Rogo does laundry, marketing to lemons, free money
Hating what you do, vibes for lemons, VC distribution glitch
Rogo does laundry
Rogo, an AI platform for automating the drudgery of investment banking, announced a $160M fundraise yesterday at a $2B valuation. The co-founders met while investment banking and started the business in 2021.
Bloomberg covered the story with the headline: “Junior Bankers Sick of Grunt Work Build $2 Billion AI Tool to Do the Job.” They go on to illustrate:
“investment bankers often toiled into the wee morning hours, crunching away on spreadsheets and rearranging logos on slide decks”
These startup stories usually start the same way: Founder hates doing something. It’s mundane, it’s dreary, it’s not living their best life. Founder looks around and realizes they aren’t the only ones that hate doing that thing. Founder then decides to do that thing for other people as a venture-backed startup and commits to 10 years of doing that thing for other people every day. Founder now is completely immersed in the thing they hated doing.
But now, it really is different.
To understand how, let’s go back to 2014 when we were in the heady days of “Uber for X.” You had a range of different startups all looking to transform industries and tasks into user-friendly conveniences with ordering available from smartphone apps. One of these startups was Washio, which was “Uber for laundry” and offered users the opportunity to get their laundry picked up with the touch of a button in an app and for it to be returned cleaned within 24 hours. Washio raised $17M to build out the vision across multiple cities.
New York Magazine did a feature on Washio and its founder, Jordan Metzner. The profile explains some of the early genesis of the startup:
Metzner scoured TaskRabbit, the website on which the broke and eager underbid one another for the chance to do tasks for the moneyed and lazy, to see what services were most in demand. A lot of people, it turned out, wanted someone to do their laundry. “That spoke to me,” says Metzner.
“I hate doing laundry,” Metzner says. “It’s [the] worst.”
The Washio operations involved collecting dirty laundry, coordinating drop off to wholesale cleaning locations, organizing the laundry, actual washing, dry cleaning for certain items that were separated out, dry and fold, repackaging, and finally redelivery to the customer. All this happened 7 days a week and on a 24 hour cycle to get customers their clothes back within 24 hours of being picked up. As a founder, you are up to your elbows in all of it to make sure it’s efficient enough to support the scaling and you are on the scene when washers break, clothes get lost, dryers fry, and workers steal.1
With physical “Uber for X” businesses, you can’t get away from the ins-and-outs of the very thing you hated to begin with when you started the business. But, now with AI, it’s different in meaningful ways.
For Rogo, there are no humans involved in producing the decks and models, just software code and AI models.2 Quality control can be executed without huge human involvement through test suites, monitoring and evaluations. There is no physical delivery or coordination—it has “bits” rather than “atoms.”
Rogo’s founders have a fundamentally different experience than what Washio’s founders and others went through 10 years ago. As they build out their product, the bandwidth and thinking that goes to simply delivering well for customers is markedly less than in that prior cycle. This frees them up to push forward aggressively on the product and getting it deeply embedded in the workflows of their investment banking customers.
The “crunching away on spreadsheets and rearranging logos on slide decks” can be completely eliminated by Rogo so that no one at Rogo actually needs to do it. The founders aren’t managing a team of people that are rearranging logos and that makes all the difference.
They have code that does it and with that they have bought their freedom. For today’s startups, it has become way more palatable to do something because you absolutely hate it.
Marketing to lemons
If your company was looking to cut corners, pull wool over eyes, or be as focused as possible on near term speed and profit, you might be looking for a slightly less reputable company to do your financial audit or SOC2 compliance work (which is needed to sell software to many enterprise customers).3
Normally, marketing for regular services involves convincing the customer that you are excellent, a great value, and you have a reputable brand established. But for services like auditing your financials or ensuring your company is SOC2 compliant, there are two distinct target customers for service providers to market to: normal companies and lemon companies that want you to let them do bad things.
First Brands Group was one such lemon. First Brands collapsed suddenly into bankruptcy in September and caused a stir in US credit markets due to the $12B in financing it had. Fraud charges were subsequently brought against the founder and his brother.
If you are a service provider marketing to the lemons, you want to do so in a way that still attracts normal customers too. A sort of vibe that only the lemons can sense would be ideal.
First Brands, according to reporting from the Financial Times, could spot the vibe:
First Brands Group selected BDO as its auditor due to the accountancy firm’s “less rigorous” approach, a former senior employee of the auto parts maker told a court-appointed examiner investigating its failure.
The FT goes on to explain:
The court filing […] cited a former senior employee who recalled a discussion “in which BDO was described as preferred among the larger firms based on its audit approach, which was viewed as less rigorous”. The executive also said BDO was characterized as having “the most unsophisticated audit process”, meaning “[w]e can do what we want with these guys”, the report said.
OK, but how did First Brands know that BDO would be great for lemons? Well:
“much” of BDO’s audit “was conducted remotely, without on-site presence at corporate locations”, according to the report. “Witnesses reported that BDO did not directly access underlying enterprise systems, such as Oracle or SAP, and did not consistently perform independent verification in certain areas, including customer confirmations, manual journal entry testing, or analysis of cash activity around period-end,” the examiner wrote.
