parallax background















The Impact of Cloud Services in the United States


Executive Summary

To learn more about the impact of cloud services on the U.S. economy, Amazon Web Services (AWS), commissioned independent consultancy Public First to undertake this quantitative research to understand the use of and benefits created by cloud services across the U.S., with a particular focus on Infrastructure as a Service (IaaS) and Platform as a Service (PaaS) forms of cloud services.

Unless otherwise stated, in this report ‘cloud services’ refers to IaaS or PaaS, and does not include Software as a Service (SaaS). By contrast, when we refer to online or cloud tools, we specifically mean SaaS tools.

By conducting a survey of over 3,000 U.S. businesses across regions and industries, including over 1,500 cloud users, Public First provides the following key results:


Broad economic impacts

  • Public First estimates that, in 2021 alone, cloud services generated over $382 billion in Gross Value Added, or GVA,1 for businesses in the U.S, quantified by increased revenues and reduced costs.2
  • Companies with an employment footprint of over 37 million jobs said their business model would not be possible without cloud services.
  • AWS users reported both cost savings and productivity improvements. On average, AWS users reported that, by migrating their services to the cloud, they were able to save 50% of their IT costs, while software developers said that they could reduce the time to develop new software by around 25%.

Who is currently using the cloud services?

A majority of businesses in the U.S. now utilize cloud services, but significant disparities in cloud adoption remain across states, industries, and business size. Based on Public First’s survey:

  • 51% of businesses said that they currently use some form of cloud services (including IaaS, PaaS and SaaS), and 21% said that they use IaaS or PaaS as a replacement or complement for their own IT infrastructure.
  • 46% of women-owned businesses in the U.S. reported using some form of cloud services (IaaS, PaaS or SaaS).
  • 58% of minority-owned businesses in the U.S. reported using some form of cloud services, including 61% of Latino-owned businesses (IaaS, PaaS or SaaS).

Small and Medium Enterprises (SMEs) and cloud services

Though adoption still lags behind larger companies, SMEs3 in the U.S. who used cloud services report that it helped them increase their competitiveness and scale up their business.

  • 85% of SMEs using cloud services surveyed agreed that the use of cloud services made it easier for them to compete with larger businesses.
  • 60% of SMEs using cloud services said that it enabled them to significantly increase their ability to scale and 47% said that it enabled them to offer new kinds of products.
  • Among study recipients, businesses in the U.S. with fewer than 10 employees were half as likely to use cloud services than businesses with over 250 employees. If the cloud service adoption rate for U.S. businesses with fewer than 250 employees increased by 10%, it would increase GVA by over $26 billion.
  • A large majority of cloud service users with increased revenue reported investing that revenue back into their businesses: 45% hired more staff, 49% invested in workforce training, and 41% made other types of investment.

Start-ups

Cloud services unlock new business models and revenue growth

  • 74% of surveyed cloud service users agreed that their business or operating model would not be possible without cloud services.

Cloud services enable collaboration and productivity

Cloud computing products and services such as online project management, employee communication or e-commerce tools now play a critical role in supporting business productivity and resilience.

  • 63% of U.S. businesses agreed that online or cloud tools made it easier for workers to keep collaborating throughout COVID-19, and 10% claimed that it would have been impossible to keep their businesses operating during COVID-19 without online or cloud services and tools.
  • Tools such as Enterprise Resource Planning (ERP) or Consumer Relationship Management (CRM) correlate with a 10% to 25% improvement in business productivity.
parallax background

Introduction

What are cloud services?

Imagine if every business had to build their own power plant. For any business, this would not only be highly expensive, but also be a distraction from its core offering. Even worse, for SMEs, the costs involved might prevent them from getting started altogether. The net result would be a slower, less innovative and less competitive economy. Cloud services make it possible for businesses to take advantage of all the power of modern Information Technology (IT) infrastructure and computing resources - without having to invest in the heavy upfront capital costs of procuring servers, storage, and other related IT equipment, which can take months to acquire and install. In the same way that the electric grid makes power available at the flip of a switch, cloud services enable flexible, pay-as-you-go, instant access to computing power.

Cloud services treat both computing resources and data storage as services: scaling them up or down to meet businesses’ on-demand needs, providing access wherever needed, empowering businesses with up-to-date IT technologies. As demonstrated later in the report, agilities have reduced costs and increased capacity or scale for businesses to grow. They have also enhanced collaboration and productivity between businesses that allow the economy to adapt to abrupt disruptions, such as the COVID-19 pandemic.

Ever since the first commercial computers, the model for computing has continually evolved as new technologies changed the balance between centralized and local computers:

  • In the 1960s and 1970s, most computing tasks ran on a mainframe, sharing centralized resources.
  • In the 1980s, the personal computer (PC) allowed workers to run their own programs locally.
  • In the 1990s, the rise of the Internet and the client-server model made it easier to connect those PCs together.
  • In the 2000s, the smartphone significantly expanded the number of computing endpoints.
  • In the 2010s, cloud services helped connect together servers, PCs, the web and smartphones.

