《硅谷銀行（SVB）：2025全球企業軟件行業狀況報告：趨勢概覽（英文版）（27頁）.pdf》由會員分享，可在線閱讀，更多相關《硅谷銀行（SVB）：2025全球企業軟件行業狀況報告：趨勢概覽（英文版）（27頁）.pdf（27頁珍藏版）》請在三個皮匠報告上搜索。

1、3Letter From Our Authors4Perspectives on the Enterprise Market5Four Themes To Watch in 20256Macro8Fundraising and Investment14Financials and Benchmarks17Spotlight:AI Special Contribution from21Exits“The transformative power of GenAI looms over the enterprise software ecosystem,but the extent of its

2、reach remains an open question.Trust,credibility and limiting bias are the keys to mass AI adoption.”3While disruption and change are constants of the innovation economy,true platform shifts come around only so often.Web adoption after the invention of the World Wide Web.Mobile following the release

3、 of the iPhone and Android.Cloud computing with the launch of AWS,Azure and Google Cloud.Today,we stand on the precipice of a seismic shift that may well overshadow the impact of these technological leaps.The transformative power of GenAI looms over the enterprise software ecosystem,but the extent o

4、f its reach remains an open question.Trust,credibility and limiting bias are the keys to mass AI adoption.Unsurprisingly,this emerging platform shift has created a frenzy of interest among investors.Roughly 40%of fundraising capital raised in 2024 came from funds that listed AI as a focus up from ju

5、st 10%in 2021.Meanwhile,AI companies commanded 45%of US venture capital(VC)investment in enterprise software startups an increase from 9%as recently as 2022.However,the bulk of this early investment has been directed toward capital-intensive hypergrowth and/or core AI businesses.More than three-quar

6、ters of capital invested into US AI startups came from rounds of$100M+.What do these numbers mean on the ground?For startups that arent raising mega-rounds(or at least not yet),the landscape remains challenging:lower Series A graduation rates,a higher bar for raising VC rounds across life stages and

7、 a continued emphasis on unit economics and smart growth.These companies feel pressure to create efficiency and solve mission-critical problems for a customer base that is increasingly reassessing overall budget and slow-rolling procurement decisions in the face of ongoing tariff threats and macroec

8、onomic uncertainty.The silver lining of this challenging landscape is that the companies carving out paths to raise capital and grow their businesses are remarkably resilient,with strong underlying metrics,enhanced product offerings addressing critical needs across the enterprise and established pro

9、duct-market fit.Much like trying to forecast the future of the web in 1998 or mobile in 2007,we are aware that the AI-native landscape is still in its infancy,and we are only just beginning to grasp what it might look like.Looking across our portfolio,we are extremely encouraged by the long-term via

10、bility of the current cohort of growth-stage software startups.Given the level of innovation,we are confident that these new startups will be the market leaders of tomorrow.Rob HelmSenior Market ManagerTech BankingSilicon Valley BAndrew McCartySenior Vice PresidentTech BankingSilicon Valley BSTATE O

11、F ENTERPRISE SOFTWARE 2025“Pinegroves broad reach across the fund and startup ecosystem gives us a unique lens into this pivotal moment for AI.On one end,our managers are energized by AIs potential to drive breakout opportunities from life sciences to financial services by augmenting human capabilit

12、y through intelligent agents that free teams to focus on higher-value work.On the other,theres caution around inflated valuations and echoes of past hype cycles.Still,with AI-native companies reaching scale at unprecedented speed,the impact feels both undeniable and enduring.”Dave MullenPartner“Sell

13、ing startup technologies to large enterprises is one of the toughest challenges because its all about trust,confidence and credibility which startups have to earn.The same applies to AI.Without explainability,human oversight and secure deployment,adoption stalls.Trust is the real barrier.The next wa

14、ve of innovation wont be driven by smarter models alone,but by smarter,more trustworthy delivery.”Darren KimuraChief Executive OfficerSTATE OF ENTERPRISE SOFTWARE 20254Ongoing tariff threats,market volatility and resulting macroeconomic uncertainty are changing the landscape for software startups.Th

15、is disruption will unfortunately leave no corner of the economy untouched,and software startups will not be immune as customers tighten budgets and stretch procurement cycles.Balancing this,though,is the reality that the software space is more insulated from direct tariff impacts than other industri

16、es,such as hardware-intensive industries tied to global supply chains.This environment also creates competitive opportunities for companies that can use their own resources more efficiently and demonstrate mission-critical value to customers.Much of the early excitement and capital has been directed

17、 toward large,name-brand,core AI businesses.However,AI company formation and disruption have permeated across all layers of the software stack,from the data layers and infrastructure,to the application layer,to the cybersecurity solutions seeking to secure organizations against evolving threats and

