Show Notes
Every time you unlock your phone, somebody makes money. It just isn't you.
Mode.Inc has raised over $100 million across Reg CF and Reg A+ offerings with 65,000-plus shareholders.
Forty to sixty hours a week. That is what the average person spends on a smartphone, which works out to somewhere between a third and half of your waking life. Big tech built trillion dollar fortunes on that time. You signed up for exactly none of it. Dan Novaes wants to change the math.
He is the co-founder and CEO of Mode.Inc, the company behind Mode Mobile, the EarnPhone, and EarnOS. He has raised over $110 million, all of it from retail investors, not a single traditional venture round in the mix. He has 65,000 shareholders who own a piece of the thing they use. He built a physical smartphone, got it into Walmart and Best Buy, then looked at the P&L and walked away from device manufacturing. He now buys “dead” consumer apps and turns them into first-party data assets for AI labs.
Oh, and he lost 35 pounds and ran a marathon while doing it. The best ideas, he says, come in the float tank.
This conversation moves fast. Ryan and Dan get into the Anthropic settlement, the Cloudflare crawler wall going up September 15, why hedge funds are buying your Amazon receipts, why the 55-plus demographic funded a company built for budget-conscious app users, and why a QR code reader was one of the smartest acquisitions Dan ever made.
Named Frameworks
The Consent Exchange
Consent is not a checkbox problem. It is a pricing problem.
- →Every product has a paying segment and a non-paying segment. Most companies just lose the second group.
- →Offer the non-payers a trade: consented, PII-stripped data in exchange for the paid tier.
- →Dan's own math on Strava: a $79.99 subscription, minus Apple's 30% cut, is roughly what a year of anonymized health data is worth on the open market. The trade nets out.
- →The reward can be cash, product access, sweepstakes entry, or credits. What matters is that the incentive is real enough that people actually opt in.
- →"You can only get consent if you align incentives." If Facebook offered you an extra like for your data, you would laugh.
M&A As User Acquisition
Dan does not think of acquisitions as acquisitions. He thinks of them as UA.
- →Every acquired app arrives with an audience, so the customer acquisition cost is baked into the purchase price rather than paid to Meta.
- →Layer EarnOS on top of a business that is already profitable from ads and subscriptions. Continued profit covers ops. No burn.
- →Each acquisition also adds a new first-party demographic to the data supply.
- →Consumer app valuations are depressed right now because bootstrapped founders are scared AI eats their category in two or three years. That fear is the discount.
The Laggard Arbitrage
Everyone is racing to serve early adopters. Dan buys the other end of the curve.
- →He bought a QR code reader in 2026. His first reaction was "who is even using this?"
- →The financials: roughly $10M revenue, $5M EBITDA, about 80% subscription, 20% cohort retention at year five.
- →The audience turned out to be boomers, using QR creation features and scanning things a phone camera would not read.
- →AOL is the extreme version of this thesis. Bending Spoons paid about $1.45 billion for a business throwing off roughly $334 million a year, and 30 million people still use it monthly.
- →"People are really quick to say that business is dead. But you're forgetting the late adopter and the laggard. That's a big space."
Data Exhaust Monetization
Mode Media exists because Dan noticed what he was already sitting on.
- →Tens of millions of people had signed up for the app over the years, most of them through Gmail. Consumer businesses churn. The emails stayed.
- →What he knew about those people: they like to earn and save money.
- →So he built a newsletter about freebies, free finds, and free samples. It started generating seven figures a year.
- →Mode Media is now a mid-to-high seven-figure annual business by Dan's account, built entirely from what he calls data exhaust.
- →The same instinct produced m0de.ai. "We're sitting on all this data. We should really create new data products, because we already have this engine going."
The Fundraising Business Unit
Crowdfunding is not a campaign. It is a department.
- →Treat capital raising the way you would treat a product team, with owned headcount, a roadmap, and a budget.
- →Do not use marketplace platforms. Dan chose DealMaker precisely because it is white-label, so he keeps the audience. DealMaker Securities acts as broker-dealer of record while the raise lives on your own domain.
- →Retail investors need a completely different pitch from VCs. Different sophistication, different vocabulary, different emotional driver.
- →The channel that worked: newsletters. Financial and stock-focused ones especially. Facebook ads alone will not carry a crowdfund because eCPMs keep climbing.
- →Founders must be willing to do webinars, tell the story on camera, and show up repeatedly.
- →Reg A+ Tier 2 lets you raise up to $75 million from retail in a 12-month window.
Zone Of Genius Pruning
Knowing what to stop doing.
- →Mode built two devices during the COVID cheap-money window, before the AI chip boom, and sold tens of thousands of units into major retail.
- →The verdict: hardware is roughly a 1% gross margin business. "You spend a hundred bucks to make a dollar." Add shipping, add tariffs, add cost of capital.
- →The pivot was not abandonment. It was licensing. Mode now works with carriers and OEMs on a Roku-style model, providing the software layer so partners can launch their own earn-enabled devices.
- →The data justified keeping the thesis: EarnPhone users showed roughly 3x higher earnings and 3x higher retention than app-only users.