Over the last decade, large numbers of businesses started utilizing cloud services, migrating away from their legacy IT infrastructure, and at the same time, a new ecosystem of SMEs built products and services directly out of cloud services, taking advantage of the greater flexibility and agility. Cloud services are now general purpose IT technologies, rather than being predominantly used by the core tech sector. Despite this, it remains relatively early days for cloud services, with many existing software processes or workflows yet to be transitioned. The Innovation Technology & Innovation Foundation (ITIF) estimates that spending on cloud computing accounts for just 7.2% of global IT spending.4

Traditional Types of Cloud Computing

The National Institute of Standards and Technology (NIST) defines cloud computing as “a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction”5. They classify cloud computing as having five essential characteristics: on-demand self-service, broad network access, resource pooling, rapid elasticity, and is a measured service.

While some organizations run their own private cloud storage, more common today is to use a public cloud provider, an organization that offers cloud computing services to external companies or members of the public.

While the lines have blurred, there are three traditional service models for cloud computing:

  • Infrastructure as a Service (IaaS). In this type of cloud computing, businesses replace their own on-premises services with hired data storage and computing power owned by a public cloud service provider. This model was first pioneered with the creation of Amazon Web Services (AWS) in 2006.
  • Platform as Service (PaaS). In the next level up of cloud computing, developers run databases and middleware in the cloud, while the underlying hardware infrastructure is abstracted away through virtualization.
  • Software as a Service (SaaS).  Finally, SaaS or cloud computing providers offer consumer facing applications or services that either run or process their data in the cloud. This allows their users to often use the same application on a wide range of platforms apart from their main computer, such as on their phone or through a direct web interface.

In recent years some industry analysts have ceased to differentiate these categories of services to the same extent. In particular, industry analysts commonly group IaaS and PaaS into one category of cloud services. For example, in 2020, Gartner revised its Magic Quadrant report for IaaS to include PaaS to reflect customers’ broad usage of IaaS providers.6 For purposes of this report, and unless otherwise specified, in this report ‘cloud users’ are companies that use either IaaS or PaaS, not SaaS alone.

How do cloud services boost economic growth?

There are multiple channels through which cloud services can boost economic growth:

  • Increasing the economies of scale for computing. While traditional on-premises servers may only see less than 20% utilization rates, servers in the public cloud typically see more than 50% utilization rates.7 At the same time, cloud services allow customers to share in the benefits of cloud providers’ centralized R&D and investments.
  • Increasing competitiveness. By removing the need for businesses to make expensive upfront capital investment in servers, IT equipment, and software licenses, the public cloud significantly lowers the cost barriers for new businesses to start up, and scale to serve customers globally.
  • Improving innovation and business dynamism. These benefits can not only reduce costs, but also accelerate the process of prototyping, developing and deploying a new service, removing the need to wait to put in place computer hardware.
  • Improving workforce collaboration. Online tools and cloud services are playing an increasingly important role in making it easier for employees to collaborate and work more flexibly, boosting labor productivity.
  • Enabling future technologies. Cloud services are critical to the development of emerging technologies, such as machine learning, blockchain, or VR. These technologies are major drivers of economic growth. McKinsey, for example, estimates that machine learning could increase U.S. GDP 21% by 2030.8 In this study, Public First estimates that 81% of businesses that used AI or machine learning also used the cloud. 61% of those businesses that used AI or machine learning said that their business or operating model would not be possible without the cloud.

How did Public First quantify the impact of cloud services?

Public First conducted a survey of over 3,000 U.S. businesses across regions and industries, including over 1,500 cloud users, to learn more about how cloud services and AWS are supporting them. The survey results presented in this study were weighted by interlocking region and business count to be representative of the US population at a national level.  Based on the results of this survey, we produced new estimates of the average Return on Investment (ROI) that businesses report from their spend on cloud services, and used this to estimate the overall economic impact across the U.S. as a whole. Unless otherwise stated, all polling results in this report are taken from this survey.

To learn more about Public First’s modeling approach, please refer to the Methodology section in Appendix A and B. Public First is a member of the Market Research Society. The full tables for all the data used in this report are available to download from our website. While AWS commissioned this report from Public First, all economic estimates are derived from official, third party and Public First’s proprietary information.

Public First estimates that, in 2021 alone, cloud services generated over $382 billion in Gross Value Added (GVA) for cloud-using businesses in the U.S. At the same time, the business model of companies with an employment footprint of over 37 million jobs would not be possible without cloud services.

parallax background

Who is using cloud services in the U.S.?

A majority of U.S. businesses are now using cloud services

The U.S. is the world’s largest largest purchaser of public cloud services, with a current spend roughly equal to the rest of the world all together. Statista estimates that global public cloud revenue was over $400 billion in 2023, with the U.S. economy making up over half of this. In 2023, the total global spend on public cloud services is expected to exceed $525 billion.9 In this study, Public First estimates that the business cloud adoption rate in the U.S. is significantly higher than that in other advanced economies such as Canada, the UK, Germany or France.10

Business Cloud Adoption per Country11
Source: Public First polling data

Cloud adoption and spending has grown rapidly in the last decade. IDC estimates that the total global public cloud revenues grew by 29% in 2021,12 while Statista estimates that the total spending on global public cloud has grown by a factor of around three over the last five years.13

In this study, Public First estimates 70% of businesses using cloud services in the U.S. increased their total spend on the cloud in the last five years, including 30% whose spending has grown by over 25%.