18、attack surfaces.AI already accounts for 35%of all US VC deal count activity.We expect this number to only grow and likely blur as AI becomes more ubiquitous.Think,for instance,of the evolution of mobile.Today,we dont draw distinctions between software companies with and without a mobile component.In

19、 the future,AI,like mobile,will be table stakes.While fundraising is normalizing to pre-pandemic levels,however,beneath the surface there is increased concentration of capital.At the same time,LP dollars are shifting toward funds focused on AI.Liquidity constraints also have LPs concentrating capita

20、l with established GPs that have proven track records.Theres broad recognition that AI is the next major technology wave following Web 2.0,mobile and cloud computing where venture historically generated outsized returns.As the pendulum swings in investors favor,GPs are selectively investing in the b

21、est of breed.This is leading to historically low Series A graduation rates and deal activity concentrated in AI-first companies showing product market fit.Exits and the liquidity they generate for founders,employees and investors have been the missing ingredient in recent years.The IPO market remain

22、s sluggish today due to market uncertainty,with only a few standout deals.However,M&A is gaining momentum.Cash runway has returned to pre-pandemic levels,and over half of US enterprise software startups are expected to raise capital in the next year.With investors being selective and a closed IPO wi

23、ndow,startups may need to explore alternative financing or exit options.Larger corporates with significant cash reserves are expected to invest in strategic technologies like AI,potentially driving further acquisitions.A recovery in exits is essential to reinvigorate venture returns and fuel the nex

24、t cycle of innovation.5STATE OF ENTERPRISE SOFTWARE 20256STATE OF ENTERPRISE SOFTWARE 2025This year has brought its own share of chaos heightened geopolitical tensions,interest rate uncertainty and public market volatility.Wading against the macro current is AI.A recent SVB poll of CFOs of VC-backed

25、 tech companies found that enterprise software startups ranked AI adoption as their top concern.This isnt to minimize other challenges such as a potential trade war,which could have knock-on effects ranging from reduced software spend to compressed margins.Rather,it highlights that despite a number

26、of possible challenges,one thing is certain:AI is the top priority.“What was done with cloud computing was amazing.But,at its core,that shift was about digitizing infrastructure.GenAI is digitizing intelligence,which is a paradigm shift more similar to the adoption of electricity or the internet,”no

27、tes Darren Kimura,CEO of AI Squared.This innovation,however,doesnt come cheap.Most new large language model(LLM)costs millions of dollars to develop,and foundational AI companies are churning out several per year.The metric that may best capture this activity is NVIDIAs revenue.The AI chipmaker has

28、cornered the market on semiconductors needed to train new models,and its sales are rising in proportion to public adoption of AI.Companies materially leveraging AI(i.e.,AI-exposed)are seeing a boost not only in revenue,but in popular operational metrics such as revenue per employee(RPE).While effici

29、ency metrics such as RPE are trending favorably,overall burn remains heavy for AI companies.STATE OF ENTERPRISE SOFTWARE 20257Notes:1)The 8th category was“Other.”Findings based on a survey from the SVB Advantage CFO report of more than 200 finance leaders across the innovation economy,conducted in F

30、ebruary 2025.2)Google Trends searches as of 3/31/2025.3)Based on total revenues and full-time employees(FTE).4)AI-exposed companies determined using SVB methodology.5)Data for US companies listed on major US exchanges.Source:Google Trends,S&P Capital IQ,SVB Advantage CFO Survey 2025 and SVB analysis

31、.AI-ExposedNon-AI-Exposed$165K$164K$170K$180K$218K$266K$273K$293K$314K$349K$444K$511K$533K$528K$559K20202021202220232024$237K$260K$286K$307K$345K$349K$339K$359K$384K$420K$639K$664K$693K$665K$808K20202021202220232024AI-ExposedNon-AI-Exposed11%15%19%8%7%20%28%27%17%19%58%79%67%79%94%53%68%64%65%61%202

32、020212022202320240102030405060708090100$0B$5B$10B$15B$20B$25B$30B$35B$40B$45B$50B20172018201920202021202220232024Enterprise SoftwareConsumer InternetFintechFrontierTechAI Adoption1325Inflation/Rates2112(Tie)Regulation3236Labor Markets4544Geopolitics56(Tie)57Supply Chain66(Tie)71Tariffs and Trade7462

33、(Tie)Middle 50%MedianAI-Exposed:Median Employee Increase YoYAI-Exposed:Median Revenue Increase YoYNVIDIA Quarterly RevenueGoogle Trends Searches for AI(indexed to 100 at peak)NVIDIA quarterly revenue has increased 6.5x in the last 8 quarters.8STATE OF ENTERPRISE SOFTWARE 2025Funds caught in between

34、established and emerging managers better knock on wood because theyre slowly being hollowed out.Since 2020,there has been a clear bifurcation in the market,where the biggest funds focus on making large investments and,in some cases,nearly“index”the venture market.On the other hand,small funds carve

