Artificial Intelligence

Conclusion: Organisations are using chatbots as information assistants, advisors, and digital services channels. Most businesses start with generic chatbots (as virtual agents), but as the demand for customer communication grows, chatbots require integration with an increasing number of backend systems and improved scalability.

The reason why most chatbot ventures fail is the inability to recognise that the chatbot principle is simple, yet complexity of deployment rises sharply over time. In addition, chatbot design must align the business and target audiences, and both will evolve. This subtle shift over time is important as organisations need to learn the role, tone, specific purpose, and personalities of their chatbots based on actual usage and feedback.

Thus, starting small with continuous, incremental development is the best strategy for chatbot development. However, this iterative approach must balance the development of chatbots with business implementation, and must consider the attributes of the existing and future deployments.

Artificial intelligence (AI) is an emerging technology that can be applied across business lines and yield significant results when aligned with business priorities. Assessing the AI maturity of your organisation can assist in providing AI roadmaps and aid in developing strategies and business cases.

The purpose of this presentation kit is to provide an AI maturity model in the analytics space. The proposed maturity model can be applied to any type of industry. Log in and click the PDF above to download the 'Analytics Artificial Intelligence Maturity Model' presentation kit and discover:

  • An IBRS AI maturity model that provides the foundation to apply the existing AI technology where it matters to the business
  • Guidelines to evolve into the future, whereby only limited data is available to make informed decisions
  • Next steps for your organisation


Implementing machine learning operations (MLOps) is complicated by several challenges: the number of the stakeholders involved in a project; the shortage of people with the necessary skills; the scope of regulatory compliance; validation of the machine learning (ML) model; and model degradation issues. Considering how these challenges will be addressed is a vital precursor for the successful implementation of MLOps.

The Latest

18 August 2021: While natural language processing AIs are becoming increasingly accurate in how they respond to questions, their ability to explain how they arrived at their answers has been limited. As The Doctor reveals, confronting a rogue AI in the Green Death, ‘Why?’ remains, perhaps, the hardest question for machine intelligence. IBM’s AI Horizons Network is developing a method to enable AIs to explain their reasoning with a common sense data set.1 

Why it’s Important.

Today, virtual service agents, both customer facing and internal IT held-desks, are effective and very efficient FAQs. They can identify a context from natural language and then provide answers to questions, as well as provide follow up answers based on the original context. However, they cannot provide details as to how they arrived at any given answer, which generally leads to a request for human manual intervention.

Specialists who develop conversation virtual service agents, work around these limitations by programmatically refining the answers AIs have available (i.e. curating the FAQ) to include reasons. E.g. “Your transaction has been declined because of XYZ.” 

IBMs work to allow AIs to report back on their reasons, may not only minimise the programming effort needed to develop virtual agents, but allow them to report decision-making in ways that organisations have not considered. 

While AI development will remain a niche activity for most Australian organisations, AI will increasingly find its way into enterprise SaaS products. Natural language AIs coupled with machine learning over knowledge assets held in core enterprise systems will see a rapid increase in the use of virtual agents, both for internal and external services. 

Who’s impacted

  • AI specialists
  • Service automation / customer experience teams
  • ICT strategy leads

What’s Next?

The rapid improvements in AI quality, coupled with their integration into most enterprise SaaS products, will make them ubiquitous for customer service delivery within the next 2-5 years.

Organisations need to start exploring the AI service agent capabilities already available in their SaaS products, and develop plans for how to leverage such capabilities. The goal should not be to deliver an ‘all-singing and dancing’ virtual agent experience, but rather to incrementally introduce capabilities over time, learning how clients and staff wish to interact, and continually leveraging advances in technology as they become available. 

Related IBRS Advisory

  1. Chatbots Part 1: Start creating capabilities with a super-low-cost experiment
  2. Preparing for the shift from digital to AI-enabled transformation
  3. BMC Adds AI to IT Operations
  4. Trends for 2021-2026: No new normal and preparing for the fourth-wave of ICT
  5. Software Agents Maturity Model
  6. Artificial intelligence Part 2: Deriving business principles



1. COMMONSENSEQA: A Question Answering Challenge Targeting Commonsense Knowledge, 2019 Association for Computational Linguistics

The Latest: 

26 June 2021: Zoho briefed IBRS on Zoho DataPrep, it’s new business-user focused data preparation which is being included in its existing Zoho Analytics tool, as well as being available separately as a tool to clean, transform and migrate data. DataPrep is in beta, and will be officially launched on 13th July 2021.