Based on the polling, Public First estimates that:

51%
of U.S. businesses use some form of cloud services
46%
of women-owned businesses use some form of cloud services
58%
of minority-owned businesses use some form of cloud services, including 61% of Latino-owned businesses.

Many businesses still have yet to take full advantage of the cloud

Despite a high overall cloud adoption rate at country-level, there remains clear disparities in adoption rate of cloud services in the U.S. across different states, industry sectors, and business sizes.

The adoption rate of cloud services varies across states. For the five largest states, based on the size of their economy and population, business’ IaaS/PaaS cloud services adoption rate ranges from 18% in Texas to 36% in New York.14

Cloud adoption by state
Source: Public First polling data

The cloud adoption rate for U.S. businesses also varies across industry sectors. Previous data from the U.S. Census Bureau suggests that U.S. businesses in finance and insurance are three times more likely to use cloud services than those in the agriculture sector.15

Relative cloud adoption (2018, US Census Bureau, 1 = All Sector average)
Source: Annual Business Survey: Extent of Technology Use of Employer Firms

More recent data from the U.S. Census Annual Business Survey in 2020 further underscores the disparity in cloud adoption across sectors. For example, businesses in professional, scientific and technical services are over 70% more likely to use cloud servers than those in the retail sector.

Businesses using cloud for different workflows (2020, US Census Bureau, % of businesses)
Source: Annual Business Survey: Technology, Financing, and Management Practices Characteristics of Businesses: 2020

There is also a significant gap in the rate of cloud services adoption between businesses of different sizes. According to Public First’s polling, businesses with fewer than 10 employees are half as likely to take advantage of cloud services as businesses with over 250 employees.

Cloud adoption by size of business
Source: Public First polling data

The future economic opportunities from cloud computing

Public First estimates that cloud service adoption is responsible for around 1.5% of the U.S. GVA. Assuming that the impacts of cloud services follow a standard S-curve technological diffusion pattern, Public First projects a 2.5% increase of U.S. GVA by 2030 at the current level of cloud services adoption.16 The central scenario assumes that complete adoption of cloud services takes 30 years, based upon the experience of other businesses adopting general purpose technologies. However, we also provide scenarios where it takes alternatively 20 years (the shorter adoption scenario), with the full benefits of cloud adoption exploited sooner, or 40 years (the longer adoption scenario), in which the benefits of cloud have much longer to play out.

GVA impact of cloud computing in the US
Source: Public First polling data

Accelerating the rate cloud service adoption could yield significant economic benefits:

  • We estimate that an increase in business adoption of cloud services across all states to the current level in New York (36%) would increase U.S. GVA by around $52 billion.
  • If 10% more U.S. businesses with fewer than 250 employees adopted cloud products and services, the U.S. GVA would grow by over $26 billion.

Other estimates of the financial benefits of cloud computing

Our estimates align well with previous studies, which generally have found that cloud services deliver significant savings for companies or productivity improvements for the wider economy:

  • McKinsey (2021) estimates that the benefits of the cloud could create over $1 trillion in value for businesses globally by 2030, including through IT cost optimization and innovation-driven growth.
  • Public First (2021) foundin Canada that 84% of AWS users reported saving money when using the cloud, compared to using traditional on-premises servers, with the top 10% of AWS customers seeing a return on investment (ROI) of over 8 to 1.
  • Public First  (2021) foundin the UK that AWS customers were receiving a return of £2 for every £1 they spend on AWS, with the top 10% of customers receiving a return of £10 or more.
  • NZIER (2020) found that a 20% increase in the adoption of cloud computing by businesses in New Zealand could boost GDP by 1.2 to 2.1%.
  • OECD  (2019) found that the use of cloud computing leads to an 8.6% improvement in firm productivity.
  • Deloitte (2018) found that public cloud customers achieve a net return on investment of $2.5 for every $1 invested in cloud services.
  • IDC (2018) estimated that AWS customers achieve a 51% lowered cost of operations and a return on investment of 637% over the first five years of adoption.
  • European Commission / Deloitte (2016) estimated that taking up cloud computing led to reduction of between 20% and 50% in total IT costs.

Data security benefits of cloud computing

Keeping data secure is a priority for businesses of all sizes. In 2021 alone, IBM estimated that the average cost of a data breach in the U.S. is $9.05 million per incident17 while the total cost of cybercrime has grown to over U.S.$1 trillion globally.18 For this study, 15% of SME survey respondents said that they had experienced some form of data loss in the last three years, 10% of hacking or a security breach, and 9% of ransomware or malware. Therefore, both businesses and public sector organizations are increasingly recognizing the importance of working with trusted vendors to keep their most valuable data secure.

By building on the cloud, businesses can automate manual security tasks, while maintaining ownership and ensuring privacy of their data, helping them feel confident that their data is safe:

94%
of cloud users responded that they are confident that cloud services help them back up data and avoid data loss.
94%
of cloud users responded that they are confident that cloud services are helping them ensure digital security and protect them from external hacking.

