Skip to content

SaaS Development: What Is SaaS and What Matters Most?

Alexander Weipprecht 6 min read 09 June 2026 2 views
Software- & SaaS-Entwicklung
SaaS Development: What Is SaaS and What Matters Most?
Illustrative image · AI-generated

SaaS (Software as a Service) is a cloud delivery model where you use a finished application over the internet and pay a subscription for it. According to NIST SP 800-145, users manage neither servers, operating systems, nor the underlying infrastructure; they only rent the software itself.

Note on terminology: this article is about Software as a Service, not the Swiss ski resort of Saas-Fee.

What is SaaS? Software as a Service explained simply

SaaS means that software runs centrally at the provider and you simply use it in the browser or via an app, without installation and without your own servers. Responsibility for operation, maintenance, and updates lies entirely with the provider, not with you.

According to the NIST definition (SP 800-145, 2011), SaaS capability is "the capability provided to the consumer to use the provider's applications running on a cloud infrastructure." The consumer "does not manage or control the underlying cloud infrastructure including network, servers, operating systems, or storage." Typical SaaS examples according to IBM include Gmail, Dropbox, and Salesforce.

What does the NIST cloud model mean? The five characteristics

Cloud computing is the foundation of SaaS and defines how IT resources are provisioned. NIST describes it as "a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort."

NIST names five essential characteristics of cloud computing:

  • On-Demand Self-Service: resources are provisioned by the user as needed.
  • Broad Network Access: access over the network using standard devices.
  • Resource Pooling: resources are shared across multiple users.
  • Rapid Elasticity: capacity scales up and down quickly.
  • Measured Service: usage is measured and billed.

In addition, NIST defines three service models (SaaS, PaaS, IaaS) and four deployment models (private, community, public, and hybrid cloud).

What is the difference between IaaS, PaaS, and SaaS?

The difference lies in how much the provider takes on and how much you manage yourself. With IaaS, you rent only the raw infrastructure; with PaaS, a development platform; and with SaaS, the finished application. The following table summarizes the three service models according to NIST SP 800-145.

ModelWhat you rentWhat the provider managesTypical user
IaaSCompute power, storage, network; you run your own operating systems and software on topPhysical hardware, virtualizationIT/infrastructure teams
PaaSPlatform for developing and running your own apps with the provider's toolsInfrastructure, operating system, runtime environmentDeveloper teams
SaaSFinished application via browser or appComplete infrastructure, platform, and applicationEnd users and companies

With IaaS, according to NIST, the consumer can "deploy and run arbitrary software, which can include operating systems and applications." With PaaS, the consumer deploys "consumer-created or acquired applications created using programming languages, libraries, services, and tools supported by the provider." With SaaS, there is no infrastructure management at all.

What is multi-tenant architecture, and why is it the heart of SaaS?

Multi-tenancy means that a single instance of the software, including the database and hardware, serves multiple customers (tenants) at the same time. According to IBM, the decisive point is that each tenant's data is isolated from the others and invisible to other tenants.

The benefits are lower ongoing costs through shared maintenance and infrastructure, good scalability, and centralized updates that roll out to all tenants at once. AWS describes multi-tenancy as a model in which resources are shared to promote agility and cost efficiency. Important: according to AWS, SaaS is primarily a business model, not necessarily a specific technical architecture; even a "stack-per-tenant" model without shared infrastructure is valid SaaS, as long as it is centrally operated and uniformly versioned.

What advantages does the SaaS model offer companies?

The biggest advantage of SaaS is that companies do not need to operate their own IT infrastructure and get software that is ready to use immediately. The main advantages at a glance:

  • No infrastructure: the provider manages servers, operating systems, and storage (NIST).
  • Lower ongoing costs: maintenance and infrastructure are shared across all tenants (IBM).
  • Centralized updates: new features reach all users at the same time.
  • Fast scalability: capacity can be adjusted elastically (rapid elasticity per NIST).
  • Predictable costs: a subscription model instead of a large one-time investment.

What disadvantages and risks does SaaS have?

The central disadvantage of SaaS is dependence on the provider, since you control neither the infrastructure nor the update cycles. Because you do not manage the underlying infrastructure, according to NIST, you give up control over data storage, availability, and feature scope.