Why it’s Important

Traditionally, cleaning and transforming data for use in analytics platforms has involved scripting and complex ETL (extract, transform and load) processes. This was a barrier to allowing business stakeholders to take advantage of analytics. However, several analytics vendors (most notably Microsoft, Tableau, Qlik, Snowflake, Domo, etc.) have pioneered powerful, drag-and-drop low-code ETL into their products.  

Zoho, which is better known for its CRM, has an existing data analytics platform with Cloud storage, visualisation and reports, and dashboards. While the product is not as sophisticated as its top-drawer rivals, it can be considered ‘good enough’ for many business user’s needs. Most significantly, Zoho Analytics benefits from attractive licensing, including the ability to share reports and interactive dashboards both within an organisation and externally. 

However, Zoho Analytics lacked a business-user-friendly, low-code ELT environment, instead relying on SQL scripting. Zoho DataPrep fills this gap by providing a dedicated, AI-enabled platform for extracting data from a variety of sources, allowing data cleaning and transformations to be applied, with results being pushed into another database, data warehouse and Zoho Analytics. 

All existing Zoho Analytics clients will receive Zoho DataPrep with no change to licensing.

However, what is interesting here is Zoho’s decision to offer its DataPrep platform independent of its Analytics platform. This allows business stakeholders to use the platform as a tool to solve migration and data cleaning, not just analytics. 

IBRS’s initial tests of Zoho DataPrep suggest that it has some way to go before it can compete with the ready-made integration capabilities of Tableau, Power BI, Qlik, and others. In addition, it offers less complex ETL than it’s better established rivals. But, that may not be an issue for organisations where staff have limited data literacy maturity, or where analytics requirements are relatively straightforward.

Who’s impacted

  • CIO
  • Development team leads
  • Business analysts

What’s Next?

The bigger take out from Zoho’s announcement is that ETL, along with all other aspects of business intelligence and analytics, will be both low-code, business-user friendly and reside in the Cloud. ICT departments seeking to create ‘best of breed’ business intelligence architectures that demand highly specialised skills will simply be bypassed, due to their lack of agility. While there will be a role for highly skilled statisticians, data scientists, and machine learning professionals, the days of needing ICT staff that specialise in specific reporting and data warehousing products is passing. 

Related IBRS Advisory

  1. Snowflake Gets PROTECTED Status Security Tick by Aussie Auditor
  2. IBRSiQ: Power BI vs Tableau
  3. Business-First Data Analytics
  4. AWS Accelerates Cloud Analytics with Custom Hardware
  5. IBRSiQ AIS and Power BI Initiatives
  6. Trends in Data Catalogues
  7. When Does Power BI Deliver Power to the People?
  8. Staff need data literacy – Here’s how to help them get it

The Latest

19 May 2021: Google has launched Vertex AI, a platform that strives to accelerate the development of machine learning models (aka, algorithms). According to Google and IBRS discussions with early adopters, the platform does indeed dramatically reduce the amount of manual coding needed to develop (aka, train) machine learning models. 

Why it’s Important

The use of machine learning (ML) will have a dramatic impact on decision making support systems and automation over the next decade. For the majority of organisations, ML capabilities will be acquired as part of regular upgrades of enterprise SaaS solutions. Software leaders such as Microsoft, Salesforce, Adobe and even smaller ERP vendors such as Zoho and TechnologyOne, are all embedding ML powered services into their products today, and this will only accelerate.

However, developing proprietary ML models to meet specific needs may very well prove critically important for a few organisations. Recent examples of this include: customise direct customer outreach with specific language tailored to lessen overdue payment, and creating decision support solutions to reduce the occurrence of heatstroke.