Lacework, a software company based in California, offers cloud-based solutions to new security problems, such as basic visibility of cloud activity for auditing or compliance reporting and using machine learning to monitor potential threats, vulnerabilities, misconfigurations, and other unusual network activity. Lacework’s use of AWS cloud services enables it to process up to 4 gigabytes of data per second from hundreds of thousands of sources, and store around 8.2 PB of data in the cloud, freeing up capital to invest in further innovation and staff productivity and scale rapidly to meet new markets. The company’s use of AWS Elastic Compute Cloud has helped Lacework to save 66 percent on computing costs. Lacework’s use of cloud services enabled the company to grow its customer base by more than a factor of five in 2021, and expand its business globally, with operations in Europe, Asia, and Australia.

Jobcase is an online community dedicated to empowering and advocating for the world’s workers, and the third-largest destination for job search in the United States. Headquartered in Massachusetts, the company uses AI and machine learning through AWS’s Redshift Machine Learning (ML) cloud-based service to generate job search recommendations that connect workers to employment opportunities. Jobcase’s data analysts and database developers used cloud services to create, train, and apply new ML models, which improved its ability to make billions of predictions in 15 minutes versus 4-5 hours. These improvements helped the company increase revenue and improve relevance of recommendations for over 10 million active job seekers per day.  

Persefoni is an Arizona-based SaaS platform that enables enterprises and financial services institutions to measure, analyze, plan, forecast, and report on their carbon footprint across their operations and portfolios in a centralized, cloud-based application. Persefoni’s cloud-based platform, which utilizes AWS services, accounts for Scope 1, 2, and 3 emissions with compliance to global greenhouse gas accounting standards, and enables users to pass audit and assurance processes for disclosure and reporting purposes. As of November 2022, Persefoni has enabled its customers to calculate more than 6.5 gigatons or 7.2 billion metric tons of CO2.

parallax background

Growth and Innovation

How cloud services support innovation

Cloud services not only enable businesses to offer products and services at lower cost, but also structure themselves in completely different ways, enabling innovation.

78% of U.S. businesses using cloud services in Public First’s survey reported that their use had made it easier for them to innovate and bring new products to market. Even more strikingly, 74% of cloud service users agreed that their business or operating model would not be possible without cloud services.

There are multiple channels through which cloud services enable innovation.

  • Speeding up development. For software developers, building on the existing software stack and infrastructure of cloud services providers, such as AWS, can significantly accelerate their productivity. On average, Public First estimates that cloud services help reduce the time needed to develop new software by around 25%.
  • Making it easier to experiment. This faster speed of development makes it much easier for businesses to prototype new products, to move to new models of continuous deployment and to make better data driven decisions. 89% of cloud service users agreed that it helped them make their business model more agile.
  • Supporting high compute or data intensive needs. From Hollywood special effects studios to scientific research, cloud services enable businesses to flexibly tap into significant pools of computing power. The cloud, for example, proved a crucial role in allowing Moderna to rapidly complete the sequencing for its mRNA COVID-19 vaccine in just two days.19 74% of cloud service-using businesses told us that the use of cloud services had made it easier to adopt new data intensive technologies, such as AI and big data.

Cloud tools are making businesses more agile

The economic benefits of online software tools running on the cloud, or SaaS, is also significant. From tools that almost every business uses (email, accounting) to more advanced tools such as Enterprise Resource Planning (ERP) or Customer Relationship Management (CRM) to dedicated sector specific tools, delivery and access of online software tools and applications via the cloud is increasingly overtaking traditional software workflows.

For businesses, cloud-based software tools and applications offer multiple advantages over the traditional model of installing and maintaining software on-premises:

  • Making it easier to join together data. 73% of businesses agreed that cloud tools had made it easier to connect and share information across different parts of the business. This in turn can help improve data reliability, security and make it easier to digitize existing business processes.
  • Enabling more flexible ways of working and collaboration. Cloud tools make it much easier to seamlessly move between a work desktop, home laptop, or smartphone.
    • The equivalent of 770,000 U.S. businesses said that it would have been impossible to keep their businesses operating during COVID-19 without online or cloud tools.
    • 21% of U.S. businesses reported that it would have been significantly more difficult or complicated to operate without online or cloud tools.
    • 63% of businesses agreed that online or cloud tools made it easier for workers to keep collaborating throughout COVID-19.
    • 68% of large businesses indicated that a significant proportion of their staff would likely continue to work remotely moving forward.
  • Boosting the productivity of businesses. Using tools such as ERP or CRM is associated with a 10% to 25% improvement in business productivity.20 Extrapolating from this, the indirect economic impact of cloud services through delivery of SaaS could be at least two times the estimated direct impact from IaaS and PaaS on the U.S. economy ($382 billion).

On average, businesses in the survey said that they currently used two to three cloud tools, but around 12% of businesses said that they were using more than ten cloud tools.

Business functions supported by cloud tools
Source: Public First polling data

Cloud tools, or SaaS, were associated with significant benefits for businesses:

  • 76% of businesses using cloud tools said that they had helped increase revenue
  • 61% of businesses using cloud tools agreed that the costs of starting a business have reduced substantially because of online tools
  • 49% of businesses using cloud tools said that they had made it easier to improve their services.
  • 43% of businesses using cloud tools said that they had made it easier to scale their business.

