Will AI Wrappers Survive? (28 Data to Understand)

We research AI Wrappers every day, if you're building in this space, get our report

Building an AI wrapper looks tempting when you see founders posting their revenue screenshots on X.

But the reality behind those numbers tells a different story.

We compiled 28 critical data points from hundreds of AI wrapper companies to understand if this business model can actually survive long-term. You can find the complete analysis in our 200-page report on AI wrappers.

In our 200+-page report on AI wrappers, we'll show you the real user pain points that don't yet have good solutions, so you can build what people want.

Will AI Wrappers Survive or Are They an Unsustainable Business Model?

• 1. 80-95% of AI wrapper businesses fail completely

  What the data shows:

  Only 5-20% of AI wrappers survive to generate meaningful revenue. Multiple sources confirm this: RAND Corporation reports 80%+ of AI projects fail, MIT's 2025 NANDA study found 95% of generative AI pilots at enterprises are failing, and S&P Global data shows 42% of companies abandoned most AI initiatives in 2024 (up from 17% in 2023).

  Why this matters for AI wrapper durability:

  This high failure rate comes from three main issues. First, anyone can build an AI wrapper in days, flooding the market. Second, features get copied fast. Third, you're completely dependent on companies like OpenAI who can change their API or add your feature anytime. Most wrappers run out of money or get replaced before they can build a real business.

  Source: Market Clarity

• 2. Just 2-5% of AI wrappers reach $10K monthly revenue

  What the data shows:

  Even among survivors, very few reach meaningful revenue. Only 30-40% of AI wrapper attempts make any money at all. Among those that do, just a tiny fraction hit $10,000+ per month. Traditional SaaS businesses do better: about 4% reach $1 million yearly, though it takes 2.5+ years.

  Why this matters for AI wrapper durability:

  The low success rate shows that most AI wrappers can't build something people will consistently pay for. Even at $10K/month, most companies lose money or barely break even. This creates a dangerous period where startups burn through their savings before they can make the business work.

  Source: Market Clarity

• 3. 10-15 new AI wrappers launch every single day

  What the data shows:

  The market has 15,000-25,000 AI wrapper products right now. 70-105 new ones launch every week. By end of 2026, we'll likely see 35,000-50,000 wrappers. Most take just 10-30 days to build. Some launch in under 48 hours.

  Why this matters for AI wrapper durability:

  When thousands of products do the same thing, standing out becomes nearly impossible. Your advantage lasts maybe 3-12 months before competitors copy your features. Categories like "chat with PDF" or content generation are packed with identical products. To survive long-term, you need either unique distribution channels or deep expertise in a specific niche, not just good technology.

  Source: Market Clarity

• 4. VCs cut investable AI companies from 75% to 40%

  What the data shows:

  Meera Clark, Principal at Redpoint Ventures, observed a dramatic shift in mid-2023. Three to four months prior, VCs believed 75-80% of new AI companies were worth investing in. That number dropped to about 40% as investors realized most companies lacked real defensibility. U.S. AI seed deals fell from $295 million in March 2023 to $179 million in July 2023.

  Why this matters for AI wrapper durability:

  Investors now see that most AI wrappers can be destroyed overnight when companies like OpenAI add competing features. This makes it harder for wrappers to raise money. Even promising companies struggle to get funding for their next round. Without investor money, more wrappers will fail faster.

  Source: Fortune


In our 200+-page report on AI wrappers, we'll show you which areas are already overcrowded, so you don't waste time or effort.

• 5. AI wrappers achieve 25-60% gross margins vs 70-80% SaaS

  What the data shows:

  According to Bessemer Venture Partners' 2025 dataset covering hundreds of companies, fast-growing AI startups average about 25% profit margins early on. Many show negative margins during growth. Well-optimized companies hit 60%+ through smart cost management and pricing. Traditional SaaS companies routinely achieve 70-80% margins.

  Why this matters for AI wrapper durability:

  This 10-25 point disadvantage directly hurts long-term profits. Lower margins mean less money for sales and marketing or slower growth. To match traditional SaaS profitability, wrappers must either charge premium prices or run extremely efficiently. Both are hard in competitive markets.