IBRS has written extensively on ML development operations (MLOps). However, the future of this disciplin e will likely be AI-powered recommendation engines that aid data teams in the development of ML models. In a recent example, IBRS monitored a data scientist as they first developed an ML model to predict customer behaviour using traditional techniques, and then used a publicly available tool that leveraged ML itself to build, test and recommend the same model. Excluding data preparation, the hand-coded approach took 3 days to complete, while the assisted approach took several hours. But more importantly, the assisted approach tested more models that the data scientist could test manually, and delivered a model that was 3% more accurate than the hand-coded solution.

It should be noted that leveraging ‘low-code’ AI does not negate the need for data scientists or the pressing need to improve data literacy within most organisations. However, it has the potential to dramatically reduce the cost of developing and testing ML models, which lowers the financial risk for organisations experimenting with AI.

Who’s impacted

  • CIO
  • COO
  • CFO
  • Marketing leads
  • Development team leads

What’s Next?

Prepare for low-code AI to become increasingly common and the hype surrounding it to grow significant in the coming two years. However, the excitement for low-code ML should be tempered with the realisation that many of the use cases for ML will be embedded ‘out of the box’ in ERP, CRM, HCM, workforce management, and asset management SaaS solutions in the near future. Organisations should balance the ‘build it’ versus ‘wait for it’ decision when it comes to ML-power services. 

Related IBRS Advisory

  1. Six Critical Success Factors for Machine Learning Projects
  2. Options for Machine Learning-as-a-Service: The Big Four AIs Battle it Out
  3. How can AI reimagine your business processes?
  4. Low-Code Platform Feature Checklist
  5. VENDORiQ: BMC Adds AI to IT Operations
  6. Artificial intelligence Part 3: Preparing IT organisations for artificial intelligence deployment

The Latest

May 2021: Talend, a vendor of data and analytics tools, released its Data Health Survey Report that claims 36% of executives skip data when making decisions, and instead go “with their gut”. At the same time, the report claims that 64% of executives “work with data everyday”. On the surface, these two figures seem at odds. However, the report goes on to claim 78% of executives “have challenges in making data drive decisions”, and this is largely due to data quality issues. However, the most interesting finding from the report is “those who produce and those who analyse data live in alternative data realities”.

Why it’s Important

At its core, this report highlights the issue of data literacy. The report was compiled from 529 responses from companies with over USD10 million in sales. A quarter of respondents were from the Asia Pacific region. However, IBRS cautions drawing Australia-specific inference, given that different markets have differing levels of data literacy maturity. No details were given for industry, which is also likely to impact data literacy maturity. In fairness, any more detailed analysis of a country or industry would not be feasible, given the sample size. 

The above concerns aside, the report does highlight the importance of data literacy: investments in big data tools are useless unless executives are knowledgeable and well versed in the key concepts of applying analytical thinking to business decisions. IBRS notes that without data literacy, the most common use of new self-service visualisation tools such as Power BI, Looker, Domo, Tableau, Qlik, Zoho and others, is to ‘prove’ executives' gut feelings. In short, too often visualisations tools are used to reinforce the ‘current ways of thinking’ rather than seek areas for improvement.  

The report’s statement that “those who produce and those who analyse data live in alternative data realities”, frequently underpins IBRS inquiries into why business intelligence and analysis programs fail to produce the expected business benefits.

Who’s impacted

  • Business intelligence/analytics teams
  • Senior line-of-business executives
  • Human resources/training teams

What’s Next?

ICT teams responsible for providing business intelligence and analytics services need to cease solely focusing on the tools and technologies and ‘getting data curated’, and spend time exploring which business decisions would most benefit from the application of analytical thinking. However, the ICT teams cannot do this alone. They need to be involved in uplifting data literacy among line-of-business executives and work closely with them to identify the decisions that not only can be addressed with data, but those that would make the biggest difference to organisational outcomes. This does not mean that all aspects of a data scientists role need to be explained to business executives. Rather, training executives in the principles of using data to inquire into issues or disprove current ways of doing things is more important.  

Related IBRS Advisory

  1. Staff need data literacy – Here’s how to help them get it
  2. When Does Power BI Deliver Power to the People?
  3. The critical link between data literacy and customer experience

IBRS interviews Dr Kevin McIsaac, a data scientist who frequently works with board-level executives to identify and prototype powerful data-driven decision support solutions.

Dr McIsaac discusses why so may 'big data' efforts fail, the role ICT plays (or rather, should not play) and the business-first data mindset.