35、out niches,targeting specific founders,technologies,stages or geographies.The strength of both ends of the spectrum is marginalizing mid-sized funds.This leaves mid-sized funds in a precarious position.Their role is less clear.Most are neither behemoths able to compete in mega-deals nor niche funds

36、in hyper-specialized markets.This could lead to consolidation and a less competitive market,with capital and talent increasingly concentrated among a few top firms.One area that most funds,large and small,have been focusing on is AI.More than 15%of US VC funds list AI as a focus twice the amount it

37、was just five years ago.While this may not seem impressive,this subset accounts for 40%of total capital raised from US VC funds in 2024.Those with an AI tilt also close above their initial target size 3x as often as funds that dont list AI as a target.These funds are also closing 17%above target.Put

38、 all together,this reflects not only the investor enthusiasm around the space,but also the funds required to properly deploy into capital-intensive hypergrowth AI startups.STATE OF ENTERPRISE SOFTWARE 20259Notes:1)Based on fund manager location.For closed US VC funds only,based on Preqin definitions

39、.2)Based on funds that list artificial intelligence as a targeted vertical.Funds may have multiple targeted verticals.3)Based on final close size and initial fund target size.Source:Preqin and SVB analysis.$114BAmount Currently Being Raised Based on Target SizeTTM US VC Fundraising22%21%20%19%15%9%3

40、5%33%22%24%29%36%201920202021202220232024US VC Funds that List AI as a Targeted VerticalUS VC Funds that Do Not List AI as a Targeted Vertical0%10%20%30%40%50%201520162017201820192020202120222023202440%of capital from US VC funds closed in 2024 came from funds that listed AI as a focus.Share from$50

41、0M FundsShare from$100M-$500M FundsShare from$100M Funds$0B$30B$60B$90B$120B$150B$180B$210B0%10%20%30%40%50%60%70%2015201620172018201920202021202220232024Year Fund ClosedContrary to what many expected,venture investment is on the upswing and back to 2022 levels.While US VC deal activity fell 8%in 20

42、24,US VC investment totaled$214B a 30%YoY increase and the third-highest annual total on record.The recovery marks an about-face for the venture ecosystem.Last year started at a low point,after eight straight quarterly declines in annual VC investment,and ended on a hot streak with three quarterly i

43、ncreases.This seems to have only continued in 2025 with mega-deals(defined here as$100M+rounds)back in full force.Through Q1,there have been 107 mega-deals the most since Q2 2022.Leading the charge were multibillion-dollar raises from AI bellwethers such as OpenAI and Anthropic.In fact,while AI has

44、only accounted for 6%of all mega-deals this year,the dollars from those deals represented 50%of the capital raised in mega-deals.Exclude AI investment and the story changes.There is no meaningful investment uptick for companies not leveraging AI,with investment for this group essentially flat for th

45、e last year.While the growing concentration in specific subsectors and deal sizes may scream“bubble”to some,history shows that you cannot tell the story of venture investment without acknowledging the innovation wave that lifts the larger VC boat.Dont fear the wave hop on board and ride it.STATE OF

46、ENTERPRISE SOFTWARE 202510Notes:1)VC investment includes early-stage,late-stage,seed,Series A+and corporate deals with a series attached to the deal.2)Selected technology verticals determined using PitchBook Data,Inc.s taxonomy.3)Based on SVB proprietary taxonomy.Source:PitchBook Data,Inc.,SVB propr

47、ietary taxonomy and SVB analysis.0%5%10%15%20%25%30%35%40%$0B$50B$100B$150B$200B$250B$300B$350B$400B199819992000200120022003200420052006200720082009201020112012201320142015201620172018201920202021202220232024TTM US VC Investmente-CommerceMobileAI0%10%20%30%40%50%60%0%10%20%30%40%50%60%20192024Artifi

48、cialIntelligenceApplicationsInfrastructureCybersecurityServicesMarket-placesShare of Enterprise Capital InvestedShare of Enterprise Deal Count41%46%55%46%42%49%46%38%59%41%51%68%201920202021202220232024US VC Enterprise Software DealsUS VC Deals excl.Enterprise Software68%of capital that went to US e

49、nterprise startups came from$100M+deals.The Series A crunch is well-documented.And while enterprise software might be a more desirable place to invest,the sector is not immune and is taking its medicine.Graduation rates have continued to trickle down across enterprise subsectors,due in part to seed

50、extensions becoming more common as well as significant growth in the amount of capital raised at a seed round.For instance,seed extension rounds accounted for 28%of all US VC enterprise software seed deals in 2024.Meanwhile,the median amount raised in an initial seed round for enterprise startups is

51、$2.8M 34%higher than the median in 2021.With more capital(and higher valuations)comes loftier expectations.This,in part,has raised the bar for Series A benchmarks expected of enterprise software seed cohorts.Take,for example,the bottom quartile for revenue at time of Series A financing in 2024($1.3M