  Source: Tanay Jaipuria

• 6. AI wrappers show -58% to +13.5% margins at $100K revenue

  What the data shows:

  At $100K monthly revenue, most AI wrapper companies lose money. Well-optimized companies with 60% margins might structure costs as: $40,000 for API costs (40%), $35,000 for fixed costs (35%), leaving $25,000 profit (25%). But companies with typical 40% margins and higher fixed costs lose substantial money at this level.

  Why this matters for AI wrapper durability:

  You need significant scale to become profitable. Most startups run out of money before hitting $100K-250K monthly revenue where the math finally works. This forces you to rely on investor funding, which is getting harder to secure. We analyze profitable pricing strategies in our report covering the AI Wrapper market.

  Source: Market Clarity

• 7. LLM API prices dropped 83% in just 16 months

  What the data shows:

  From March 2023 (GPT-4 launch) to July 2024, OpenAI cut GPT-4o pricing by 83% for output tokens and 90% for input tokens. Prices went: GPT-4 ($60/$30 per million tokens) to GPT-4 Turbo ($30/$10) to GPT-4o ($15/$5) to GPT-4o updated ($10/$3). FutureSearch's analysis shows OpenAI's own profit margins dropped from 75% in June 2024 to 55% after August 2024.

  Why this matters for AI wrapper durability:

  Falling API costs sound good, but they create constant pressure. When prices drop over 50% yearly, you must keep lowering your prices to stay competitive. This makes maintaining margins extremely hard. Even worse: if OpenAI with massive scale sees margins fall to 55%, wrapper businesses face existential pressure as their biggest cost becomes commoditized.

  Sources: Cursor, FutureSearch

• 8. High churn yields LTV of just $288-400 in first year

  What the data shows:

  With typical pricing of $20-50/month, 80% profit margins, and 3.5% monthly churn (34-46% yearly), first-year customer value calculates to about $288-400. Full lifetime value at 28.6 months averages $686, but most value comes in year one. Getting a customer costs $50-200 for consumers and $200-500 for small businesses. You need at least a 3:1 value-to-cost ratio to be sustainable.

  Why this matters for AI wrapper durability:

  Many AI wrappers operate at 1:1 or worse ratios, meaning they spend as much (or more) to get customers as those customers are worth. High churn (2-3x worse than enterprise SaaS) creates a leaky bucket where you need constant new customers just to maintain revenue. Only wrappers with under 3% monthly churn and higher pricing ($99-299/month) can make the math work.

  Source: Market Clarity

• 9. Power users create 100:1 cost variance destroying profitability

  What the data shows:

  Monthly costs per user vary wildly: Light users cost $1-7, Medium users $17, Heavy users $51, and Power users $169-270. Anthropic reportedly lost "tens of thousands per month" on some $200/month users. Some startups got $800,000 cloud bills on just $250K yearly revenue.

  Why this matters for AI wrapper durability:

  This unpredictable cost structure creates huge risk for fixed-price subscriptions. A small group of power users can destroy your entire profit margin. You must either add usage limits, charge based on usage, or risk financial disaster. But usage-based pricing makes fewer people sign up and increases cancellations.

  Source: Tanay Jaipuria


In our 200+-page report on AI wrappers, we'll show you which ones are standing out and what strategies they implemented to be that successful, so you can replicate some of them.

• 10. OpenAI lost $5B on $3.7B revenue in 2024

  What the data shows:

  OpenAI lost about $5 billion in 2024 on $3.7 billion revenue (losing $1.35 for every dollar made). Anthropic lost $5.3 billion. Some analyses suggest AI companies collectively lose "nearly $40 for every dollar they make." OpenAI's projected losses could hit $115 billion by 2029.

  Why this matters for AI wrapper durability:

  If even massively-funded leaders like OpenAI and Anthropic with direct access to AI models lose money at scale, it shows current AI business models might be fundamentally broken without continuous investor money. For wrapper businesses, this creates a crisis: if your suppliers are unprofitable at scale, wrappers built on top have virtually no path to profitability.

  Source: Medium

• 11. Jasper's revenue collapsed 54% from $120M to $55M ARR

  What the data shows:

  Jasper AI, one of the biggest early AI wrapper success stories, peaked at $120 million yearly revenue in 2023 before dropping to $55 million in 2024. Despite raising $131 million at a $1.5 billion valuation (October 2022), hitting 85% customer retention, and serving 100,000+ customers, revenue fell dramatically. The company cut its valuation by 20%, revised forecasts down 30%, did layoffs in July 2023, and replaced the CEO.