The growing maturity of data handling and analytics is driving interest in data catalogues. Over the past two years, most of the major vendors in the data analytics field have either introduced or are rapidly evolving their products to include data cataloguing.

Data catalogues help data users identify and manage their data for processing and analytics. Leading data cataloguing tools leverage machine learning (ML) and other search techniques to expose and link data sets in a manner that improves access and consumability.

However, a data catalogue is only beneficial when the organisation already has a sufficient level of maturity in how it manages data and analytics. Data literacy (the skills and core concepts that support data analytics) must also be established in the organisation’s user base to leverage full benefits from the proposed data catalogue.

Organisations considering data catalogues must have a clear picture of how to use this new architecture, and be realistic in how ready they are to leverage the technology. Furthermore, different organisations have unique and dynamic data attributes, so there is no one-type-fits-all data catalogue in the marketplace.


The deployment of machine learning (ML) solutions across a broad range of industries is rising rapidly. While most organisations will benefit from the adoption of ML solutions, ML’s capabilities come at a cost and many projects risk failure. Deployment of ML solutions needs to be carefully planned to ensure success, to minimise cost and time, but also to deliver tangible results and assist decision-making.

The Latest

29 April 2021: Cloud-based analytics platform vendor Snowflake has received ‘PROTECTED’ status under IRAP (Australian Information Security Registered Assessors Program).  

Why it’s Important

As IBRS has previously reported, Cloud-based analytics has reached a point in cost of operation and sophistication that it should be considered the de facto choice for future investments in reporting and analytics. However, IBRS does call out that there are sensitive data sets that need to be governed and secured to a higher standard. Often, such data sets are the reasons why organisations decide to keep their analytics on-premises, even if the cost analysis does not stack up against IaaS or SaaS solutions.

The irony here is that IT professionals now accept that even without PROTECTED status, Cloud infrastructure provides a higher security benchmark than most organisations on-premises environments.

However, security must not be overlooked in the analytics space. Data lakes and data warehouses are incredibly valuable targets, especially as they can hold private information that is then contextualised with other data sets.

By demonstrating IRAP certification, Snowflake effectively opens the door to working with Australian Government agencies. But it also signals that hyper-scale Cloud-based analytics platforms can not only offer a bigger bang for your buck, but greatly improve an organisation's security stance.

Who’s impacted

  • CDO
  • Data architecture teams
  • Business intelligence/analytics teams
  • CISO
  • Public sector tech strategists

What’s Next?

Review the security certifications and stance of any Cloud-based analytics tools in use, including those embedded with core business systems, and those that have crept into the organisations via shadow IT (we are looking at you, Microsoft PowerBI!). Match these against compliance requirements for the datasets being used and determine if remediation is required.

When planning for an upgraded analytics platform, put security certification front and centre, but also recognise that like any Cloud storage, the most likely security breach will occur from poor configuration or excess permissions.

Related IBRS Advisory

  1. Key lessons from the executive roundtable on data, analytics and business value
  2. VENDORiQ: AWS Accelerates Cloud Analytics with Custom Hardware
  3. IBRSiQ: AIS and Power BI Initiatives
  4. VENDORiQ: Snowflakes New Services Flip The Analytics Model

The Latest

7 May 2021: Analytics vendor Qlik has released its mobile client Qlik Sense Mobile for SaaS. During the announcement, Qlik outlined how the new client enables both online and offline analytics and alerting. The goal is to bring data-driven decision-making to an ‘anywhere, anytime, any device’ model. 

Why it’s Important

While IBRS accepts that mobile decision support solutions will be of huge value to organisations, this needs to be tempered with an understanding that not all decisions should be made in all contexts. There is a very real danger that in the hype surrounding analytics, people will start making decisions in less than ideal contexts. Putting decision support algorithms (i.e. agents), KPI dashboards and simply modelling tools on mobile devices will likely be the next wave of analytics. In short, mobile big data/AI driven solutions that support specific, narrow mobile work tasks will be a very big deal in the near future.

However, creating and diving into data - that is, data exploration - is or should be, a process rooted in deep, careful, considered scientific thinking. That is a cognitive task that is not well suited to a mobile device experience. This is not just due to the form factor, but also the working context. Such deep thinking requires focus that a mobile work context does not provide.

