Published in


The State of Automation: Unlocking Unstructured Data

We continued our “State of Enterprise Tech” Webinar Series this week spotlighting “The State of Automation” with a panel of experts from the startup and corporate ecosystem, including Harald Collet, CEO and Co-Founder of Alkymi, Preeti Asthana, Director of Innovation & Partners, Data & Analytic Services at Aon, and Elizabeth Granda, Director of Digital Automation at CIT.

During the panel, we discussed how Fortune 500s are thinking through automation strategies today and in the future, and the biggest challenges for adoption. Check out our YouTube here for a recording of the full webinar and see a recap of our top 3 burning automation questions answered below:

How do you identify an opportunity to be automated?

“Technology that is driving growth is definitely a priority vs. automating an existing process.” — Preeti Asthana, Aon

A lot of companies focus on routine resource-consuming processes that move data from one place to another. But what’s more important is focusing on automation processes that can actually improve data as an asset.

“Automation will fail unless tied to business strategy.” — Elizabeth Granda, CIT

Elizabeth further explained that identifying the best opportunities all comes down to the company’s overall business strategy and what they’re looking to accomplish. As she pointed out, 5 years ago companies invested millions of dollars into RPA and then were “looking under rocks for things to automate” and then a few years later they had to “de-robot.” While they took out manual processes, they added complexity and cost to IT through these application dependencies.

One area where CIT has found great return on their investment is in test automation. CIT is a fast paced and changing environment, and recent acquisitions provided them new sources of data, so investing in test automation not only provided them with the ability to support the release of data and better understand it, but it also provided efficiencies back to the business and the technology teams with no harm testing and new release testing.

How do you measure success?

“Focus on the outcome…A lot of these initiatives are 70% business strategy and 30% technology.” — Harald Collet, Alkymi

When thinking through how to measure an automation project’s success, Harald’s observed a shift over the past couple years. Many enterprise customers used to focus on automating 100% of the process, whereas now it’s more important to focus on the automation’s outcome.

For example, Alkymi’s Data Inbox works to automate document understanding and free data from email and documents. One of their Fortune 100 customers came to Alkymi to automate their account rollover processes. This was a manual method of processing statements full of financial data to trigger a transfer of funds, and incorporating an automation component to eliminate that laborious first step allowed them the opportunity to improve their customer service and acquire more assets, faster.

According to Harald, enterprises are increasingly thinking through their larger, strategic vision to transform their processes. This makes the current automation environment more agile and puts automation startups in a better position to make an impact.

Where is the biggest automation challenge in the enterprise?

“Unstructured data is our biggest challenge. If you can solve the problem for unstructured data and leverage something that has AI and gives you the ability to more accurately get at that data and build on it, you are building something new you can grow with.” — Elizabeth Granda, CIT

While most robotic processing automation (RPA) tools are used to simply automate a defined business processes, unlocking the potential of unstructured data generates net new value that can create completely new solutions and boost productivity. Businesses are under extreme pressure to perform quickly in this new COVID-19 work environment and being able to free up workforce capacity can help companies better leverage their employee base for business-critical work.

Preeti doubled down on the challenge of unstructured data, elaborating that a few years ago, “[Aon] had a lot of data waiting to be manually entered into systems. But with new automation capabilities, we can now utilize about 95% of that unstructured data that was not being utilized previously and we can input those into underwriter models. That has had a major impact on how we look at analytics…”

For more insights on how automation is impacting the enterprise and opportunities for automation startups, see a full recording of our webinar below.

Stay up-to-date on upcoming “State of Enterprise Tech” Webinar Series’ topics and speakers by following our Enterprise Weekly Newsletter every Friday. Future webinar topics include Cloud, DevOps, and more.


Thank you to Kira Colburn for her work on this piece.




Work-Bench is an enterprise technology VC fund in NYC. We support early go-to-market enterprise startups with community, workspace, and corporate engagement. Sign up to get our digest of top content & industry news weekly:

Recommended from Medium

Can TTS change your life?

A Perspective of the Future

Google I/O 2018 Summary — Don’t be evil… be AI first

Intelligent AI Adoption: Rethinking the Dynamics of Adopting AI

Accelerate to AI, data-driven business with IBM Watson Data Platform

Artificial Intelligence technologies for the Diagnosis of Diseases: the Experience of Russian…

Anticipate financial regulatory trends is not magic… but science

Voice is the next big thing : Why Now?

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Kelley Mak

Kelley Mak


More from Medium

Organising for customer centricity & scale

From Data Geeks to Datanauts: Why we changed

Veza sheds some light on a dark corner of your business

IBM Cloud Code Engine with R