Founder Experiment
Run this over the next 30 days using your existing analytics. Pull your conversion rate on your paywall: Dan's acquired spam-filtering app converted about 1 in 20 visitors, which is genuinely good for a hard paywall, and that still means 19 out of 20 people walked. Now build a second door: same product, free tier, one condition, the user consents to share a specific, bounded data signal, purchase receipts, usage patterns, health metrics, whatever your product legitimately touches. Three things to get right: name the exact data, since vague permissions kill opt-in rates (“we see what you buy on Amazon and Walmart” converts better than “we may collect certain usage information”); strip PII before it leaves your system and say so on the consent screen; and make the value obvious and immediate, not “help us improve” but a real unlock they wanted anyway. Measure two numbers after 30 days: what percentage of the 19 took the trade, and what that cohort is worth per user per year against your paid cohort. If the free-with-consent cohort clears even 40% of paid ARPU, you just found revenue in the segment you were writing off. Bonus round: go look at your dead leads, everyone who signed up and churned or never converted, the way Dan turned exactly that list into a seven-figure media business, and ask what single interest all of those people demonstrably share.
Glossary
EarnOS
Mode's proprietary rewards and monetization layer, designed to sit on top of any consumer app.
EarnPhone
Mode's smartphone, sold through Walmart, Best Buy, and other retail. Discontinued as a first-party hardware product, now the basis of an OEM licensing model.
m0de.ai
Mode's data division, which sells consented supply to AI labs, enterprises, brands, researchers, and financial buyers. Spelled with a zero.
Reg A+ (Regulation A Tier 2)
An SEC exemption letting a company raise up to $75M from accredited and non-accredited investors in a 12-month period. Requires audited financials and ongoing reporting. Preempts state blue sky review.
Reg CF (Regulation Crowdfunding)
Caps at $5M in any 12-month period. Must run through a registered funding portal or broker-dealer. Mode's first raise.
DealMaker
White-label capital-raising infrastructure. The issuer runs the raise on its own domain while DealMaker Securities acts as broker-dealer of record. Roughly half of all Reg A+ dollar volume in 2025.
Alt data
Non-traditional datasets bought by hedge funds and CPG companies for investment and product decisions. SKU-level purchase data is one of the most valuable categories.
Consented data
Data supplied with explicit, informed user permission. The regulatory and commercial bar that frontier AI labs increasingly demand.
Mixed-use crawler
A bot that blends search indexing, real-time agent retrieval, and model training. Cloudflare's September 15, 2026 default blocks these on ad-carrying pages.
Pay Per Use
Cloudflare's successor to Pay Per Crawl. Pays publishers when their content shapes an AI answer rather than when a bot fetches the page.
UA (user acquisition)
The paid and organic motion of getting new users. Dan reframes M&A as a UA channel.
Cohort retention
What percentage of users who joined in a given period are still active later. The QR code app held 20% at year five, which is exceptional.
Float tank
Sensory deprivation tank. Dan credits it with "tens of millions of dollars" worth of ideas.
Q&A: What Founders Ask After This Episode
What is the consent economy?
A model where users are paid, in cash, product access, or credits, for data and attention they currently give away for free. It exists because AI labs now need consented, legally clean training data, and because the old extraction model is running into settlements like Anthropic's $1.5 billion payout to authors.
Can you actually raise serious money through equity crowdfunding?
Yes, but almost nobody does it well. Mode.Inc has raised over $110 million across Reg CF and Reg A+ offerings from roughly 65,000 investors. Dan is explicit that this required treating fundraising as a standing business unit with paid acquisition, not a campaign you launch and hope goes viral.
Why do founders get rejected at Series A and B even with a good business?
Pattern matching. Dan's seed round worked because seed investors buy founder grit. Later-stage investors buy legible category fit, and a company with hardware, software, media, and M&A all running at once does not pattern match cleanly. The fix was not becoming more legible. It was finding capital that valued the story instead.
How do you monetize users who will never pay?
Offer a consent trade. Free tier in exchange for a specific, bounded, PII-stripped data signal that has real market value. Dan's benchmark: even a hard paywall converting at 1 in 20 is leaving 19 people on the table.
Is buying an old, unsexy app a good acquisition strategy in the AI era?
It can be a great one. Valuations for consumer apps are depressed because founders fear AI disruption. Late adopters and laggards represent a large, sticky, underserved market. And every acquired app brings both an audience and a fresh first-party data set.
Why is hardware such a hard business?
Roughly 1% gross margins, brutal shipping costs, tariff exposure, and rising component prices as AI chip demand pulls up the whole electronics stack. Dan built and shipped two devices, learned a lot, and moved to a licensing model with carriers and OEMs.
Who actually invests in equity crowdfunding campaigns?
Frequently not your users. About 85% of Mode's investor base was over 55, a group that was not the target app demographic at all. They connected with the story about big tech extracting value from their kids and grandkids.
How do AI companies get training data legally in 2026?
Licensing and consent. Reddit gets roughly $60M a year from Google. Cloudflare's September 15 default flips a large slice of the web to blocked-unless-paid. Companies like Mode.Inc are building consented supply as a product.