52、),which was the median revenue at time of Series A financing in 2021.Looking forward,until the seed cohorts get digested(i.e.,raise,fail or exit),graduation rates are likely to remain muted.Couple this with potential ongoing market uncertainty,and we may see a“flight to quality”where investors conti

53、nue to be extremely selective in their capital deployment potentially making it even harder to raise the next round.Those left in the dust will need to seek an exit.More on that later.STATE OF ENTERPRISE SOFTWARE 202511Notes:1)Sector and subsectors determined using SVB proprietary taxonomy.2)Delta s

54、ymbolizes“change.”Tech includes all other tech sectors as defined by SVB proprietary taxonomy excluding Enterprise Software.3)Revenue at Series A excludes extension rounds.Source:PitchBook Data,Inc.,SVB proprietary data and SVB analysis.0%5%10%15%20%25%30%06121824303620242023202220212020201920172016

55、20152018Year vs.Tech excl.ES220153.6%20163.7%20172.7%20181.4%20196.5%20204.7%20213.6%20222.7%20231.6%20241.4%Months to Graduate0%5%10%15%20%25%30%20162017201820192020202120222023202420250%5%10%15%20%25%30%35%40%$0M$1M$2M$3M$4M$5M$6M$7MBottom QuartileMedianTop QuartileWithin 3 YearsWithin 2 YearsWith

56、in 1 Year20192020202120222021202220232024ApplicationsInfrastructureAICyberThe herd of US VC-backed enterprise software unicorns continues to grow,albeit at a slower pace relative to previous years.The current cohort accounts for 40%of all US VC-backed unicorns by count(in line with historical norms)

57、and aggregate valuation(nearly 2.5x the share a decade ago).Where recent cohorts separate themselves is their speed in achieving unicorn status.For the 2024 cohort,the median time to reach unicorn status from a companys founding year was 3.6 years a year quicker than consumer and frontier startups,a

58、nd 2.5 years quicker than fintechs.To no surprise,within enterprise software,AI startups continue to stand out.Since 2021,on a median basis,AI companies have achieved unicorn status in 4.5 years,which is much quicker compared to applications and infrastructure(8 years),and cybersecurity(6 years)star

59、tups.While new unicorn creation has been impressive in recent years,unicorn exits are a different story.The enterprise unicorn herd stands above 300,with few exiting,closing their doors or taking a down round below a$1B post-money valuation.This has pushed the distribution to skew older,with the med

60、ian age standing at 11.5 years.With the growth of the herd,so too comes growing demand for liquidity.But with the IPO window effectively shut,and large scale M&A being tougher to execute with the overhanging macro climate,many run the risk of ending up in no mans land.This is evident in the uptick i

61、n“Zombiecorns those with poor revenue growth and unit economics.STATE OF ENTERPRISE SOFTWARE 202512Notes:1)Sector and subsectors determined using SVB proprietary taxonomy.2)Based on annualized quarterly revenue and operating margins.Source:PitchBook Data,Inc.,SVB proprietary data and SVB analysis.Cu

62、rrent2016 2017 2018 2019 2020 2021 2022 2023 2024 20252015 Total$0.0B$0.2B$0.4B$0.6B$0.8B$1.0B$1.2B0123456789102024202320222021202020192018Years0%5%10%15%20%25%0-2 yrs.2-4 yrs.4-6 yrs.6-8 yrs.8-10 yrs.10-12 yrs.12-14 yrs.14-16 yrs.16-18 yrs.18-20 yrs.20+yrs.0%5%10%15%20%25%30%35%Q1Q3Q1Q3Q1Q3Q1Q3Q1Q3

63、Q1Q3Q12019202020212022202320242025US Tech Unicorns excl.Enterprise SoftwareUS Enterprise Software Unicorns 2019 DistributionCurrent Distribution2021 DistributionCurrent distribution of US Enterprise Software unicorns skews older.STATE OF ENTERPRISE SOFTWARE 202513Capital Raised1,2Pre-Money Valuation

64、s1,2$2.0M$2.0M$2.1M$2.6M$3.0M$3.0M$2.0M$2.0M$2.4M$3.0M$2.8M$3.5M201920202021202220232024$8M$9M$11M$12M$11M$13M$9M$10M$12M$15M$12M$19M201920202021202220232024$7M$7M$9M$10M$11M$12M$8M$8M$10M$12M$14M$16M201920202021202220232024$23M$25M$38M$42M$38M$42M$21M$30M$44M$49M$40M$69M201920202021202220232024$18M

65、$25M$30M$30M$26M$28M$25M$21M$30M$39M$45M$45M201920202021202220232024$40M$45M$100M$75M$44M$67M$60M$43M$75M$70M$50M$45M201920202021202220232024$70M$82M$130M$140M$108M$122M$85M$86M$136M$120M$275M$233M201920202021202220232024$240M$292M$800M$555M$322M$475M$300M$280M$550M$350M$250M$945M2019202020212022202

