Judge Leonie Brinkema forced Google to open key ad-tech seams, Jeff Green says Google should leave the open internet, and Google says breakup was the wrong cure — but the startup opening is still a cramped little crawlspace, not a gold rush
What you’re reading: a real Opportunity Dossier from our AI research pipeline — Ada spotted the signals, Bea shaped the candidates, Cora dug deep, Edie tried to kill it and owns the score, and Della wrote it up. New to our scoring? Start with How Our Star Score Works. Research date: September 2026 — facts move; check dates before betting money.
Fluidity Rating: ★☆☆☆☆ (1/5). Real legal change, real niche need, but slow proof cycles, trust-heavy sales, ugly integrations, and entrenched incumbents make this a poor Return-on-Founder-Time bet for a stranger.
Jeff Green, the co-founder of The Trade Desk, says Google should “exit the open internet.” Lee-Anne Mulholland, Google’s vice president for regulatory affairs, says the court rightly rejected a breakup. Judge Leonie Brinkema says breaking up Google’s ad business is “neither realistic nor needed” and that behavioral remedies can “effectively pry open to competition” the harmed markets. Rajeev Goel, the co-founder of PubMatic, says the ruling could open one of the biggest pools of ad demand to real competition. The fight is not about whether something changed. The fight is about whether the change creates startup oxygen, or just reshuffles power among companies that already live in the ducts. Axios
“I do think that it makes sense for Google to exit the open internet and to focus on YouTube.” Axios
“We’re very pleased the Court rejected the DOJ’s proposal to break apart tools that help small businesses reach new customers and grow.” TechCrunch
Structural remedies are “neither realistic nor needed,” and behavioral remedies can “effectively pry open to competition” the harmed markets. FindLaw
The AdWords remedy “could open one of the market’s largest pools of buy-side demand to competition on the merits of technology and performance.” AdExchanger
Adtech competitive openings after legal adjustments means startup ideas built around the court forcing Google’s publisher ad tools to interoperate more cleanly with rivals, especially for open-web display ads on publisher websites. A stranger should care because Google’s ad plumbing quietly decides how a big chunk of the web gets paid, and a founder should care because legal shocks sometimes create small but real seams where a new tool can wedge in. Our verdict: the seam is real, but the seam is narrow.
The court opened a door, but the door leads into a crowded boiler room.
Publishers will not buy a political story; publishers will buy a revenue lift story.
The best business here is probably an ugly audit-and-migration tool, not a glamorous new ad empire.
The allure is easy to feel. For years, critics said Google’s control of publisher ad servers, ad exchanges, and buy-side demand let Google steer too much traffic through Google-owned pipes. Then a federal judge did not blow up the stack, but did order real changes. The U.S. Department of Justice said on September 16, 2026 that Judge Brinkema’s remedies require that “Google must create and support integrations between AdX and Prebid, and DFP and Prebid,” while also stopping preferential bidding by Google Ads into Google-owned tools and expanding publisher data access. For founders who know the history, that language lands like a crowbar at exactly the right joints. Justice Department
The believers’ dream goes like this: if Google must let rival pathways connect more cleanly, publishers finally get a fairer test of what inventory is worth, rival tools get cleaner access to demand, and a neutral control layer becomes valuable. A founder does not need to build a whole new exchange. A founder only needs to become the trusted dashboard, migration helper, experiment runner, or “show me where money is leaking” layer for publisher revenue teams. Judge Brinkema herself wrote that behavioral remedies could “effectively pry open to competition” the harmed markets, which is unusually explicit language for anyone hunting a commercial opening. FindLaw
The money is not imaginary. One eMarketer forecast, which we could not independently confirm, puts open web programmatic display ad spending at $42.53 billion in 2027, up from $31.57 billion in 2023. A founder does not need all of that market. A founder needs a sliver of publisher software spend or managed-service spend attached to that market. If 8 to 20 publishers pay roughly $25,000 to $100,000 per year for observability, routing analysis, or ongoing support, a real small business can exist. The investor lens can see that shape clearly enough: a services-plus-software company can emerge if the tool becomes embedded in publisher revenue operations. EMarketer
But the skeptic gets the next turn, and the skeptic has ugly facts on the skeptic’s side. The court did not order a breakup. The court ordered six years of behavioral remedies, a monitor, and a technical committee. A behavioral remedy creates interface work, compliance work, and migration work. A behavioral remedy does not create a blank field for a new full-stack giant. The court also limited the liability finding to publisher ad server and ad exchange markets in open-web display advertising, not the whole ad market. That scope matters because founders can easily hear “Google ad-tech ruling” and hallucinate a much larger greenfield than reality supports. FindLaw