Based on the data from the polling, Public First identified clear trends whereby companies that reported greater use of cloud tools were more likely to report higher revenue growth in the previous year. In particular, companies that were fully digitally native and reported using cloud tools for over ten purposes were particularly likely to report recent growth, with an average of 6% revenue growth in 2021 (see graph below).

Average Revenue Growth by Number of Cloud Tools Used
Source: Public First polling data

Proscia is a health care start-up, based in Pennsylvania, founded by a team from Johns Hopkins University and the University of Pittsburgh. The company developed a digital pathology software solution, Concentriq, that runs on AWS cloud. Whereas cancer research and diagnosis has been historically performed by human pathologists examining glass slides through a microscope, Proscia’s software solution is helping labs worldwide shift to computational pathology using artificial intelligence to analyze digitized pathology images. Concentriq provides researchers and pathologists with a single interface for reading images, rendering diagnoses, managing imaging data, and collaborating instantly and remotely with experts around the world.

Arch Platform Technologies is a small business that runs an AWS cloud-based visual effects studio utilized by film and movie studios. By operating its studio in the cloud, Arch and its clients have been able to drastically reduce their capital expenditures that would typically be spent on equipment, space, networking, cooling, and maintenance. Running in a cloud environment has also changed the way production studios can hire talent – without geographical location as a hiring factor, studios can choose the best talent regardless of where they’re located. Artists are also able to sign on with more studios around the world.

Seafloor Systems, a California based startup, builds and manages a fleet of autonomous robotic boats, which are used by researchers and hydrographers to survey and collect data to chart bodies of water worldwide. Using AWS Internet of Things (IoT) solutions and AWS RoboMaker, Seafloor Systems is able to simulate, test, and securely deploy robotic applications, while pushing updates to robots in the field. Use of cloud technologies also enabled Seafloor Systems to perform simulations during the COVID-19 pandemic as well as deploy applications to robots in remote locations in times ranging from minutes to 1 hour, compared to the days or weeks previously. The company estimates that it saves approximately $1,500 per robot deployment by using AWS.

parallax background

Small and Medium Enterprises and cloud services

The importance of cloud services for small and medium enterprises

Small and medium enterprises (SMEs) are businesses with less than 500 employees. While adoption of cloud services among SMEs often lags behind larger enterprises, those SMEs that use cloud services reported significant benefits. The flexibility and scalability of cloud services allows small businesses to share in the wider benefits of R&D and investment by cloud platforms, reducing the barriers to entry to develop a product or enter a new market. 85% of cloud-using SMEs agreed that the use of cloud services had made it easier for them to compete with larger companies.

According to Public First’s polling, the reported savings and additional revenue from SMEs led on average to higher Returns on Investment (ROI) from cloud services spending than those reported by larger enterprises.

Return on Investment from Spending on Cloud Services (IaaS or PaaS), by business size
Number of EmployeesMean Return on Investment
0 to 93.3
10 to 193.2
20 to 993.7
100 to 2491.6
240 to 5001.6
500+1.6

On average, SME cloud service users self-reported direct savings and revenue:

  • Saving an average of $34,000 a year by investing in cloud services
  • Earning an additional $24,000 in revenue a year from their investment in cloud services.
Average reported savings and additional revenue from cloud services (IaaS or PaaS) by business size, measured by number of employees
Employee Size BandCloud Savings ($)Cloud Revenue ($)
0 to 927,73413,197
10 to 1940,00831,889
20 to 9976,95093,470
100 to 24970,222113,602
250 to 49998,132149,458

Cloud-using SMEs reported multiple ways in which cloud services helped them earn additional revenue:

60%
said that cloud services enabled their company to significantly increase their ability to scale up.
59%
said that cloud services enabled their company to be far more agile with the goods or services they offer.
47%
said that cloud services enabled their company to offer new kinds of products.
39%
said that cloud services had enabled their company to access global markets.

Cost savings and revenue generated from the use of cloud services can free up resources for wider job creation and reskilling. Of the cloud-using businesses that reported increased revenue in the previous year, 45% of those businesses invested in hiring more staff, and 49% invested in workforce training.

Many of the benefits from cloud services are particularly important to cloud-using SME with particularly fast revenue growth,21 of which:

85%
said that the use of cloud services had enabled them to significantly increase capacity or scale
85%
said that the use of the cloud services is increasingly important to stay competitive in their market segment
85%
said that the use of cloud services had made it possible for them to take their business global

The benefits of cloud services for SMEs

The top-three benefits of cloud services reported by SMEs were increased reliability (91%), automatic back-ups (91%), and higher security (90%).