  Why this matters for AI wrapper durability:

  Jasper's collapse is the defining warning for AI wrappers. Despite apparent advantages (brand recognition, large customer base, significant funding, strong retention), the company couldn't compete when OpenAI improved ChatGPT and launched ChatGPT Plus at $20/month versus Jasper's $80/month. Even market leaders with substantial resources face existential threats when platform providers launch competing features.

  Sources: Latka, Maginative

• 12. 12-month product head start now equals three prompts

  What the data shows:

  According to venture firm Virta, "a 12-month product head start in 2022 might equal just three well-engineered prompts and a polished interface today for many workflow products." What once took months to build now gets replicated in days or weeks. Foundation model pricing collapsed 98% within a single year, the "fastest technology commoditization cycle we've ever seen."

  Why this matters for AI wrapper durability:

  Competitive advantages now last weeks instead of years. Traditional startup strategies relying on first-mover advantage don't work anymore. AI wrappers must execute extremely fast while building deeper advantages beyond core AI features (workflow integration, proprietary models, distribution). You have 3-6 months to establish a defensible position before competitors copy everything, compared to 12-24 months for traditional SaaS.

  Source: Virta Ventures

• 13. AI reduced data migration costs to near zero

  What the data shows:

  Traditional software switching costs built on data migration complexity have been eliminated by AI. What previously required tedious manual work or 200+ developer hours costing $20,000+ now takes "seconds of compute time." Hand ChatGPT an export file from one app, and seconds later you get an import file ready for another app.

  Why this matters for AI wrapper durability:

  AI wrappers can't rely on traditional lock-in that kept customers trapped. Data portability is now essentially free. You must build retention through deep workflow integration, proprietary data ownership, or real network effects rather than artificial barriers. The irony: the same AI technology enabling wrapper businesses also destroys their ability to retain customers. We detail effective retention strategies in our 200-page report covering everything you need to know about AI Wrappers.

  Source: Fillout

• 14. Customer support AI tools suffer 76% annual churn

  What the data shows:

  Monthly churn varies wildly by category: Customer support and chatbots see 6-12% monthly churn (53-76% yearly), Marketing and CRM tools see 3-7% monthly (31-58% yearly), while Financial and Fintech AI achieves 2-5% monthly (22-46% yearly). Healthcare and HR tools fall in between at 3-8% monthly.

  Why this matters for AI wrapper durability:

  Domain expertise and compliance provide much stronger retention than pure AI features. Financial tools show 2-3x better retention than customer support tools. AI wrappers in crowded horizontal categories (chatbots, content generation) face the worst retention and thus poorest survival odds. Success depends on deep vertical specialization in domains with real barriers beyond the AI model itself.

  Source: LiveX AI

• 15. Data value decreases as AI corpus grows larger

  What the data shows:

  Research from Andreessen Horowitz shows that for most AI applications, "the cost of adding unique data to your corpus may actually go up, while the value of incremental data goes down." Data network effects have "limited value as a defensive strategy" for enterprise startups, with defensibility often eroding as the data corpus grows rather than strengthening.

  Why this matters for AI wrapper durability:

  AI wrappers claiming data advantages face growing investor skepticism. Foundation models use mostly public data, and private data adds limited value in most cases. True defensibility requires either: (1) exclusive access to proprietary data that can't be replicated, (2) domain-specific data that materially improves performance in specialized niches, or (3) rapid accumulation of usage data creating compounding advantages before competitors catch up.

  Source: Andreessen Horowitz


In our 200+-page report on AI wrappers, we'll show you the real challenges upfront - the things that trip up most founders and drain their time, money, or motivation. We think it will be better than learning these painful lessons yourself.

• 16. OpenAI PDF support killed multiple $500K+ MRR businesses

  What the data shows:

  In October 2023, OpenAI added native PDF support to ChatGPT, immediately threatening startups like PDF.ai (making $500,000+ monthly), ChatPDF, and AskYourPDF. One founder reported that after polling users, most said his plugin would "see less usage." ChatOCR reported being a "victim" with major usage drop. The November 2023 GPT Store launch let anyone create custom GPTs for $20/month, eliminating the need for specialized wrapper apps.