Where This Idea Came From
The origin story starts in January 2023, when the Justice Department and plaintiff states sued Google over ad technology. The case page puts the complaint on January 24, 2023. That complaint set up the central human fight: regulators and publisher allies arguing that Google controlled too many steps of the transaction, Google arguing that stronger remedies would break useful tools, and independent ad-tech companies hoping a court would weaken Google’s advantage without wrecking the market around them. Justice Department
The board state changed on April 17, 2025, when Judge Brinkema found Google had monopolized the publisher ad server and ad exchange markets for open-web display advertising and had unlawfully tied Google’s publisher ad server and exchange products. That liability finding did two things at once. The liability finding gave critics a legal win. The liability finding also sharply narrowed the actual opportunity, because the harmed markets were specific plumbing markets inside open-web display. FindLaw
The loudest public prophet of “Google should recede” has been Jeff Green of The Trade Desk, who said at an Interactive Advertising Bureau leadership meeting in January 2025, “I do think that it makes sense for Google to exit the open internet and to focus on YouTube.” Jeff Green is not a neutral umpire. Jeff Green is one of the best-known champions of the “open internet versus walled gardens” story, and Jeff Green’s company benefits if more demand flows outside Google’s own channels. The quote still matters because the quote captures the fantasy many founders immediately feel when Google gets constrained: maybe the open web is back. Axios
Then the remedy landed, and the remedy was much smaller than the fantasy. On September 2, 2026, the court rejected breakup and ordered operational changes instead. Google’s Lee-Anne Mulholland quickly framed the result as a win against disruption: “We’re very pleased the Court rejected the DOJ’s proposal to break apart tools that help small businesses reach new customers and grow.” Google’s position is not just public relations. Google’s position is the practical sales obstacle any startup faces: many publishers already run on Google, know the pain of changing ad setups, and will ask why a new vendor should touch any revenue-sensitive workflow at all. TechCrunch
Two weeks later, the Justice Department made the opening concrete. The Justice Department said Google must support AdX-Prebid and DFP-Prebid integrations, let rival ad servers receive AdX bids, give publishers more access to Google-owned data, and stop letting Google Ads bid preferentially into Google-owned tools because Google owns those tools. That paragraph is the entire “why now” thesis. Without that paragraph, most of this startup conversation collapses. With that paragraph, a founder can at least point to an actual forced change in the pipes. Justice Department
Rajeev Goel of PubMatic became one of the clearest commercial evangelists for the opening. AdExchanger reported Rajeev Goel saying the AdWords part of the remedy “could open one of the market’s largest pools of buy-side demand to competition on the merits of technology and performance.” Rajeev Goel also wrote in a 2025 PubMatic post that the court had confirmed Google “unlawfully acted as a monopolist, rigging a marketplace that should have supported content creators by rewarding innovation and fair competition.” Rajeev Goel’s incentives are obvious. PubMatic wants more contestable demand. Rajeev Goel still matters because Rajeev Goel names exactly where incumbents think value could move. AdExchanger
Jason Kint of Digital Content Next added the publisher-side pressure. Digital Content Next argued after the ruling that the remedies show “why Congress must act,” which is a revealing reaction. Publisher advocates saw progress, but publisher advocates did not see a finished victory. A founder reading the room should hear two things in that reaction: publishers still want more leverage, and publishers still do not trust the current market enough to relax. Public Now
There is also a useful self-tension in Jeff Green’s rhetoric. In January 2025, Jeff Green said Google should “exit the open internet.” In 2026, Adweek summarized Jeff Green saying 2026 would be the “best year yet” for the open internet. The tension is not hypocrisy. The tension is timing. Jeff Green’s bullish story gets stronger if Google retreats hard, but Judge Brinkema did not force a retreat that hard. The remedy may improve the market without delivering the cleaner world Jeff Green’s broad argument seems to want. Adweek
Who Wants to Build It
The builder signal is thinner than the legal signal. We did not find a big visible founder wave racing to build “post-remedy Google openings” as a category. That absence is a signal by itself. If a giant clean startup category had truly appeared overnight, more named founders would already be planting flags.