How important or unimportant are the following benefits of cloud computing to your company?
Source: Public First polling data

By giving SMEs in the U.S. access to the economies of scale in computing resources that were once only available to large businesses, cloud services make it possible for SMEs to rapidly grow, work more flexibly, and deploy new software solutions. Of the cloud-using SMEs polled for this report:


57%
said that cloud services had helped improve their company’s business resilience
57%
said that cloud services had helped make their company’s business model more agile
50%
said that cloud services had helped increase their company’s customer base

In addition, cloud services also help an increasing number of SMEs boost productivity through digital transformation, automating existing processes, developing new challenges and monitoring data more systematically. Of the SME cloud users in the business poll:


47%
used cloud services to replace paper or manual processes with digital systems
42%
used cloud services to automate existing business processes
42%
used cloud services to develop new types of digital applications
38%
used cloud services to start monitoring new or existing metrics in real time

There remains, however, significant potential to further increase productivity through digitalization among SMEs. A recent report by the Bipartisan Policy Centre, drawing on Public First polling, found that microbusinesses in particular were more likely to be conservative with their use of digital tools, which tend to use less tools than other SMEs.22

To support their use of cloud services, SMEs reported multiple forms of complementary investment:


59%
said they had invested in digital skills training for staff
44%
said they had invested in hiring specialists with cloud skills
44%
said they had invested in developing new internal applications

On average, SMEs cloud service users reported spending $10,000 to $30,000 in complementary investments to cloud computing.


Echelon Fitness is a Tennessee-based company that provides smart, connected exercise equipment in the United States, Canada, France, Germany, and the United Kingdom, through which users can access live, on-demand, studio-quality workouts through a cloud-based fitness application hosted on AWS. Live classes are distributed globally using Amazon CloudFront, a fast content delivery network service that securely delivers data, videos, and applications to users globally. Use of AWS cloud allowed Echelon to scale its offerings when the company experienced exponential growth during the COVID pandemic in 2020, while also reducing costs. In December 2020, the company hosted over 2 million virtual classes, compared to 800,000 classes in 2019, each of which can handle 10,000 or more attendees without issue. These live classes were distributed globally using Amazon CloudFront, a fast content delivery network service that securely delivers data, videos, applications, and APIs to users globally with low latency and high transfer speeds.

ĪTEGRIA, an Illinois-based small business recently acquired by RIA in a Box, utilized Amazon Workspaces, a cloud based Virtual Desktop Infrastructure solution, to provide compliance solutions to small and midsize financial advisory firms. The company was able to create secure, individualized environments for its clients, increasing security and simplifying compliance with the U.S. Securities and Exchange Commission (SEC). ĪTEGRIA has over 120 virtual private cloud environments and about 2,000 cloud desktops. The company’s ability to scale contributed to annual growth of 20 percent a year.

Digital skills shortages are acting as a constraint on increasing cloud adoption

According to the polling, 85% of all businesses surveyed said that digital skills were important to the day to day operation of the businesses. However, the lack of staff with necessary cloud digital skills remains a major barrier that prevents U.S. businesses from fully realizing the benefits of cloud services. This can be particularly important for medium sized businesses. 63% of medium sized businesses23 who were currently not using cloud services said that the shortage of trained staff was an important barrier to cloud adoption.

The lack of workers with digital skills also poses a challenge for companies that are using cloud services – 58% of SME cloud service users said that a shortage of digital skills was a significant constraint to their business being able to either increase its revenue or reduce costs.

In December 2020, AWS committed to help 29 million people globally grow their technical skills with free cloud computing training.24 As part of this, the company offers a library of over 500 free courses, a range of AWS Certifications, and the AWS re/Start program to help people from underrepresented communities move into tech jobs.

parallax background

Appendix A) Estimated GVA impact of the cloud by state

These results are based on our national level estimates for economic impact for different sizes of businesses, apportioned to each state based on their number of businesses by business size class and adjusted for their wider region’s level of adoption of cloud.


StateGross Value Added generated by the cloud ($ bn)Estimated Jobs supported by the cloud
Alabama4.944,114
Alaska1.06,770
Arizona7.156,922
Arkansas3.029,076
California48.0277,000
Colorado7.554,794
Connecticut4.225,103
Delaware1.911,263
District of Columbia2.714,174
Florida20.4165,037
Georgia11.484,744
Hawaii1.713,143
Idaho2.119,769
Illinois14.8101,363
Indiana7.360,002
Iowa3.729,647
Kansas3.428,471
Kentucky4.541,444
Louisiana5.041,179
Maine1.412,180
Maryland8.555,769
Massachusetts8.652,201
Michigan9.979,392
Minnesota6.851,997
Mississippi2.727,879
Missouri6.354,578
Montana1.412,939
Nebraska2.519,423
Nevada3.829,714
New Hampshire1.712,532
New Jersey12.780,878
New Mexico2.319,329
New York29.3152,265
North Carolina11.187,368
North Dakota1.29,519
Ohio12.297,007
Oklahoma4.136,847
Oregon5.240,946
Pennsylvania16.5124,684
Rhode Island1.29,445
South Carolina5.246,394
South Dakota1.29,787
Tennessee7.562,248
Texas30.2217,127
Utah4.032,459
Vermont0.76,661
Virginia10.073,481
Washington10.259,407
West Virginia1.816,226
Wisconsin6.757,061
Wyoming0.97,262

Appendix B) Methodology FAQ

What does your study measure?

The core measure in our study is an estimate of the economic activity, or gross value added (GVA), generated by cloud services in the U.S.