66、32024Median-AIMedian-Enterprise excl.AIMiddle 50%-Enterprise excl.AISeedSeries ASeries BSeries C+SeedSeries ASeries BSeries C+Notes:1)Series defined by PitchBook Data,Inc.2)Enterprise Software and AI defined by SVB proprietary taxonomy.Source:PitchBook Data,Inc.and SVB analysis.Median-AIMedian-Enter

67、prise excl.AIMiddle 50%-Enterprise excl.AI14STATE OF ENTERPRISE SOFTWARE 2025Cash remains king.In climates like todays,it cannot be understated how important burn and cash runway management is to a startups success and longevity.Across stages,on a median basis,smaller enterprise software startups ar

68、e seeing runway trickle lower while later-stage companies are holding steady.This is likely due in part to earlier-stage startups having fewer levers to pull.The quickest way to extend runway is to cut costs and improve margins,which every single subsector and life stage accomplished last year.For e

69、arlier-stage startups,walking the tightrope of managing burn to extend runway without compromising revenue growth through product development or go-to-market(GTM)strategies remains a key challenge.Revenue growth is a critical metric for raising the next funding round.These challenges will only be ex

70、acerbated by subdued investment appetite,should investors continue to get more selective with deal flow and lack of exits slow the liquidity flywheel.From a subsector standpoint,AI startups were the only cohort of startups to see their runway increase on a median basis.This is reflective of both the

71、 cash injections AI startups have received from investors looking to capitalize on the current opportunity as well as the boost in revenue that AI startups have experienced from increased customer appetite and usage.STATE OF ENTERPRISE SOFTWARE 202515Notes:1)Revenue buckets based on latest annualize

72、d quarterly data.2)Runway based on Q4 data of the respective year.3)Subsectors based on SVB proprietary taxonomy.4)EBITDA margin based on Q4 data of the respective year.Source:SVB proprietary data and SVB analysis.2019202020212022202320241216102291618300369121518212427303336$0-$10M$10M-$25M$25M-$50M

73、$50M+1211131414101411036912151821AICyberApplicationsInfrastructure-215%-63%-38%-25%-152%-41%-13%-7%-350%-300%-250%-200%-150%-100%-50%0%$0-$10M$10M-$25M$25M-$50M$50M+-253%-144%-89%-105%-169%-97%-40%-79%-350%-300%-250%-200%-150%-100%-50%0%AICyberApplicationsInfrastructure201920202021202220232024201920

74、202021202220232024201920202021202220232024Months of Cash RunwayMonths of Cash RunwayRevenue growth is declining across life stages.As firms pull back on enterprise spending,startups are feeling the pinch.Balancing profitability and growth will be the key.For late-stage US enterprise software startup

75、s,this is often viewed through the Rule of 40.1 This metric has continued to slide as growth rate declines have outpaced the progress in margins.The median Rule of 40 for US enterprise software startups with$50M+in revenue was 9%in 2024,down from 21%in 2021.Additionally,the share of startups above t

76、he Rule of 40 threshold has fallen to 13%,down from 31%in 2021.However,when it comes to attracting capital and investor interest,the best of breed will always find a way.Using a mix of proprietary data and qualitative analysis,we built a cohort of top-performing US enterprise software startups.These

77、 metrics give a glimpse under the hood of the fastest growing companies we work with.Notably,we have yet to see much substance to back up the“SaaS is dead”narrative.The vast majority has stuck with a SaaS model,though theres been a shift toward a core subscription offering with consumption features

78、layered on.As AI Squared CEO Darren Kimura notes,“The principles behind SaaS still apply,but the dynamic is shifting toward value,which is driving the rise of consumption-based pricing.Enterprise software still demands security,scale and predictability,which supports a SaaS-like foundation.But beyon

79、d that,features should be priced based on the value they deliver.Its all about finding the right balance to meet evolving enterprise needs.”Notes:1)The Rule of 40 is a commonly used financial metric for enterprise software companies that states the sum of the revenue growth rate and the profit margi

80、n should be 40%or higher.2)Revenue buckets based on latest annualized quarterly data.3)Revenue growth and EBITDA margins based on annualized Q4 data of the respective year.4)Annual recurring revenue.Source:SVB proprietary data and SVB analysis.-45%-40%-35%-30%-25%-20%-15%-10%-5%0%0%5%10%15%20%25%30%

81、35%40%45%Revenue Growth YoYEBITDA Margin202320222021202020192024Bubbles scaled by share of US VC-backed enterprise software startups that are above the Rule of 40 for a given quarter.0%10%20%30%40%50%60%70%80%$0-$10M$10M-$25M$25M-$50M$50M+Revenue BucketSTATE OF ENTERPRISE SOFTWARE 202516$10M-$25M$50