What does exist is edge activity. Indie Hackers featured Ad Control under the theme “building an ad platform shouldn’t take a year,” with a team member drawing on experience from Google’s ads organization. That is not proof of breakout demand. That is proof that builders still see pain inside ad operations and ad-serving workflows. Indie Hackers
BiddingStack has been publishing around a “Prebid Sales Agent,” which points in the same direction: not “we built the next empire,” but “we built a helper around the existing open-source standards layer.” The standards layer matters because the court remedy explicitly names Prebid. A lot of founder energy is likely to cluster around tools that help operators navigate the standards layer, not around trying to replace every major incumbent. BiddingStack
Other visible vendors sit in the same broad wedge: Setupad, Pubstack, Asteriosoft, Freestar, and similar firms frame value around wrappers, managed optimization, contextual targeting, analytics, or a publisher operating system. That lineup is the real market shape. Founder activity exists, but founder activity mostly aims at making ad operations less painful, not at creating a whole new market from scratch. Freestar
Scorecard
#### Time-to-Truth — 1/5 Time-to-Truth means how fast a founder can get a trustworthy answer to “should this business exist?” Here the answer is painfully slow. A believable yes or no probably takes 4 to 9 months and roughly $25,000 to $120,000 because publishers usually want live traffic tests, side-by-side revenue comparisons, legal review, and integration work before trusting any “we can improve yield” claim.
#### Time-to-MVP — 2/5 Time-to-MVP means how fast a founder can build the first version somebody can actually use. A narrow observability or migration tool may take 2 to 5 months. Anything that touches auction routing, bidding intelligence, or production integrations across Google Ad Manager, DoubleClick for Publishers, AdX, Prebid, and multiple sell-side platforms can stretch to 6 to 12 months. The integration surface is the product.
#### Time-to-First-Customer — 1/5 Time-to-First-Customer means how long before money arrives from a real buyer. Expect 4 to 10 months for a cold founder. Only founders with warm publisher relationships already in hand are likely to compress the path to 2 to 6 months. Mid-market publishers and monetization teams need trust, proofs, and usually a pilot before paying.
#### Capital Requirements — 1/5 Capital Requirements means how much cash the founder personally needs and how much total funding the business likely needs. The founder-funded wedge is roughly $40,000 to $150,000 personally invested for 6 to 12 months of cloud costs, logging, security basics, contract review, travel, and contractor help. A business robust enough to matter likely needs $500,000 to $2,500,000 across 12 to 24 months for one to two ad-tech engineers, one publisher-solutions or ad-operations lead, infrastructure, security, and runway. Every one of those numbers is an estimate, but the shape is clear: too heavy for casual bootstrapping.
#### Execution Complexity / Operational Risk — 1/5 Execution Complexity means how much ugly reality stands between “works in demo” and “works in a customer’s money flow.” The burden is high. Integrations are brittle. Customer workflows are revenue-sensitive. Data handling raises legal and security concerns. The target is moving because remedy implementation is still unfolding. If a product lowers revenue or adds too much delay to auctions, reputation dies fast.
