What to Consider When Evaluating Automation Solutions

Rory O'Brien
Rory O'Brien
February 11, 2023
June 7, 2021
min read
What to Consider When Evaluating Automation Solutions

Over the last few years, companies all over the world have invested heavily in implementing and evaluating automation solutions — specifically platforms designed to streamline manual tasks or processes.

In fact, “heavily” may be an understatement. According to MarketWatch, the global market for workflow automation software solutions is expected to reach $29.26B by 2026 — with a CAGR of 21.4%. Additionally, Gartner reports that Robotic Process Automation is the fastest-growing segment of the global enterprise software market, growing 63.1% and reaching $1.3B in 2019.

Despite the investment, however, the world of workflow or process automation remains difficult to understand; certainly many who’re right now evaluating automation solutions misunderstand it.

It’s a world populated, for one thing, by many different kinds of platforms that purport to address common operational problems — they’re sold as a means of saving time, reducing cost, improving consistency, and facilitating collaboration — but in actuality, all solve more specific problems. In selecting the best automation platform, it is important to understand the differences in scope.

I know this because, in a previous life, I was a Sales and Marketing Ops consultant, and before that, I was the Director of Marketing Ops for a fast-growing startup. In those roles, I was tasked with identifying technology solutions that could do things like integrate data systems, streamline internal processes, and automate tasks. I spent untold hours shopping around and hacking solutions with virtual duct tape. And that experience taught me a lot — both about how the automation space is structured, as well as what the capabilities, applications, and limitations of the available solutions generally are.

As it turns out, almost every automation solution available today can be grouped in one of three high-level categories, each designed to solve a different specific problem. To truly tap into the potential of automation software — to use automation technology in a way that actually optimizes processes to increase efficiency, lower costs, and minimize change management — we need to apply the right solutions in the right situations.

If you, as I once was, are scouring the market for available automation software to help you address certain internal problems right now, identifying what these high-level categories are — and what companies are operating inside them — is an important place to start. It’s where I wish I would have started, to be sure. Here’s a quick breakdown of how to go about evaluating automation solutions.

Evaluating Automation Solutions: Integration Platform As a Service (iPaaS)

A primary category in the automation solution space is iPaaS, which is comprised of platforms designed to allow companies to connect — or integrate — the various, typically siloed SaaS apps or tools they use across departments and environments. It’s generally useful for moving data from system A to system B with “triggering” events.

What iPaaS is designed to do:

  • Allow users to utilize APIs to create, update, or delete data in SaaS systems.
  • Remove manual and redundant data entry in multiple systems.
  • Reduce data inconsistency across systems.
  • Add automated business logic based on “triggering” events when creating, updating, or deleting data.

What iPaaS is not designed to do:

  • Build a new app to automate a net-new process that does not yet involve systems (i.e. email-based workflows, people-centric workflows, etc.).
  • Maintain state. Since they are focused on connecting systems, they typically rely on the underlying systems to provide context.
  • Provide end-user capabilities (i.e. reaching out to users via existing or new interfaces like email or chat and following up to chase information).
  • Work with unstructured data like emails, documents, or freeform text.

iPaaS in a sentence: Platforms designed to automate manual processes necessitated by siloed applications.

Evaluating Automation Solutions: Workflow Automation — or, Business Process Automation (BPA)

Workflow automation solutions allow organizations to define and automate processes by building custom forms or apps atop linked workflows and business logic. It’s generally useful for organizations who want to build a net new app to automate what is currently a manual business process.

What Workflow Automation is designed to do:

  • Allow business teams to design, build, and deploy forms or apps without the need for developers.
  • Create custom tailored apps that fit the specific needs of the organization.
  • Automate processes that are currently handled manually via spreadsheets, email, etc.
  • Map out internal processes and automate specific steps of workflow logic.

What Workflow Automation is NOT designed to do:

  • Connect or integrate existing systems, as BPA platforms typically have very limited integration capabilities.
  • Automate system-level actions, such as repetitive data entry or record creation.
  • Build automated logic or workflows on top of existing systems, as they focus on building net-new apps.