82、M+201920202021202220232024-18%-14%Logo Churn69%29%Logo Growth$10M-$25M$50M+-14%-10%ARR Churn98%40%ARR Growth$10M-$25M$50M+$10M-$25M$50M+Special Contribution from17STATE OF ENTERPRISE SOFTWARE 2025Despite the decreasing cost of software and higher-efficiency scale for AI startups,capital requirements

83、 for AI may remain elevated due to high compute costs,expenses of attracting top-tier talent,and building proprietary infrastructure for training,fine-tuning and deploying models.When analyzing proprietary Pinegrove Venture Partners data,the average AI startup is scaling to$100M in 2 years(for the 2

84、022 cohort),which is nearly a quarter of the time from just a decade ago.However,the amount of capital to get there has kept up pace.The average AI startup is simultaneously burning through$100M in 3 years(for the 2022 cohort),which is roughly half the time it took a decade ago.In that light,AI comp

85、anies will maintain the need for massive capital investment from venture capital,private equity and corporate investors.Insert the large multi-stage managers with built-up war chests.Large multi-stage managers may continue to capture increasing amounts of value accruing to AI startups with their abi

86、lity to meet these funding needs,relative to smaller competitors in market.However,dont count out the competition and their ingenuity.Smaller venture funds are competing in AI by building deeper founder relationships and offering hands-on technical support.They often provide unique value like GPU ac

87、cess,infra credits or AI-specific GTM help.Many are deeply embedded in AI research or open-source communities,giving them early access to emerging talent.Some also differentiate through global sourcing,flexible check sizes and curated founder ecosystems.STATE OF ENTERPRISE SOFTWARE 202518Notes:1)Dat

88、a tracks all venture deals for the years associated and measures the total value generated to date.Value is defined as the growth in post-money valuation minus the increase in capital raised.Large multi-stage investors are identified from a proprietary list of firms actively tracked by Pinegrove Ven

89、ture Partners.Source:Pinegrove Venture Partners proprietary data and SVB analysis.$0M$20M$40M$60M$80M$100M$120M01234567891020152016201720182022202120202019Years to$100M in RevenueYears to$100M in Burn$0M$20M$40M$60M$80M$100M$120M0123456789102015201620172018202220212020201962%67%64%66%61%58%70%68%88%

90、89%2015201620172018201920202021202220232024Pre-Gen AIPost-Gen AISTATE OF ENTERPRISE SOFTWARE 202519Source:Pinegrove Venture Partners analysis.Plateau will be reached in:0-2 years2-5 years5-10 yearsMore than 10 yearsExpectationsInnovation TriggerPeak ExpectationsDisillusionmentSlope of EnlightenmentP

91、lateau of ProductivityFirst Principles AIAutonomous VehiclesPrompt EngineeringCloud AI ServicesEmbodied AIQuantum AIAI EngineeringEdge AISynthetic DataGenerative AIFoundation ModelsModelOpsIntelligent ApplicationsKnowledge GraphsAI-Ready DataAutonomous AI AgentsAI SimulationNeuro-Symbolic AIComposit

92、e AIDecision IntelligenceSovereign AIAI TRiSMArtificial General IntelligenceResponsible AISmart RobotsNeuromorphic ComputingMultimodal AIAI Governance ToolAI Coding AssistantsAI-Augmented AnalyticsMultiagent SystemsCausal AISTATE OF ENTERPRISE SOFTWARE 202520Notes:1)Importance score determined using

93、 Pinegrove Venture Partners proprietary ranking system based on investment discussions.Source:Pinegrove Venture Partners proprietary data.GrowthBusiness ModelCapital RaisingPortfolio ManagementEnterprise AdoptionProduct Market FitScaling ChallengesCapital RequirementsPortfolio ConstructionDefensibil

94、ityGTM ChallengesGPU ComputerMarket CompetitionTalent AcquisitionScalingLimitationsCompetitionAccessibilityFinancingRiskEthical IssuesProduct Market FitScaling ChallengesCompetitionEfficiencyLabor ShortagePatentsSafetyEnterprise AdoptionCapital DeploymentSecurity AutomationHealthcare AIAI Automation

95、Character AINetwork EffectsAI PrimitivesData InfrastructureAutonomous SystemsRoboticsDeep TechDataSecurityAIInfrastructureWorkflowOptimizationDigital HealthSyntheticBiologyRoboticsAutonomousVehiclesWeb3Digital HealthTelemedicineBlockchainAIDesignPopulation HealthOpportunitiesChallenges20212022202320

96、24Using notes and materials from five years of manager conversations,we applied our proprietary AI systems to identify the most cited challenges and opportunities in AI by year.This analysis reflects recurring themes,ranked by importance and tracked over time to highlight how priorities have evolved