#### Competitive Intensity — 1/5 Competitive Intensity means how crowded and defended the target space already is. Magnite, PubMatic, OpenX, Index Exchange, Amazon’s Transparent Ad Marketplace and Unified Ad Marketplace, Prebid-linked vendors, and managed monetization firms already operate at these seams. A startup does not get an open field. A startup gets one narrow wedge inside somebody else’s map. Magnite 10-K
#### Market Momentum — 2/5 Market Momentum means whether the market wind is helping or fighting the founder. Open-web display is still large, and the legal remedy creates a near-term tailwind. But the stronger growth stories in digital advertising have been shifting toward connected television, often shortened to CTV, and retail media rather than classic open-web display plumbing. The market is alive, but the hottest growth is elsewhere. Digiday
#### AI Leverage — 2/5 AI Leverage means whether AI meaningfully improves the economics or defensibility of the business. AI can help with anomaly detection, report summaries, support work, and ad-operations copilots. AI is much less likely to be the core edge in the live auction path, where latency and integrations dominate. AI helps compress labor more than AI creates a moat.
#### Defensibility — 2/5 Defensibility means how likely a lasting moat is. A moat can emerge if a company becomes the neutral control plane for publisher revenue teams and accumulates cross-publisher benchmarking data. That moat only appears after distribution and data volume. Early versions are more likely to be copyable software-plus-service wedges than hard technical monopolies.
Distribution compounding: weak. One customer does not make the next customer dramatically cheaper early on. Publisher sales are trust-based, proof-based, and relationship-heavy.
Base rating: 1 star. Real opening, wrong shape.
Cold-start rating: 1 star. A cold founder pays a large relationship-building tax before learning much.
Your money: founder personal investment of $40,000 to $150,000 versus total business capital need of $500,000 to $2,500,000 over 12 to 24 months (estimate).
Time to break-even: 18 to 36 months to cover costs, assuming 8 to 20 customers paying roughly $25,000 to $100,000 annual contract value or equivalent managed-service fees, and assuming gross margin stays software-heavy rather than exchange-heavy (speculative estimate).
The audience you’d need first: about 200 warmly reachable humans — meaning mid-market publisher revenue leaders, ad-operations managers, monetization consultants, and a few design-partner publisher groups that the founder can already reach directly or through trusted second-degree introductions.
What moves the needle: for a cold stranger, the rating stays 1 star. With an audience already trusted by publisher revenue teams, the rating rises to 2 stars. With capital alone, the rating rises to 2 stars. With both audience and capital, the ceiling is still 2 stars. No amount of money or audience fixes the fact that the opening is narrow, mature, incumbent-heavy, and slow to validate.
The person who wins here is: an ex-publisher monetization lead, ad-operations operator, or senior ad-tech integration person who knows buyers by name, can sell audits before product, tolerates ugly workflow details, and has enough runway for slow proof cycles.
The person who’ll be tempted but shouldn’t be is: the smart generalist software founder or consumer-AI founder who sees “Google got constrained” and imagines a fresh greenfield, but dislikes service work, long sales, and skeptical revenue operators.
#### What It Would Take to Raise This Score
The fastest honest kill test is 30 days long. Talk to 20 publisher-side ad-operations or revenue leaders and pitch one narrow paid pilot, such as neutral routing observability or migration benchmarking. Require 5 letters of intent or paid pilots at $5,000 to $15,000 each. If fewer than 2 buyers will pay or commit, even after concrete workflow discussion and sample outputs, the opening is probably being absorbed by incumbents or is not painful enough to support a new vendor.
Go-to-Market
The best channels are direct outreach, partnerships, and industry associations or events.
1) Cold outreach / sales — strong fit. The buyer list is small and specific: publisher revenue leaders, heads of ad operations, and monetization consultants. Early cost per serious pilot is mostly founder time plus roughly 10 to 30 warm conversations and $1,000 to $5,000 in travel, event, and sales-tool costs per early customer.
2) Partnerships / co-marketing — strong fit. Partnerships with monetization consultants, Prebid implementers, analytics vendors, and publisher advisory firms borrow trust from people who already sit in buyer workflows. Effort is high relationship work, but the trust transfer is real.