  Why this matters for AI wrapper durability:

  These documented events show the existential platform risk in wrapper businesses. OpenAI systematically identifies successful use cases built on their APIs, copies those features directly, and offers them cheaper or free, destroying the wrapper's value. Sam Altman explicitly warned: "If you're just wrapping GPT-4, we're going to steamroll you." No defensive advantage exists for simple API wrappers when the platform can replicate features at will.

  Sources: Gizmodo, TechCrunch

• 17. $55.3B in AI M&A deals in first 7 months of 2025

  What the data shows:

  AI M&A deal value hit $55.3 billion in January-July 2025, an 11% increase over all of 2024 ($50B total). Deal volume reached 240 transactions, already exceeding 50% of 2024's total. Between 2014-2023, CSET Georgetown tracked 4,354 total AI transactions, with annual deals more than doubling from 225 in 2014 to 494 in 2023. Big Tech leads: Apple (28 AI acquisitions), Alphabet (23), Microsoft (18), and Meta (16).

  Why this matters for AI wrapper durability:

  Rapid consolidation by tech giants creates multiple threats to independent wrappers. Large platforms systematically acquire capabilities and integrate features that eliminate wrapper value. The concentration among top companies shows they're absorbing AI capabilities rather than partnering with independents. While this creates potential exit opportunities, most acquisitions are talent-focused "acqui-hires" at modest valuations rather than validation of wrapper business models.

  Sources: CSET Georgetown, PYMNTS

• 18. 80% of Fortune 500 adopted ChatGPT in 9 months

  What the data shows:

  Over 80% of Fortune 500 companies adopted ChatGPT within the first nine months of launch. OpenAI's revenue split shifted from 75% consumer to about 50/50 consumer/enterprise, showing rapid enterprise capture. ChatGPT Enterprise launched in August 2023 with SOC 2 compliance, data privacy guarantees, and 2x faster performance.

  Why this matters for AI wrapper durability:

  Platform providers have overwhelming enterprise advantages: established brand trust, existing relationships, compliance certifications, pricing power through scale, and feature breadth that wrappers can't match. The 80% Fortune 500 adoption in just nine months shows enterprises rapidly consolidate to platform providers rather than using multiple point solutions. Generic horizontal wrappers attempting enterprise sales risk losing deals mid-cycle when platforms launch similar features.

  Source: Andreessen Horowitz

• 19. $1.2B valuation Builder.ai filed bankruptcy in 2025

  What the data shows:

  Builder.ai raised $445 million with Microsoft backing but filed for bankruptcy in May 2025 after claiming $220 million in 2024 revenue when actual revenue was only $55 million (300% exaggeration). Creditor Viola Credit seized $37 million from accounts, leaving just $5 million. The company owed $85M to Amazon and $30M to Microsoft, having employed 1,500+ people at peak before laying off 770 employees.

  Why this matters for AI wrapper durability:

  This spectacular failure shows that even heavily-funded AI wrappers with major corporate backing can collapse when built on inflated metrics and weak technology. Builder.ai was actually using hundreds of engineers in India and Ukraine rather than genuine AI automation, essentially a services business pretending to be a technology platform. The case shows wrapper valuations often exceed fundamental business value, creating unsustainable expectations that lead to catastrophic failures.

  Sources: TechStartups, Bloomberg


In our 200+-page report on AI wrappers, we'll show you dozens of examples of great distribution strategies, with breakdowns you can copy.

• 20. One-third of global VC funding went to AI in 2024

  What the data shows:

  Total AI funding in 2024 exceeded $100 billion (up 80%+ from $55.6 billion in 2023), representing nearly one-third (32%) of all global venture funding ($314 billion total). However, funding was highly concentrated. In Q4 2024, roughly half of late-stage funding went to just three companies: Databricks ($10B), OpenAI ($6.6B), and xAI ($6B). Among the 49 U.S. AI startups that raised $100M+ in 2024, funding went primarily to infrastructure and foundational model companies.

  Why this matters for AI wrapper durability:

  While AI receives massive investment, the concentration in foundational models and infrastructure rather than applications suggests VCs are betting on underlying technology rather than wrappers built on top. This creates a capital drought for wrapper businesses: abundant funding exists for AI broadly, but application-layer companies struggle to raise follow-on rounds as investors grow skeptical of defensibility. The funding imbalance will accelerate wrapper failures as infrastructure companies are well-capitalized while wrappers face money constraints.