Our priority was to look at the benefits enjoyed by customers of cloud services. We did not seek to measure the direct, indirect, or induced impact of the investment by cloud providers in data centers or offices in the U.S., or impact of sales by cloud service providers.

As is standard practice in economic impact studies, our results look only at the gross economic impact of cloud services rather than comparing it against a hypothetical where the cloud does not exist.


What is the main source of data for the report?

In order to estimate the impact of cloud computing in the U.S., we:

  • Ran an in-depth business poll of 2,000 senior business decision makers from across the U.S. Working with multiple panel providers including Dynata, MSI and EMI Research Solutions, Public First used a combination of quotas and weighting to ensure the final possible sample was representative of the U.S. business by business class size and cloud adoption. This sample was used to look at the adoption of cloud computing and AWS among different business types, as well as exploring the wider usage of digital tools (SaaS) and barriers to greater adoption of cloud services. Please note that the survey asked about both online and cloud tools in most questions, which is why both are mentioned throughout the report.
  • Ran a second booster poll to increase our sample of cloud service-using businesses. This second poll focused only on businesses that confirmed they used IaaS or PaaS, and was used to gain more granular information on the business impacts from using AWS and other cloud providers. In order to reduce confusion about the type of cloud that businesses were using, which can be particularly prevalent among small business users, Public First ran multiple checks on the data, including asking all self-described AWS users to explain in one sentence what they used the service for. Respondents who could not answer this were excluded from our final samples.

As part of our survey, we asked businesses to estimate:

  • Their business revenue, and its average growth in the last three years.
  • Their total spend on cloud services.
  • The amount they saved from using cloud services as compared to in-house IT.
  • The additional revenue they earned from using cloud services, not including the above cost saving.
  • Their level of confidence in the latter estimate.

All our estimates are self-reported and weighed by level of confidence.

For example, in order to estimate the savings of businesses using cloud services, we asked:

In an average year, roughly speaking, how much do you estimate your business saves by using a cloud service for your IT infrastructure, as compared to more traditional on-site servers?


How many businesses did you speak to?

The breakdown of our sample size is as follows:


Headcount of BusinessSample size in representative sampleSample size in cloud users sample
1-497162
5-928846
10-49288226
50-9936161
100-24936230
250-49990282
500+90893
Total17991901
SMEs (under 500)17091007

What was the breakdown of your sample by industry?

The table below summarizes the number of businesses we spoke to by sector. Given that many of these samples are not large enough to be statistically significant, we do not break out per sector data in this report.


 Sample size
Accommodation and Food Services27
Administrative and Support and Waste Management and Remediation Services19
Agriculture, Forestry, Fishing and Hunting20
Arts, Entertainment, and Recreation60
Construction199
Educational Services82
Finance and Insurance137
Health Care and Social Assistance117
Information118
Management of Companies and Enterprises54
Manufacturing131
Mining, Quarrying, and Oil and Gas Extraction13
Professional, Scientific, and Technical Services239
Public Administration49
Real Estate and Rental and Leasing69
Retail Trade152
Transportation and Warehousing25
Utilities34
Wholesale Trade40

How was the polling data turned into an estimate of the economic activity created by cloud services?

In order to generate our headline estimate of cloud services’ economic impact, we first created an estimate of the ratio of economic activity created to cloud spend for each business:

  • We converted estimates of additional revenue into additional gross value added by applying a GVA effects multiplier, drawn from the latest OECD input-output tables. This allows us to account for both additional costs for each business from servicing new revenue enabled by cloud services, and additional economic activity enabled in each company’s wider supply chain.
  • We added this to each company’s estimate of cost savings, and then divided by their estimated spending on cloud services.

As an additional check, we combined and averaged out our estimates of cloud service spending drawn from the polling data with third party estimates of total spending on IaaS and PaaS.

In order to generate an estimate of ROI for each business:

  • We converted self-reported estimates of additional revenue into profitability by applying an average operating margin for their sector.
  • We added this to each company’s estimate of cost savings, and then divided by their estimated spending on cloud services.

In order to aggregate our estimates for the U.S. economy as a whole, we averaged our estimates for ten size classes of business:

  • Sole traders
  • 2-4 employees
  • 5-9 employees
  • 10-24 employees
  • 25-49 employees
  • 50-99 employees
  • 100-249 employees
  • 250-499 employees
  • 500-999 employees
  • Over 1000 employees

Where possible, we used the average directly from our polling data for each size class. For the ROI estimates, where the sample size was too small to allow for this, we instead used a grouped estimate for ROI for businesses across different size classes.

We then multiplied each average by:

  • The Bureau of Economic Analysis estimate of the number of businesses in each size class.
  • The proportion of businesses in each size class using cloud services, taken from our polling.

In order to estimate the economic activity created for each state, we:

  • Apportioned out the national impact by that state’s share of total U.S. GVA.
  • Applied an adjustment to this value, based on the region’s relative cloud share, drawn from our polling.

In order to estimate the potential impact of higher adoption of cloud services, we assume that the value per cloud user remains constant, and total the potential aggregate impact.


How did you estimate the total number of jobs in companies enabled by the use of cloud services?