As organisations embrace self-service analytics and more staff are engaged in creating and consuming visualisations and reports, data maturity will become an increasingly important consideration. However, data literacy is not just a set of skills to learn: it requires a change in culture and demands staff become familiar with rigorous models of thinking. It also requires honest reflection, both of the organisation’s activities and individually. 

While mobile analytics will be a growing area of interest, it will fail without a well-structured program to grow data literacy within the organisation and without granting staff the time and appropriate work spaces to reflect, explore and challenge their assumptions using data.

Who’s impacted

  • CDO
  • HR directors
  • Business intelligence groups

What’s Next?

Organisations should honestly assess staff data literacy maturity at a departmental and whole or organisation level. Armed with this information, a program to grow data literacy maturity can be developed. The deployment of data analytics tools, and indeed data sets, should coincide with the evolution of data literacy within the organisation. 

Related IBRS Advisory

  1. Staff need data literacy – Here’s how to help them get it
  2. When Does Power BI Deliver Power to the People?
  3. The critical link between data literacy and customer experience

IBRSiQ is a database of Client inquiries and is designed to get you talking to our advisors about these topics in the context of your organisation in order to provide tailored advice for your needs.

IBRSiQ is a database of Client inquiries and is designed to get you talking to our advisors about these topics in the context of your organisation in order to provide tailored advice for your needs.

The Latest

16 April 2021: BMC has released a new edition of its Helix Platform, which leverages machine learning algorithms to support AI-driven IT operations (AIOps) and AI-driven service management (AISM) capabilities. The introduction of these algorithmic features enable IT service and operations teams to predict and resolve issues more effectively.

Why it’s Important

The use of algorithms to both categorise and predict events in IT operations is a growing trend. Such AI capabilities will be increasingly embedded in existing IT operations suites. As vendors enter a new ‘AI-powered’ competitive phase, these new AI capabilities will be included as part of regular upgrades and maintenance, rather than as add-on components.

Getting value from the new AI capabilities requires planning very human responses.  

For example, the predictive capabilities of algorithms, especially when using multi-organisational data, can provide op teams with alerts well in advance of problems becoming apparent. But unless op teams are resourced and given budget to respond to such ‘predictive maintenance’ issues, these predictive capabilities will be relegated to little more than an alarm clock with a snooze button. 

Likewise, the ability to correctly leverage and continually train advisory from resolution support algorithms, will demand both training of, and input from, the support team. The algorithms are only as good as the information and the contexts they can draw on. Support team people play an intimate role in ensuring the right information is selected for training the algorithm and, most importantly, the right contexts. This is especially pertinent as virtual agents (chatbots) are introduced for self-help capabilities.

Who’s impacted

  • CIO
  • IT operations staff
  • Support desk

What’s Next?

Begin to track the new AI capabilities available in IT operations support platforms, not just for the platforms used by your organisation, but in the competitive landscape. While there is no critical priority to adopt AI-powered IT operations or service management capabilities (just yet), it is important to understand what is coming and what may already be available as part of your current licensing agreements.

Assemble a working group to explore how AI capabilities could positively impact IT operations and service management, and the changes in process and roles that would be required to leverage them.

In short, start planning for AI-powered operations and a service management future.

Related IBRS Advisory

  1. Running IT-as-a-Service Part 55: IBRS Infrastructure Maturity Model
  2. Sustaining efficiency gains demands architecture risks mitigation Part 2
  3. Artificial intelligence Part 3: Preparing IT organisations for artificial intelligence deployment
  4. IBRSiQ: Approach to identifying an ITSM SaaS Provider


The decision to integrate machine learning (ML) into systems and operations is not one that is made lightly. Aside from the costs of acquiring the technology tools, there are added considerations such as staff training and the expertise required to improve ML operations (MLOps) capabilities.

An understanding of the ML cycle before deployment is key. Once requirements and vision are defined, the appropriate tools are acquired. ML specialists will then analyse and perform feature engineering, model design, training, and testing and deployment. This is also known as the dev loop. At the implementation stage, the ML model is deployed and the application is subsequently refined and enhanced. The next stage is the monitoring and improving stage where the organisation refines the model and evaluates the ROI for its data science efforts. This stage triggers the retraining of the model through data drift and monitoring.

