. At that pace, ThesisAI would have fallen below $60,000 within only a few months.

That did not happen.

Recent VibeCrowd data still places ThesisAI at approximately $120,000 a month, while the product remains active and continues operating its recurring affiliate program. The business appears to have absorbed a sharp correction and then stayed around six figures.

ThesisAI deserves to be remembered as a big monthly drop. Calling it one of the SaaS businesses losing the most revenue right now would overstate what the newer data shows.

Get the biggest database of
profitable internet businesses

We mapped 300+ proven digital businesses so you can skip the blind trial and error. For each one, you get the site, the revenue numbers, the distribution strategy, the repeatable patterns, and ideas to recreate the model in a different niche, channel, or angle.

Get the full database →

Which apparent indie SaaS losers have already bounced back?

Several businesses that recently looked like revenue losers are growing again today, which shows how quickly an MRR-decline article can become stale.

TrustMRR itself is a good example. PulseMRR previously showed the one-person business around $20,500 MRR with an -8.1% reading. TrustMRR's live payment data now puts it at roughly $26,500 MRR and almost $47,000 of revenue over 30 days.

1Lookup also appeared around $199,000 MRR in an earlier snapshot and is currently above $230,000 MRR. HypeProxies can move down by a few thousand dollars in a short-term tracker while still showing strong positive monthly growth and collecting roughly $350,000 over 30 days.

Indie business Earlier negative reading Fresh reading What happened
TrustMRR ~$20.5K MRR, -8.1% ~$26.5K MRR Clear rebound
1Lookup ~$199K MRR ~$232K MRR Recovered well above the earlier level
HypeProxies Short-term MRR dips ~$350K 30-day revenue Monthly trend remains strong

Are indie AI SaaS products getting crushed right now?

Indie AI SaaS is not broadly collapsing today; the market is producing large winners at the same time that individual products such as Photo AI, Interior AI and Chariot AI are pulling back.

The freshest TrustMRR statistics are useful here. Its SaaS category contains more than 4,000 verified businesses and currently shows positive average 30-day growth. The same database includes AI businesses ranging from tiny experiments to companies doing hundreds of thousands of dollars per month.

GojiberryAI, for example, is currently above $400,000 MRR in TrustMRR-derived data and has continued growing. Chatbase sits above $800,000 MRR. LLM Gateway has moved into the hundreds of thousands of dollars of monthly revenue. Those businesses are expanding in the same environment in which some celebrated indie AI products are shrinking.

That gives us a more interesting picture than an “AI SaaS is dying” story. Revenue is becoming extremely uneven.

Building an AI product has become easier, competition arrives faster, and technical differentiation can disappear quickly. At the same time, products with strong distribution or a valuable business workflow can reach revenue levels that were exceptional for bootstrapped SaaS only a few years ago.

Get the biggest database of
profitable internet businesses

We mapped 300+ proven digital businesses so you can skip the blind trial and error. For each one, you get the site, the revenue numbers, the distribution strategy, the repeatable patterns, and ideas to recreate the model in a different niche, channel, or angle.

Get the full database →

What do the biggest indie SaaS revenue drops have in common?

The clearest indie SaaS declines tend to come from businesses that became unusually dependent on one product, one audience or one acquisition engine.

That concentration is part of why indie software can be so profitable. A tiny company does not need five sales teams, ten product lines and a global marketing operation. One founder can serve one audience with one product and keep enormous margins.

Revenue can also move quickly when the thing driving demand weakens.

Blabla gives us a fresh example outside the main ranking. The AI social-inbox SaaS built a large multilingual SEO footprint and signed recognizable enterprise customers, but its founder now openly says the team struggled with distribution. Payment data currently shows about $8,200 MRR, while recent cash collections are considerably lower. The product itself is still operating; finding enough new customers appears to be the harder part.

We see similar exposure across the wider group. Some products depend heavily on search traffic. Others depend on a founder's audience, affiliates or a host platform such as Facebook. AI products have an extra problem because competitors can reproduce features extraordinarily quickly.

The common vulnerability is concentration. The same narrow focus that lets a tiny SaaS become highly profitable can make its revenue change much faster than people expect.

Which indie SaaS have lost the most revenue lately?

As seen above, Photo AI is currently the strongest named answer in absolute dollars: its publicly disclosed monthly revenue has fallen from $105,000 earlier this year to roughly $80,000 now, with another noticeable drop visible in the most recent public snapshots.