3) Associations, trade shows, events, and integration directories — strong fit. Publisher associations and ad-tech events concentrate the exact humans who compare vendors, share horror stories, and trade referrals. Integration directories matter because the buyer wants proof that the tool fits into the standards layer and existing stacks.
In the first 90 days, weeks 1 through 3 should be 20 to 40 interviews with publisher ad-operations and revenue teams. Weeks 4 through 6 should pre-sell 3 to 5 paid audits or design-partner pilots around a promise like: “We will compare your current Google Ad Manager, AdX, and Prebid routing and identify 1 to 3 revenue or workflow wins in 30 days.” Weeks 7 through 12 should turn repeated findings into a narrow dashboard, migration report, or experiment-management product.
One thing a day: message one publisher operator; review one auction workflow; publish one sharp technical observation; ask one partner for one introduction; or turn one pilot result into one case study.
Considered and set aside: search engine optimization and content, generative-engine optimization, paid search, paid social, influencer marketing, own podcast or YouTube, newsletter cross-promotion, press, Product Hunt, awards, app stores, generic listing sites, local press, built-in virality, affiliates, direct messages as a main motion, freemium as a main motion.
Illustrative Case Study — not a real company
A former publisher monetization director leaves a mid-size digital media company after spending six years inside yield meetings. The former publisher monetization director knows exactly what annoys heads of ad operations: every vendor promises lift, almost no vendor makes apples-to-apples comparisons easy, and every routing change feels like touching the oxygen supply. The former publisher monetization director does not pitch a miracle. The former publisher monetization director pitches a paid 30-day audit for $10,000 to compare current Google Ad Manager and AdX behavior against one alternative path and one Prebid setup.
Three publishers take the call. One publisher buys. The first month is ugly. The product is mostly spreadsheets, exports, a simple dashboard, and too many manual checks. But the audit finds one repeated workflow leak: the publisher cannot cleanly compare which routing path wins after latency and setup mistakes are considered. The founder turns that pain into a tiny product: a neutral experiment log, revenue-comparison dashboard, and migration checklist.
The business is alive at that moment, but still dangerous. The founder now needs five more publishers without breaking trust, and every future sale still depends on proof, references, and integration tolerance. The lesson is not “startup unlocked.” The lesson is “small hard business maybe unlocked.”
The Skeptic’s Case
The strongest reason not to build this business is that founders will confuse a legal headline with a startup category. The remedy is real. The remedy mostly creates interoperability work, compliance work, migration work, and benchmark work inside a mature market. Mature markets with entrenched workflows are not friendly to cold founders.
Another reason not to build the business is buyer psychology. Publishers do not buy ideology. Publishers buy measurable revenue lift, labor savings, risk reduction, or some combination. A founder cannot walk into a sales conversation saying “the court changed the rules” and expect a contract. A founder needs evidence that a workflow changed for the better. That evidence usually means pilots, traffic allocation, analytics work, and uncomfortable side-by-side tests.
Another reason not to build the business is incumbent absorption. Magnite reported 2025 revenue of $668.2 million. PubMatic reported fourth-quarter 2025 revenue of $85.5 million. The Trade Desk reported 2025 revenue of $2.9 billion. Those companies, plus OpenX, Index Exchange, Amazon, and service-heavy publisher partners, already have buyer relationships, engineering teams, and distribution into the exact seams the court just opened. Magnite Investor Relations
We also could not honestly verify a broad founder rush into the exact post-remedy niche. We found some edge signals and relevant builders, but not a thick field of named founders building in public around this precise wedge. That absence weakens the “obvious breakout category” thesis.
The Thesis-Killer Alternative
The easier and cheaper bypass is simple: publishers do nothing new. Publishers let existing stack vendors, managed monetization partners, or the open-source Prebid ecosystem absorb the remedy.
That bypass is powerful because the core costly requirement here is not raw software creation. The core costly requirement is trusted integration into a live money flow. Existing vendors already have the relationships, support teams, and integration surfaces. If the incumbent or service partner can say, “we already handle your wrapper, demand paths, and reporting; we will update the setup as the remedy takes effect,” the startup loses the main wedge before the startup even starts.