Workflow Automation in a sentence: Platforms to allow business teams to build net new apps.

Evaluating Automation Solutions: Robotic Process Automation (RPA)

The third primary category in the automation space is RPA. RPA solutions utilize software robots, AKA bots, to automate routine tasks within software applications that are normally done by humans.

What RPA is designed to do:

  • Offload manual human-to-system tasks (i.e. entering physical forms to digital systems, scanning images, etc.).
  • Handle both structured and unstructured data, such as data collected through chat, voice messages, spreadsheets, or documents.
  • Automate tasks on top of hard-to-integrate systems (i.e. systems without APIs), which are handled today by human inputs.
  • Provide assisted or unassisted bots that either work independent of people or are dependent upon a human trigger (i.e. person pushing a button to activate a bot).

What RPA is NOT designed to do:

  • Integrate systems via APIs. RPA focuses on solving challenges with systems that don’t have APIs. This makes RPA less applicable to modern SaaS applications, as APIs are typically a faster and more consistent means of automating system-level tasks or workflows.
  • Create a new interface or app for people to use.
  • Automate workflows. Bots focus on automating specific tasks. To automate a complete workflow, you must string together multiple bots, systems, and people.
  • Maintain state. Since bots work at a task level, contextual data still must be maintained in a central database or underlying system.

RPA in a sentence: Platforms that allow companies to automate previously un-automatable tasks on legacy systems or with unstructured data.

A Critical Missing Element: People

Now that you see the different uses for each of these types of automation platforms, you can easily understand why operations and technology teams need to deploy each of these platforms at different times to solve manual or inefficient processes. It is important to apply the right platform at the right time.

However, even with all these types of automation platforms, there is one critical element that is still missing from these equations: people. While automating system-to-system connections with iPaaS, building net new apps with Workflow Automation platforms, or eliminating repetitive manual tasks with RPA all make processes more efficient, none of these platforms respect people as a key element of improving processes.

iPaaS platforms don’t solve how data gets from people to systems to begin with. (Salesforce Training cited lack of adoption as one of the most common reasons for the failure of CRM systems.) Workflow automation platforms have a similar challenge. Introducing net new interfaces forces people to learn how to use those new interfaces, which itself requires costly change management and education. RPA platforms, meanwhile, focus on reducing manual human dependency in processes, but do little to actually help people get more out of their enterprise systems.

We believe folks need to think differently about automation from the start. When evaluating automation solutions and strategizing around how to implement them, you must think first and foremost about the impact those solutions will have on your people. (Here at Tonkean, we have codified a framework to build processes around people, called People-First Process Design.)

My hope is, equipped with a deeper understanding of automation in general, along with a new appreciation of the importance of thinking people first, you’ll be able to shop smarter than I was able to when I started out, such that the improvements you bring to your ops teams and your internal processes are scalable and long-lasting.

Want to learn more about how Tonkean’s process orchestration platform can help your company? Click here.

Share this post
Read more posts
The Promise and Practical Potential of Intake Orchestration in Procurement
min read

The Promise and Practical Potential of Intake Orchestration in Procurement

If you work in procurement, it’s likely you’ve been hearing a lot about the promise and potential of intake orchestration software. But what exactly is intake orchestration, what does it look like in practice, and what makes it so exciting?
Read post
Procurement Ops Digest #4: Practical Applications of Process Orchestration in Procurement
min read

Procurement Ops Digest #4: Practical Applications of Process Orchestration in Procurement

The word of 2024 in procurement so far is “process orchestration.” Everybody—from procurement experts to executives at Goldman Sachs—is talking about it. But… why? What are they so excited about? Here's why.
Read post

Stay up to date

Get experts articles & updates to your inbox!

Create a process experience that works.

Maximize adoption, compliance, and efficiency.
Transform your internal processes with powerful AI and personalized experience. 100% no-code.