97、.EngineeringSynthetic BiologyBlockchainSignal DetectionEmbedded TechComputer VisionCapital NeedsFintech21STATE OF ENTERPRISE SOFTWARE 2025$0B$20B$40B$60B$80B$100B$120B$140B$160B$180B20152016201720182019202020212022202320242025$0B$5B$10B$15B$20B$25B$30B$35B$40B$45B201520162017201820192020202120222023

98、20242025Heading into the year on the back of a successful public exit for ServiceTitan,most investors were optimistic about a prosperous IPO market in 2025.CoreWeave provided a glimmer of hope it would kick-start the market,but“Liberation Day”brought things to a screeching halt.Arguably one of the b

99、iggest knock-on effects for the innovation economy of heightened trade tensions is the ongoing uncertainty limiting exit options.Potential acquirers are less likely to make large,transformational acquisitions with less clarity on how to value a business.Simultaneously,companies may be hesitant to ex

100、it amid the market volatility if theyre not confident theyll receive a favorable public debut and valuation multiple.To the latter point,revenue multiples at IPO have already come down dramatically since the IPO boom when valuations were at their frothiest.The IPO bar is higher today,and few enterpr

101、ise software startups surpass it.Most are either too small,growing too slowly or dont have a clear path to profitability.Couple this with plenty of deployable late-stage capital for those in need that also meet the proper criteria,and there is less of a need to go public.Liquidity concerns are real

102、for investors,though.However,until the IPO window thaws,look for secondary markets or M&A activity to provide some relief.Notes:1)Enterprise Software defined using SVB proprietary taxonomy.2)Value defined as aggregate IPO post-valuation.3)Revenue multiple determined using TTM revenue and the company

103、s post-valuation at IPO.4)Based on SVBs proprietary data.Source:PitchBook Data,Inc.,S&P Capital IQ,SVB proprietary data and SVB analysis.Too Small:Revenue$25M71%13%3%13%29%16%6.2x6.9x7.2x15.2x11.7x4.6x201720182019202020212022-2025TTM IPO Aggregate ValuationIPO(Sized by Valuation at IPO)STATE OF ENTE

104、RPRISE SOFTWARE 202522US VC-Backed Enterprise Software StartupsMost Active Acquirers of Enterprise Startups 2024-25Rank Company12345678910Multiple InvestorsAfter constant gripes of M&A not holding up its end of the bargain,the tides are finally turning.Driving the enthusiasm was Googles$32B acquisit

105、ion of Wiz its largest deal ever.Not only did it provide much-needed liquidity,but it also potentially signaled where enterprise spend is headed security and AI.While splashy deals grab headlines,most are happening at the early stage.Looking at M&A deal activity for US enterprise software startups,a

106、 growing share is exiting at the seed stage.This is a function of both larger strategics scooping up tech and talent early,as well as seed startups failing to meet the necessary benchmarks to raise a Series A.More acquisitions may be on the way.Darren Kimura,CEO of AI Squared,who splits his time as

107、an investor,highlighted that“Theres a noticeable shift in mindset and a growing appetite for exit opportunities.In the past,many would have doubled down on growth.Today,if an M&A option looks reasonably attractive,teams are more willing to consider it,particularly if it means they can continue build

108、ing technology that contributes to the rise of this AI revolution.”The AI revolution has also paved the way for VC-led roll-ups,a PE-style play that allows AI-native startups to transform legacy businesses and services driving growth and liquidity.Add in a sluggish IPO market,the need for nearly hal

109、f of all US enterprise software startups to raise(or exit)in the next year and S&P 500 Tech cash at$604B,1 and its a recipe for more activity.The liquidity will be much needed to reinvigorate venture returns and jump-start funding for the next wave of startups.Notes:1)Data as of 3/31/2025.Tech defin

110、ed by SVB.2)Enterprise tagging based on SVB proprietary taxonomy.3)Not all deals have disclosed valuations.4)2024-2025 includes data from 1/1/2024-4/30/2025.5)Pre-seed includes accelerator/incubator and angel rounds.Early-stage includes Series A and B as well as early-stage as defined by PitchBook D

111、ata,Inc.Late-stage includes Series C+as well as late-stage as defined by PitchBook Data,Inc.6)Runway data as of latest available quarterly data.Source:S&P Capital IQ,PitchBook Data,Inc.,SVB proprietary data and SVB analysis.$0B$8B$16B$24B$32B$40B$48B$56B$64B$72B$80B$88B050100150200250300350400450500

112、55020172018201920202021202220232024$40B Pending Acquisition Value0%10%20%30%40%50%60%70%80%90%100%0-33-66-99-1212-1515-1818-2121-2424-2727-3030-3333-36Months Until Need To RaiseShare of US Enterprise Software Startups That Need To Raise as of 2024TTM M&A CountTTM M&A Aggregate ValuationHalf of US En