Does the bypass kill the opportunity? Mostly yes for any broad startup thesis. Not completely for a narrow tooling thesis. What survives the bypass is neutral observability, migration tooling, experiment management, fairness or compliance analytics, and cross-vendor benchmarking. What does not survive well is the fantasy of building “the next great independent ad-tech platform” just because Google got constrained.
Business Model, MVP, Time-to-Truth Test, Risks, Reasons Not to Pursue
The cleanest business model is software plus service. Start with a paid audit or pilot at $5,000 to $15,000. Turn recurring pain into a dashboard or workflow product priced at $25,000 to $100,000 annual contract value, with optional managed support. Avoid taking exchange risk early. Avoid becoming another opaque middleman early.
The best minimum viable product is not a new exchange or a full ad server. The best minimum viable product is a narrow observability or migration tool: compare path A against path B, show where discrepancies sit, document setup changes, and help the publisher decide whether a routing or configuration change is worth making.
The time-to-truth test is brutally concrete: can the founder get 5 scoped pilots or letters of intent from 20 interviewed buyers inside 30 days? If not, the market is probably not starving for a new vendor.
The risks are plain. Google may comply narrowly. Google may appeal. Publishers may avoid churn. Incumbents may ship the same helpful features first. The strongest secular ad-spend growth may continue to move toward connected television and retail media instead of the publisher display plumbing where this remedy bites hardest.
Reasons not to pursue are even plainer. Do not pursue the business if the founder lacks warm publisher relationships. Do not pursue the business if the founder hates services. Do not pursue the business if the founder wants product-led growth, fast loops, or mass-market upside. Do not pursue the business if the founder’s only thesis is “Google got hit, therefore greenfield.”
The Full Top 3 Takeaways
The court opened a door, but the door leads into a crowded boiler room. The remedy is real and specific. The Justice Department says Google must support AdX-Prebid and DFP-Prebid integrations, allow more data access, and stop preferential Google Ads bidding into Google-owned tools. That creates workflow demand. The same remedy is behavioral, not structural. A founder gets a wedge in a mature subsystem, not a reset of the industry. Justice Department
Publishers will not buy a political story; publishers will buy a revenue lift story. Every hard part of the business sits downstream of proof. Buyers need measurable revenue gains, labor savings, or lower risk. That means pilots, side-by-side tests, trust, and references. The founder who wins here behaves like a surgeon and consultant first, software founder second.
The best business here is probably an ugly audit-and-migration tool, not a glamorous new ad empire. The strongest surviving wedge is neutral control-plane work: observability, migration support, experiment management, cross-publisher benchmarking, and compliance evidence. A software-plus-service company could become durable in that lane. A generic “new ad-tech winner” thesis almost certainly dies under incumbent pressure.
The real unresolved question is the only question that matters now: when Google’s forced integrations and anti-preference rules hit real publisher workflows, will meaningful revenue actually reroute into rival paths, or will the existing vendors and service partners quietly absorb the whole change before any startup becomes trusted enough to matter?
Provenance: Super Fluidity Opportunity Dossier, method version current as of 2026-09-27. AI-assisted report with public-source verification and confidence hedging. Facts are source-verified where cited; single-source and estimate-level claims are presented as such.
Why five ratings? Because the same idea is a different bet for different people. As a side gig — no money, no audience, no special skill, a few hours a week beside a job — the question is simple: is there a realistic path to $2,000–$5,000 a month within a year? As a business, starting with no money and no audience, most ideas are traps — even ideas that became famous companies rate only 2 stars from that starting line. With money, an audience, or both, the same idea can be a real opportunity. So every dossier rates all five, and you can find your own row.
Dossiers rated 0–2 stars for a founder with no money or audience are published free, right here. The rarer 3-, 4-, and 5-star finds go to paid subscribers first — that tier is opening soon, and free subscribers get the first invitation. Join the free list. Or have us investigate your own idea, privately: Request a Dossier ($199).
