Watson’s prescription for knowing your data

Published on the 02/08/2023 | Written by Heather Wright

Watson's prescription for knowing your data

Pronto embeds IBM smarts into data security…

AI might be the sexy technology of the moment, but Jacqui Drew has a decidedly unsexy message for Australian and New Zealand companies: Know your data, where it is and who has access to it at all times. 

Drew, who is general manager of the BI focused business unit at Australian ERP vendor Pronto Software, admits it’s not a particularly trendy message – but it is a crucial one for local businesses spurred on by the current hype around ChatGPT and generative AI. 

“Data security has been around for a very long time, but certainly in the age of AI it is more and more important to establish that trust in your data and trust that you know who is accessing it,” Drew says.

“Being able to verify the information is key for a lot of businesses at this stage.”

“Knowing your data and knowing who has access, not just to the data you provide, but the data that is generated from you – the derivative information – and understanding who owns that derivative information is critical. Is it the publicly available algorithm, or is it something that should be owned by the organisation itself. 

And then understanding how you develop trust in what that algorithm is telling you. Whether it is just restructuring conversations, whether it is restructuring communications with customers and suppliers or whether it is deriving insights and linkages with data, how do you verify it?” 

She told iStart ChatGPT’s rise has had a profound impact in increasing interest across Australian and New Zealand businesses as everyone can now see what AI can provide businesses. 

It’s interesting because it is by no means the first, or the most powerful, algorithm out there, but it is the one that is most accessible, so it has really opened the conversation and changed the dialogue. 

“A lot more people are curious now about what AI can do for businesses and also the risks that ChatGPT and sharing sensitive context and IP can have on businesses.” 

That’s bringing the trust economy to the fore.  

What we are seeing with business intelligence is how do you trust the insights that are being generated and the data you are feeding into these algorithms. Being able to verify the information is key for a lot of businesses at this stage.” 

A big part of that is data governance and knowing what data is being used to derive the data insights.  

Several products on the market advertise that they have data insights and data insight generation using AI toolsets. 

IBM, which is a close long-term partner for Pronto Software, is among those companies, with Drew noting Pronto Software has been ‘playing’ with the insight derived toolset for about 10 years.  

“The difference I see in the marketplace is that a lot of the IBM toolsets have been designed for the organisation in mind which does not compromise information security,” she says. 

“Businesses, SMEs in particular, do need to be careful about where they’re giving their intellectual property and where they are giving their information.  

“This is not just about facts and figures, this is more about where they are feeding their email data and actually getting sentiment analysis – making sure that is keep on-prem and can’t be shared or compromised beyond their enterprise boundaries.” 

That issue of data sovereignty is a critical piece, Drew says.  

“A lot of businesses have a very profound understanding that their data security is key to their business, so understanding then what they can do within those confines to actually maximise their investment in AI, but also keeping a very close watch on what is happening in the industry. 

“Forecasting this five years into the future, I think we will have a large array of commercial toolsets that have that data sovereignty baked into it, that you can feed in and actually generate your own business insights without compromising your data.” 

Pronto Software has just launched a data integration suite of offerings, Pronto Xi BI Unlimited with IBM Cognos Analytics with Watson, which it says delivers advanced analytics and comprehensive insights. 

“We are leveraging the Watson AI engine that is built into Cognos Analytics, and using it against all variety of data which will then help businesses to generate insights and customer sentiment analysis from within the boundaries of the business. So they’re not outsourcing the toolsets and not providing a weakness to their data security.” 

The data can come from a range of sources, including Pronto Xi ERP, external financials systems, sales tools, online websites, weather and weather forecast, foot traffic trends, PLC controllers and IoT. 

The company says the integration addresses a critical challenge many businesses face – a limited focus on the scope of data that was deemed important. 

Drew says the system also allows businesses to become ‘truly self-reflective’, enabling them to feed in not only sentiments and analysis for customers, but how quickly the business is adapting – the metadata analysis of who is logging on when, how they derive impact and harnessing that to make it more efficient.  

“It is moving towards using the toolsets to be self-reflective, rather than just using it as a customer toolset,” she says.  

Drew says her conversations with customers are often focused around the capabilities to make normal processes more streamlined and more effective in their language capabilities and pattern recognition. 

“It is all about driving more efficiencies and catching trends before they are actually noticed. So using toolsets to pick up on the trends and subtle indicators that can impact business performance.” 

One area she says standing out as an area of opportunity is being able to use sentiment analysis to drive support desk interaction, identifying whether a person needs an automated response, can be directed to a knowledge base article, or whether their sentiment is such that it warrants someone picking up the phone and making a personal connection. 

“Those types of indicators have never really been available without human interaction. So having that as an automated process certainly has helped. 

“Even feeding a knowledge base into an algorithm so someone can converse with a chatbot to drive that information, so almost like a guided step – that is coming into the fray and the conversation quite a lot as well. 

“So, it’s using the standard toolsets and grabbing all the transactional and historic information we store in an ERP or support desk or CRM system and using that to supercharge the process and the interaction with the customer, even if it means a human has to pick up the phone rather than having an algorithm do the work.” 

If you’re think every Australian and Kiwi business is jumping into AI, think again. Drew says with the current economic conditions, companies are hesitant to be first movers. 

“They’re keen to see what commercial toolsets are available to them and how they can see demonstrated value, rather than experimenting and using budget without any guarantee of impact.” 

For those who have already seen the value of data driven decisions and have data governance in place already, adoption of the new technology can, however, be rapid, she says.  

“Those customers are on the front foot and they’re bringing the conversation to us and asking how they adopt this, how they get the real time bandwidth between what staff are doing, the transactions, the online toolsets and these insight generators?”

Post a comment or question...

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.


Thank you! Your subscription has been confirmed. You'll hear from us soon.
Follow iStart to keep up to date with the latest news and views...