  Sources: Crunchbase, TechCrunch

• 21. Copy.ai achieved 480% revenue growth with only $17M raised

  What the data shows:

  Copy.ai grew revenue from $42K (2020) to $23.7M (2024), achieving 480% growth in 2024 alone with 4+ consecutive months of 20%+ expansion. The company reached 16 million users by February 2025, having raised only $16.9M total, far less than competitors like Jasper ($131M). Success came from rapidly pivoting from simple copywriting tool to enterprise Go-to-Market AI platform with complex workflow automation.

  Why this matters for AI wrapper durability:

  Copy.ai shows that AI wrappers can succeed by rapidly evolving beyond simple API wrappers into comprehensive workflow platforms with real product differentiation. The capital efficiency (achieving similar scale to Jasper with 1/8th the funding) suggests the winning strategy involves product focus and customer value rather than aggressive marketing spend. This proves that wrappers adding differentiated features beyond base models and focusing on workflow integration can build durable businesses, but requires significant product sophistication.

  Sources: Latka, Copy.ai

• 22. Perplexity reached $100M ARR in 20 months after Pro launch

  What the data shows:

  Perplexity AI hit $100M yearly revenue by March 2025, just 20 months after launching Pro subscriptions, approaching $200M by September 2025. The company raised $665M total and was valued at $20B (September 2025), up from $9B (December 2024). With 22 million monthly users processing 400M queries monthly, Perplexity differentiated by adding citation features, building a proprietary search index, and using open-source models (LLaMA-2, Mistral) alongside commercial APIs.

  Why this matters for AI wrapper durability:

  Perplexity shows that AI wrappers can achieve massive valuations by building genuine proprietary technology layers (search index, multi-model routing, citation systems) rather than simply providing UI over third-party APIs. Success came from creating defensible differentiation that platform providers can't easily replicate. However, even at $20B valuation, Perplexity faces platform risk as Google and Microsoft invest heavily in AI-powered search, highlighting that even differentiated wrappers remain vulnerable to well-resourced competition.

  Sources: TechCrunch, Analytics India Magazine

• 23. FormulaBot reaches 87.5% profit margin on $500K ARR

  What the data shows:

  Founder David Bressler generates $42K monthly ($504K yearly) with 87.5% profit margins from 10,900 paid users out of 650K+ freemium users. Built the MVP in 6 weeks, made $20K in first 2 months, hit $3K monthly within 2 months, and scaled to $500K yearly by 2024. The exceptional margins result from very low token costs due to short formula text (converts text to Excel formulas rather than long-form content).

  Why this matters for AI wrapper durability:

  FormulaBot's exceptional 87.5% margins (approaching traditional SaaS levels) show that AI wrappers with simple, low-token-usage use cases can achieve favorable economics. However, this is a clear outlier. Most AI wrappers handling complex reasoning, long-form content, or multi-turn conversations achieve only 40-60% margins. The case shows that highly specialized, low-token-consumption use cases can build economically viable wrapper businesses, but this strategy only works for narrow applications.

  Source: Market Clarity

• 24. Harvey AI: $75M ARR and $5B valuation in legal AI

  What the data shows:

  Harvey AI achieved $75 million yearly revenue and a $5 billion valuation by focusing exclusively on legal workflows with deep domain specialization. The company built proprietary legal knowledge, specialized models, and workflow integrations specifically for law firms and legal departments. Harvey charges premium prices ($299+ per month) for specialized capabilities that general-purpose AI tools can't match, serving a niche that's substantial enough to support a large business but too specialized for platform providers to prioritize.

  Why this matters for AI wrapper durability:

  Harvey represents the clearest path to AI wrapper durability: deep vertical specialization in domains with meaningful expertise barriers, proprietary data, regulatory requirements, and workflow complexity that platform providers won't prioritize. By focusing on a specialized vertical with high willingness to pay, Harvey built genuine advantages that extend beyond the underlying AI models. This suggests AI wrappers can survive by serving niches too small or specialized for OpenAI or Anthropic to address directly, but requires domain expertise and substantial product sophistication. We analyze similar vertical strategies in our market clarity report covering AI Wrappers.