Other typical concerns include data protection and compliance (where is the data stored?), vendor lock-in when switching providers, customizability with standard SaaS, and dependence on a stable internet connection. With multi-tenancy, clean tenant isolation is also security-critical: the separation of customer data must work reliably.

What does SaaS development involve, and what matters technically?

SaaS development involves building a cloud-based application that serves multiple customers centrally, securely, and scalably. Unlike with traditional software, architecture, tenant separation, billing, and update strategy all need to be considered from the very start.

What matters most when developing SaaS products:

  • Tenant isolation: data from different customers must be securely separated (IBM).
  • Scalability: the architecture must be able to grow elastically (rapid elasticity).
  • Uniform versioning: all tenants run on one centrally operated version (AWS).
  • Subscription and billing logic: measured service requires measurable, billable usage.
  • Security and data protection: authentication, encryption, and GDPR-compliant data storage.

How big is the SaaS market? Is it worth building your own SaaS product?

The SaaS market is the largest segment of the public cloud and continues to grow at double-digit rates. Gartner forecast worldwide end-user spending on SaaS of 247.2 billion US dollars for 2024, an increase of around 20% over the previous year (as of: Gartner forecast, May 2024).

In total, Gartner expected worldwide public cloud spending of 675.4 billion US dollars for 2024 (2023: 561 billion US dollars), an increase of 20.4%. The cloud is also standard in Germany: according to the Bitkom Cloud Report 2024, 81% of companies use cloud computing, another 14% are planning or discussing its use, and only 5% reject it. Currently, 38% of IT applications run from the cloud, and in five years that figure is expected to reach 54%. Data basis: representative survey of 603 companies with 20 or more employees (survey weeks 12-19/2024).

Frequently asked questions about SaaS and SaaS development (FAQ)

What does SaaS mean, explained simply?

SaaS stands for Software as a Service and means you use finished software via subscription over the internet instead of buying it and installing it on your own servers. The provider operates, maintains, and updates the application; you simply use it in the browser.

What is the difference between SaaS and PaaS?

With SaaS, you use a finished application; with PaaS, a platform on which you develop and run your own applications. PaaS is aimed at developer teams, while SaaS is aimed at end users and companies that need software ready to use immediately.

Is SaaS the same as the cloud?

No. Cloud computing is the umbrella term, and SaaS is one of three service models within it, alongside IaaS and PaaS. NIST defines cloud computing with three service models and four deployment models (private, community, public, hybrid).

What is multi-tenancy in SaaS?

Multi-tenancy means that a single software instance serves multiple customers at the same time, while each customer's data stays isolated and invisible to others. This lowers costs and allows centralized updates for all tenants simultaneously.

Is SaaS always multi-tenant?

Not necessarily. According to AWS, SaaS is primarily a business model: centrally operated, with all customers on one version. Even a "stack-per-tenant" model without shared infrastructure can be valid SaaS, as long as it is centrally managed and uniformly versioned.

What are some well-known SaaS examples?

Typical SaaS applications, according to IBM, include Gmail, Dropbox, and Salesforce: tools used purely via browser or subscription, without customers running servers or updates themselves.

Is it worth building your own SaaS product?

The market is large and growing: Gartner forecast worldwide SaaS spending of 247.2 billion US dollars for 2024, with growth of around 20%. Whether your own product is worthwhile depends on your target audience, differentiation, and a viable, scalable architecture.

Note: this article is general, technical guidance on SaaS and SaaS development and does not replace individual architecture or legal advice. Market figures are marked as forecasts or point-in-time values (as of: June 2026).

Sources

Share this article

Stay up to date

Get the latest articles, insights and industry updates straight to your inbox.

Unsubscribe at any time. See our privacy policy.

Decide for yourself what Google shows you

Google lets you choose which sources appear more prominently in your search results: in Top Stories and in AI answers. Two clicks, and you see the sites you trust.

Add provimedia.de to my preferred sources

Ready for your AI competence certificate?

Get your AI certificate – flexible, online, with a documented certificate of participation for building AI literacy (Art. 4 EU AI Act).