113、terprise Software startups need to raise in 12 months.STATE OF ENTERPRISE SOFTWARE 202523Share of enterprise startups getting acquired at seed stage has jumped 13 percentage points since 20190%10%20%30%40%50%60%70%80%90%100%2019 2020 2021 2022 2023 24-25ServicesInfr.AppsAI/MLCyberMarketplaces16%16%1

114、6%17%14%13%14%17%26%25%27%27%45%41%34%35%42%40%26%25%23%24%17%19%2019202020212022202324-25Pre-SeedSeedEarly-StageLate-StageSTATE OF ENTERPRISE SOFTWARE 202524Artificial Intelligence:Technology that enables machines to do tasks that would otherwise require human intervention.Includes core AI(i.e.,fou

115、ndational models),CyberAI,AI infrastructure and AI applications.Cybersecurity:Technologies and services designed to protect networks,computers,applications and data from attack,damage or unauthorized access or attack.Includes network management software.Enterprise Applications:Technologies that are

116、built on top of the underlying infrastructure.Includes categories such as sales software,operations software,marketing software,HR tech,communication software,collaboration software and vertical specific software.Enterprise Infrastructure:The foundational framework that underpins the underlying tech

117、nological platform.Includes categories such as business intelligence and analytics,developer tools,microservices and data infrastructure.Enterprise Software:Software that is developed in order to satisfy the needs of businesses and organizations.Main subsectors include artificial intelligence,cybers

118、ecurity,enterprise applications and enterprise infrastructure.Rob HelmSenior Market ManagerTech BankingSilicon Valley BAndrew McCartySenior Vice PresidentTech BankingSilicon Valley BRob Helm is head of Consumer and Enterprise Relationship Management in Northern California.In this role for the SVB Co

119、mmercial Bank,he oversees leadership,client relationships and strategy.Andrew McCarty is a Senior Vice President on SVBs NorCal Enterprise Software team.In this role,he advises startups and their founders on creative banking and financing solutions while leveraging SVBs network to maximize clients p

120、robability of success.Andrew Pardo,CFASenior Analytics ResearcherSVB Market InsightsSilicon Valley BAndrew Pardo and the SVB Market Insights team leverage SVBs proprietary data,a deep bench of subject-matter experts,and relationships with world-class investors and founders to develop a holistic view

121、 of the innovation economy.Dave MullenPartnerVenture CapitalPinegrove Venture Partnersdavepinegrove.vcPinegrove is a venture investment platform designed to deliver tailored solutions for fund managers,founders and limited partners in the venture capital ecosystem.Our expertise spans fund of funds,v

122、enture debt funds,venture secondaries and co-investments.Alexey LagerbergSenior Data ScientistVenture CapitalPinegrove Venture Partnersalexeypinegrove.vcSTATE OF ENTERPRISE SOFTWARE 202525Silicon Valley Bank(SVB),a division of First Citizens Bank,is the bank of some of the worlds most innovative com

123、panies and investors.SVB provides commercial banking to companies in the technology,life science and healthcare,private equity and venture capital industries.SVB operates in centers of innovation throughout the United States,serving the unique needs of its dynamic clients with deep sector expertise,

124、insights and connections.SVBs parent company,First Citizens BancShares,Inc.(NASDAQ:FCNCA),is a top 20 U.S.financial institution with more than$200 billion in assets.First Citizens Bank,Member FDIC.Learn more at .Silicon Valley B See complete disclaimers on the following page.2025 First-Citizens Bank

125、&Trust Company.Silicon Valley Bank,a division of First-Citizens Bank&Trust Company.Member FDIC.26STATE OF ENTERPRISE SOFTWARE 202526The views expressed in this report are solely those of the authors and do not necessarily reflect the views of SVB.This material,including without limitation to the sta

126、tistical information herein,is provided for informational purposes only.The material is based in part on information from third-party sources that we believe to be reliable but which has not been independently verified by us,and,as such,we do not represent the information is accurate or complete.The

127、 information should not be viewed as tax,accounting,investment,legal or other advice,nor is it to be relied on in making an investment or other decision.You should obtain relevant and specific professional advice before making any investment decision.Nothing relating to the material should be constr

128、ued as a solicitation,offer or recommendation to acquire or dispose of any investment,or to engage in any other transaction.All non-SVB named companies listed throughout this document,as represented with the various statistical,thoughts,analysis and insights shared in this document,are independent t

129、hird parties and are not affiliated with Silicon Valley Bank,a division of First-Citizens Bank&Trust Company.Any predictions are based on subjective assessments and assumptions.Accordingly,any predictions,projections or analysis should not be viewed as factual and should not be relied upon as an acc

130、urate prediction of future results.Investment Products:Are not insured by the FDIC or any other federal government agencyAre not deposits of or guaranteed by a bankMay lose value 2025 First-Citizens Bank&Trust Company.Silicon Valley Bank,a division of First-Citizens Bank&Trust Company.Member FDIC.STATE OF ENTERPRISE SOFTWARE 202527