The Latest

23 February 2021: The appetite for crowdfunding of tech startups looks to remain strong, with the fledgling accounting software vendor Thrive securing AU$3 million through the Birchal service.  

Why it’s Important

There are two lessons to take from this announcement. 

First, commercial crowdfunding is a growth area that will favour niche tech start-ups. As more success stories emerge, this has the potential to re-invigorate the Australian startup community, which has been lagging. 

Second, it highlights the likely capabilities to be introduced in SaaS-based financial solutions: namely AI-powered automation and machine-learning decision support.

Who’s impacted

  • CIO
  • CFOs
  • Individual investors

What’s Next?

There is the potential for larger organisations to set aside funds to invest in startups. CIOs and CFOs may wish to watch the crowdfunding space that may provide relevant solutions to their needs, or secure services that may complement or even compete with their organisation. While IBRS acknowledges this strategy will not be suitable for the majority of organisations it works with, there is the possibility this will become more common over the next decade, especially for startups in security, Cloud management and cost control, AI-powered automation and machine learning-based decision support systems.

While Thrive is unlikely to be of interest to CIOs, being targeting squarely at SMEs and sole traders, the vendor’s goals leverage AI to automate much of the account process and provide recommendations, highlighting where development dollars will be going for many SaaS-based accounting solutions.

Related IBRS Advisory

  1. CIOs seek ready-made over DIY AI solutions
  2. How can AI reimagine your business processes?
  3. Salesforce Einstein automate
  4. The evolution of SaaS offerings for legacy systems

The Latest

2 February 2021: Google has announced general availability of Dialogflow CX, it’s virtual agent (chatbot) technology for call centres.  The service is a platform to create and deploy virtual agents for public-facing customer services. Google has embraced low-code concepts to allow for rapid development of such virtual agents with a visual builder. The platform also allows for switching between conversational ‘contexts’, which allows for greater flexibility in how the agents can converse with people that have multiple, simultaneous customer service issues.

Why it’s Important

While virtual agents are relatively easy to develop over time, two key challenges have remained: 

  1. the ability to allow non-technical, customer service specialists to be directly involved in the creation and continual evolution of the virtual agents
  2. the capability of virtual agents to correctly react to humans’ non-linier conversational patterns.

Google’s Dialogflow CX has adopted aspects of low-code development to address the first challenge. The platform offers a visual builder and the way conversations are developed (contexts) can be described as ‘program by example’. While there are third-party virtual agent platforms that further simplify the development of agent workflows (many of which build on top of Dialogflow), the Google approach is proving sufficient for non-technical specialists to get heavily involved in the development and fine-tuning of virtual agents

Who’s impacted

  • CIO
  • Development team leads
  • Business analysts

What’s Next?

If not already in place, organisations should establish a group of technical and non-technical staff to explore where and how virtual agents can be used. Do not attempt a big bang approach: keep expectations small, be experimental and iterative. Leverage low-code ‘chatbot builder’ tools to simplify the creation of virtual agent workflows, while leveraging available hyperscale cloud platforms for the back end of the agents. 

Related IBRS Advisory

  1. Chatbots Part 1: Start creating capabilities with a super-low-cost experiment
  2. Virtual Service Desk Agent Critical Success Factors
  3. SNAPSHOT: The Chatbot Mantra: Experimental, experiential and iterative
  4. New generation IT service management tools Part 1
  5. Artificial intelligence Part 3: Preparing IT organisations for artificial intelligence deployment
  6. VENDORiQ: Tribal Sage chatbot


For enterprises and small to medium businesses (SMBs), Artificial Intelligence (AI) opportunities are widespread and industry-specific. Each industry will grapple with conversations to understand how AI can:

  1. Create competitive advantage.
  2. Complement existing business.
  3. Disrupt, or even destroy the business model that exists today.

What businesses need to plan for is that AI engineering and AI ops are destined to be the essential umbrella to govern AI in the coming decade. Hyper-automation (HA) of business processes will see some business models fail whilst others thrive into the 2030s.