  Source: Andreessen Horowitz

• 25. PDF.ai: 400K users vs competitor's 4K users with better tech

  What the data shows:

  PDF.ai reported 400,000 users despite a competitor having "superior technology" (11+ AI models vs. PDF.ai's single model, 39+ document types supported) having only 4,000 users, a 100x difference attributed entirely to distribution advantages. Similarly, PhotoAI founder Pieter Levels attributes 50% of traffic to his 500,000+ Twitter following. Analysis shows distribution creates 10-100x revenue impact, mattering far more than technology quality.

  Why this matters for AI wrapper durability:

  This reveals that AI wrapper success depends more on distribution advantages (social media followings, SEO expertise, existing audiences, viral mechanics) than on technological superiority or product quality. Wrappers with unique distribution channels (influencer founders, established communities, superior growth marketing) can build sustainable businesses even with commoditized technology, while technologically superior products with weak distribution fail. However, distribution advantages can also erode quickly as platforms consolidate attention and SEO becomes saturated.

  Source: Market Clarity


In our 200+-page report on AI wrappers, we'll show you the best conversion tactics with real examples. Then, you can replicate the frameworks that are already working instead of spending months testing what converts.

• 26. SOC 2 Type 2 compliance costs exceed $100K total

  What the data shows:

  Achieving SOC 2 Type 2 compliance (required by most Fortune 500 companies) costs over $100,000 including: external CPA audit fees ($15,000-$50,000+), compliance platform subscriptions ($25,000-$100,000+ yearly), security tools, employee training ($10,000-$30,000), documentation and policy development ($20,000-$50,000), and ongoing monitoring ($20,000-$40,000 yearly). Additional costs include average data breach expenses of $4.88 million.

  Why this matters for AI wrapper durability:

  Compliance costs create a significant barrier to enterprise sales for early-stage wrappers, often representing 6-12 months of money. This creates a chicken-and-egg problem: you need compliance to win enterprise deals, but need enterprise revenue to afford compliance. Startups that fail to achieve SOC 2 or ISO 27001 lose access to 80% of Fortune 500 companies. The compliance burden significantly favors larger, well-funded players and platform providers who spread these costs across massive user bases.

  Source: Tiebreaker AI

• 27. Consumer AI wrappers reach $4.2M ARR vs $2M enterprise

  What the data shows:

  According to Andreessen Horowitz data across hundreds of companies, median consumer AI companies reach $4.2 million yearly revenue in their first year and raise Series A funding within 8 months after starting to charge. Median enterprise AI companies reach $2 million yearly in first year, raising Series A in 9 months, giving consumer wrappers a 110% revenue advantage initially. However, consumer wrappers face 46% yearly churn versus 34% for B2B.

  Why this matters for AI wrapper durability:

  This data reveals a critical tradeoff: consumer AI wrappers show higher initial revenue velocity but face significantly higher churn and lower lifetime values, while enterprise wrappers grow slower but build more durable businesses through contracts and lower churn. Consumer wrappers optimizing for rapid early growth create leaky bucket businesses that struggle with retention, while enterprise-focused wrappers that survive the slower ramp build more sustainable long-term models. However, both remain vulnerable to platform competition.

  Source: Andreessen Horowitz

• 28. Median AI wrapper survivors reach only $50K-$200K ARR Year 1

  What the data shows:

  Among the 5-20% of AI wrappers that survive and generate revenue, median performers reach just $50,000-$200,000 yearly after their first year, with top 25% hitting $500,000-$1M. Top performers (1-2%) reach $1M-$5M+ in year one. In the first 3 months, median performers make $0-$2,000 monthly while top 10% reach $15,000-$50,000 monthly. Stripe data shows AI startups reach $1M annualized revenue in a median of 11 months versus 15 months for traditional SaaS.

  Why this matters for AI wrapper durability:

  While top performers can achieve significant revenue quickly, the median survivor generates relatively modest revenue ($50K-$200K yearly), making long-term sustainability challenging. At this revenue level, companies operate at negative or barely-positive margins, creating a period where most run out of money. The bifurcated outcomes (median at $50K-200K vs. top performers at $1M-5M+) suggest AI wrappers face winner-take-most dynamics where category leaders capture disproportionate value while the middle struggles to survive.

  Source: Market Clarity

In our 200+-page report on AI wrappers, we'll show you the ones that have survived multiple waves of LLM updates. Then, you can build similar moats.