Ah, OK. So BDO didn’t really do much of anything (allegedly). So it probably wasn’t so hard for First Brands to know they found a compatible auditor.
It turns out that some startups have also been looking for compatible SOC2 compliance providers and also found one (allegedly) in Delve. SOC2 compliance is all about security: proving you can securely manage client data, processing integrity, confidentiality, and privacy.
If you were not doing these things but still wanted the SOC2 stamp of approval to be able to sell to enterprise customers, you might start looking for service providers which gave off a certain vibe. And you found it, according to Techcrunch and reporting by the anonymous ‘DeepDelver’ on Substack:
Delve “achieves its claim of being the fastest platform by producing fake evidence, generating auditor conclusions on behalf of certification mills that rubber stamp reports, and skipping major framework requirements while telling clients they have achieved 100% compliance.”
DeepDelver went into considerable detail about those claims, accusing the startup of providing customers with “fabricated evidence of board meetings, tests, and processes that never happened,” then forcing those customers to “choose between adopting fake evidence or performing mostly manual work with little real automation or AI.”4
So, basically a smart structural hack allows Delve to serve normal companies as well as lemons. Delve allows the people that want thorough review to manually do all the work for it before it gets processed by a rubber stamping mill. But for the lemons, you could just adopt the fake evidence. And this would be pretty easy to spot during the buying process for their services.
Three different Delve customers have had major security breaches (the type SOC2 compliance checks are designed to prevent) despite their certification by Delve.
Of course, much of this is allegations at this point. Which may be helpful for advertising to lemons for both BDO and Delve? Or maybe not, because they’ll have to tighten things up now?
Regardless, seems inevitable that someone will always be selling compliance that looks like compliance but is actually just a rubber stamp. The product isn’t always the product.
Conflict creates free money
Meta’s purchase of AI startup Manus for $2.5B is being blocked by China on national security grounds, according to reporting in the Wall Street Journal. Manus has already disbursed the sale proceeds to its investors after the deal closed in December.
One of Manus’ venture capital investors, Benchmark, has already distributed the proceeds of the sale to its own limited partners. According to The Information:
Benchmark, which led the $75 million round of financing at a roughly $500 million valuation last spring, has already distributed the returns from the transaction, which closed last year, to its investors, according to a person familiar with the matter.
So, to recap: Meta bought Manus in December. The cash hit Manus’ account. Then Manus sent the cash from the acquisition to shareholders (including Benchmark) and started integrating with Meta and onboarding staff to Meta. Shareholders like Benchmark then passed the money they got from Manus to the limited partners (investors) in Benchmark’s fund. So now the money is no longer with Benchmark or Manus, it’s basically in the hands of pensioners and foundations.
And China is now pushing for the whole deal to be unwound.
Meta will now have to “divest” or sell Manus to a Chinese company, likely for far less than Meta paid. Meta will have paid $2.5B for nothing.
But Benchmark’s investors have their money free and clear, according to reporting from The Information:
“it would be difficult for Meta to get its money back from Manus’ investors,” Frank Aquila, an M&A lawyer at Sullivan & Cromwell, said.
Unwinding the closed transaction would set a “dangerous” standard for companies hoping to do cross-border deals and would be “close to unprecedented and almost impossible to put the ‘genie back in the bottle.’”
So it seems that the horse is out of the barn already on this one. This isn’t the first time that conflict created free money.
In 2020, Sequoia led the Series B round into a fintech startup, Finix. Sequoia’s check was $21M into a $35M round. The round closed in February 2020 and, according to Techcrunch, in early March 2020 they walked away:
Sequoia Capital has, for the first time in its history, parted ways with a newly funded company over a purported conflict of interest and, almost more shockingly, handed back its board seat, its information rights, its shares and its full investment.
It wasn’t a small check. According to sources close to the situation, it just gave up $21 million.
The conflict in question here was that Sequoia was an investor in Stripe. After the deal closed, Sequoia felt that the competitive conflict between Finix and Stripe was too material but had already wired the money.
Finix just kept the money and cancelled Sequoia’s shares. This was like receiving cash from the sky with absolutely no dilution or obligations. Benchmark’s investors might end up with the same type of windfall: free money from a sale that never really happened.
There are free lunches, after all.
Like this newsletter? Subscribe and forward it to your friends. Have feedback, thoughts to share or comments? Comment or reply to the newsletter.
By 2016, Washio had shut down operations and its assets were purchased by Rinse, which also offers laundry and dry-cleaning on demand via its smartphone app. OK, so in this case, it wasn’t 10 years of overseeing daily laundry for Metzner, but it would have been had it been successful.
Had Rogo been built 10 years earlier, it would have likely involved teams of people in the Philippines.
Auditors and compliance firms exist because lenders and customers can’t verify a company’s claims themselves, so they outsource trust to a third party.
Further according to DeepDelver and Techcrunch:
Those firms, they said, are just rubber-stamping reports that were generated by Delve. As a result, DeepDelver said the startup “inverts” the normal compliance structure: “By generating auditor conclusions, test procedures, and final reports before any independent review occurs, Delve places itself in the role of both implementer and examiner. This is not a technicality. It is a structural fraud that invalidates the entire attestation.”