Chariot AI stands out on percentage decline, with MRR roughly 40% below its recent high. BuildMyAgent.io and FriendFilter + GroupFilter give us cleaner evidence of recurring contraction because their payment-verified MRR has fallen alongside other operating metrics. Interior AI is also moving down quickly, while Nomads.com has suffered a much larger decline when we use a longer comparison window.

Aplano changes the story in the opposite direction. Fresh data no longer supports calling it a major current loser, despite an older dashboard still showing a huge negative percentage. ThesisAI also looks far less alarming now than its earlier -25.9% month suggested.

So the best current answer is fairly specific: several well-known indie SaaS businesses really are losing meaningful revenue, and Photo AI has the largest named recent dollar decline we can confidently reconstruct. We do not see evidence of a general indie SaaS collapse. What we do see these days is something that used to be much harder to observe publicly: successful bootstrapped products reaching serious scale, peaking, and then giving back $5,000, $10,000 or $25,000 of monthly revenue while remaining perfectly viable businesses.

Get the biggest database of
profitable internet businesses

We mapped 300+ proven digital businesses so you can skip the blind trial and error. For each one, you get the site, the revenue numbers, the distribution strategy, the repeatable patterns, and ideas to recreate the model in a different niche, channel, or angle.

Get the full database →

OUR METHODOLOGY

Determining which indie SaaS businesses are losing the most revenue sounds simple, but the public picture is surprisingly unclear. Revenue data is fragmented, businesses report different metrics, comparison periods vary, and a dramatic negative number can sometimes be nothing more than an outdated snapshot.

Rather than relying on anecdotes, reputation or a vague sense of which products seem to be struggling, we broke the question into measurable dimensions: absolute revenue lost, percentage decline, changes in recurring revenue, recent cash collections, subscriber movements, the length of the decline, the freshness of the underlying data, and evidence of subsequent recovery.

For each dimension, we gathered the freshest relevant evidence we could find and prioritized the strongest available data. Payment-connected revenue was especially useful for MRR, collections and subscriber counts, while founder disclosures gave us a direct view where verified payment data was unavailable.

We then checked older readings against newer ones instead of assuming a previously reported decline was still continuing. Aplano is the clearest example of why that matters: an older negative reading looked severe, but fresher data was strong enough to remove it from the current-loser ranking.

We assessed the indicators together rather than letting one dramatic percentage or one bad month decide the ranking. A company can suffer the largest dollar decline without having the largest percentage decline, while another can collect less cash in a given month without its underlying recurring base weakening by the same amount.

The final conclusions therefore come from aggregating several recent indicators company by company, then comparing them across the wider group. That also lets us separate genuine recurring contraction from billing noise, stale snapshots, rebounds and broader claims about the indie SaaS or AI software market.

The goal is not to manufacture a perfectly standardized leaderboard from imperfect public data. It is to build the strongest ranking the available evidence supports, while keeping the underlying differences in metrics visible rather than hiding them.

Key sources used for this analysis include TrustMRR's explanation of revenue verification, TrustMRR's API documentation, the TrustMRR SaaS database, the TrustMRR AI database, Chariot AI's verified revenue profile, BuildMyAgent.io's verified revenue profile, FriendFilter + GroupFilter's verified revenue profile, TrustMRR's own revenue profile, 1Lookup's verified revenue profile, HypeProxies' verified revenue profile, LLM Gateway's verified revenue profile, Blabla's verified revenue profile, Pieter Levels' direct disclosure that Photo AI was generating $105K/month, Pieter Levels' current project portfolio, Photo AI's first-party founder page, Levels' update on Nomads.com's pricing and business-model evolution, Aplano's first-party product site, ThesisAI's first-party product site, and Roman's verified founder portfolio for GojiberryAI.

Get the biggest database of
profitable internet businesses

We mapped 300+ proven digital businesses so you can skip the blind trial and error. For each one, you get the site, the revenue numbers, the distribution strategy, the repeatable patterns, and ideas to recreate the model in a different niche, channel, or angle.

Get the full database →
Steal What Works

Who wrote this?

STEAL WHAT WORKS TEAM

We study profitable internet businesses, take them apart, and write down what actually works: pricing, distribution, growth, packaging. We turn 300+ proven examples into a database so founders can stop testing random ideas and start from proof. Explore the database →

Back to blog