In order to estimate this, we estimated and multiplied together for each business size class:

  • Their average number of employees
  • The proportion of businesses using cloud
  • The proportion of cloud users who strongly agreed that their business or operating model would not be possible without cloud.

How did you estimate the potential future impact of the cloud on economic growth?

To estimate this, we calibrated a logistic S-curve diffusion model against:

  • Our existing estimate of the total economic impact of cloud services for the U.S.
  • A central assumption that complete adoption will take 30 years, based upon the experience of other businesses facing general purpose technologies. In the conservative and optimistic scenario we instead assume that cloud adoption will take 20 or 40 years respectively.

In order to keep the calculation tractable, we conservatively assumed that the average benefit per company remained constant over time. While early adopters are likely to have gained a higher marginal return from adopting cloud than the median firm, the benefits from cloud services are also growing over time as technology and the wider ecosystem improve, enabling. More use cases.

How reliable are self-produced estimates for this kind of data?

Given the timeline and scope of our study, we believed that using an anonymous online business survey was the most cost proportionate research methodology.

However, as with all self-reported data, polling data comes with inevitable limitations:

  • Respondents to an online survey may not be representative of businesses a whole.
  • Businesses may misremember, simply not know, or inaccurately estimate key financial data.
  • Many business leaders may poorly understand the cloud or how their business uses it.

In order to guard against these possible shortcomings, we took these steps:

  • We sourced data from multiple panel providers, including Dynata, MSI, EMI, and Cint.
  • As part of our standard polling quality checks, we excluded respondents who answered too quickly, ‘straight lined’, failed attention checks (“Please select this option if you read this”) or did not confirm that they were a senior business decision maker.
  • We manually reviewed every survey response given to the open text questions, excluding responses where respondents give contradictory, confused or meaningless answers.
  • We weighted the cost and revenue estimates by the respondents’ self-reported level of confidence in their estimate:
    • “Very confident” - 1
    • “Somewhat confident” - 0.66
    • “Not very confident” - 0.33
    • “Not at all confident” - 0
    • “Don’t know” - 0
  • We included only respondents who confirmed that they bought services from a cloud provider, with a lengthy explanation to try and head off any confusion:

“Cloud computing is the on-demand delivery of IT resources over the Internet with pay-as-you-go pricing. Instead of buying, owning and maintaining their own data centers and servers, organizations can now acquire technology such as compute power, storage, databases and other services on an as-needed basis from an external cloud provider through the Internet. It is similar to how consumers flip a switch to turn on lights in their home and the power company sends electricity.  Capacity can grow or shrink instantly and organizations only pay for what they use.”

  • After respondents had read this definition and confirmed that their business used cloud computing, we then only included those who went on to confirm that they used cloud computing for IaaS / PaaS, rather than SaaS.


  1. Gross Valued Added is the value that producers have added to the goods and services they have bought.
  2. As part of cloud services, we are including value created by both IaaS and PaaS.
  3. Unless otherwise specified, by SME we mean a business with fewer than 500 employees.
  4. https://itif.org/publications/2021/06/01/secrets-cloud-computings-first-stage-action-agenda-government-and-industry
  5. https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-145.pdf
  6. Raj Bala, “The Magic Quadrant for Cloud Infrastructure and Platform Services,” Gartner, March 2, 2020, accessed September 29, 2021, https://blogs.gartner.com/raj-bala/2020/03/02/magic-quadrant-cloud-infrastructure-platform-services/.
  7. https://itif.org/publications/2021/06/01/secrets-cloud-computings-first-stage-action-agenda-government-and-industry/
  8. https://www.mckinsey.com/~/media/McKinsey/Featured%20Insights/Artificial%20Intelligence/Notes%20from%20the%20frontier%20Modeling%20the%20impact%20of%20AI%20on%20the%20world%20economy/MGI-Notes-from-the-AI-frontier-Modeling-the-impact-of-AI-on-the-world-economy-September-2018.pdf
  9. https://www.statista.com/outlook/tmo/public-cloud/worldwide
  10. Based on recent polling carried out by Public First in Canada, the UK and across Europe.
  11. Based on recent polling carried out by Public First in Canada, the UK and across Europe.
  12. https://www.idc.com/getdoc.jsp?containerId=prUS49420022
  13. https://www.statista.com/outlook/tmo/public-cloud/worldwide
  14. The polling sample was only large enough to be representative for the five largest states, with it only being possible to analyze cloud uptake at a regional level for other parts of the country.
  15. Annual Business Survey: Extent of Technology Use of Employer Firms.
  16. Public First modeling. For more details, please see the methodology appendix.
  17. https://www.ibm.com/security/data-breach
  18. https://www.mcafee.com/enterprise/en-us/assets/reports/rp-hidden-costs-of-cybercrime.pdf
  19. https://aws.amazon.com/solutions/case-studies/innovators/moderna/
  20. See Table A for more details of relevant estimates.
  21. Defined as those growing by 6% a year, or above.
  22. https://bipartisanpolicy.org/report/small-businesses-go-digital/
  23. Defined as business with 100-499 employees.
  24. https://www.aboutamazon.com/news/workplace/amazon-to-help-29-million-people-around-the-world-grow-their-tech-skills-with-free-cloud-computing-skills-training-by-2025