The Latest

5 December 2020: Australian education solution vendor Tribal, has upgraded its digital learning design chatbot. The move is illustrative of how chatbots can be leveraged to aid complex tasks - in this case, learning content, delivery, and leaner coaching.

Why it’s Important

Chatbots are not unique to Tribal. However, Tribal is demonstrating how such agents can deliver new capabilities into the LMS market, which can be glacial in the adoption of innovation. The Tribal chatbot is aimed at improving knowledge transfer inside an organisation. It assists domain experts to build learning content and share knowledge by recommending approaches to online training.

Who’s Impacted

  • CIO / CTO
  • Service delivery teams 

What’s Next?

Like most forms of AI, chatbots will make their way into organisations through their addition to existing software solutions, either via paid upgrades or as part of the ongoing improvements of SaaS solutions. Chatbots will increasingly act in an advisory manner or as a guide for complex processes inherent in the vendors’ solutions. 

As a result of this trend, staff will be presented with a growing number of chatbots embedded in different vendor’s solutions, each serving a specific purpose. This itself will present a new challenge for digital maturity and staff satisfaction.

Related IBRS Advisory

The Latest

2 December 2020: Salesforce Einstein is being extended into the Mulesoft automation and data integration platform. The newly announced Flow Orchestrator enabled non-technical staff to transform complex processes into industry-relevant events. The new AI-assisted MuleSoft Composer for Salesforce will allow an organisation to integrate data from multiple systems, including third-party solutions.

Why it’s Important

AI enables business process automation as a key technology enabler that favours organisations with a Cloud-first architecture. Salesforce will leverage its experience and connections with selling to organisation’s non-IT executives to secure a strong ‘brand leadership’ position in this space.

Who’s Impacted

  • CIOs
  • CTOs
  • CRM Leaders

What’s Next?

In mid-2019, IBRS noted a significant upswing in interest in Mulesoft and integration technologies more broadly from the non-ICT board-level executives. In particular, COOs and CFOs expressed strong interest in, and awareness of, process automation through APIs.  

Digging deeper, IBRS finds that Salesforce account teams, who are well-known for bypassing the CIO and targeting senior executive stakeholders, are also bringing Mulesoft into the business conversation. Also, Microsoft is expected to double-down on AI-enabled business process automation with the PowerPlatform. 

As a result, the addition of Salesforce Einstein AI into the discussion of automation and integration is expected to land very well with COOs and CFOs. 

CIOs need to be ready to have sophisticated discussions with these two roles regarding the potential for AI in process automation. Expectations will be high. Understanding the possible challenges of implementing such a system takes careful consideration. CIOs should be ready to build a business case for AI-enabled business process automation.

Related IBRS Advisory

Conclusion: Machine learning operations (MLOps) adapts principles, practices and measures from developer operations (DevOps), but significantly transforms some aspects to address the different skill sets and quality control challenges and deployment nuances of machine learning (ML) and data engineering.

Implementing MLOps has several benefits, from easing collaboration among project team members to reducing bias in the resulting artificial intelligence (AI) models.

IBRS was asked to present on the AI market for 2018 - 2019. This advisory presents an AI market overview for this time with an outlook towards 2025. How has your organisation's AI journey progressed?

Conclusion: The recent use of artificial intelligence (AI) solutions has demonstrated the value of this type of technology to consumers and organisations. It resulted in the recent discovery of new antibiotics, the emergence of self-services (e. g. virtual agents) and the ability to analyse unstructured data to create business value. However, releasing AI solutions without integrating them into the current IT production environment, the corporate network and Cloud will limit the value realisation of artificial intelligence deployments.

NewsIBRS advisor Dr Joseph Sweeney has been tracking the three major Cloud vendors capabilities in AI and said Google is right to believe it has an edge over AWS and Microsoft when it comes to corpus (the data that 'feeds' certain AI applications) and also in AI application infrastructure cost and performance. However, he said this advantage was not materialising into significant gains in the Australian market.

Full Story.

Conclusion: As Australia’s use of consultancy services continues to grow, so too does the need for businesses to obtain value from these engagements quickly and effectively. Key to obtaining this value is the organisation’s ability to easily and rapidly provide consultants and contractors with the specific context of your business, your customers and your unique challenges.

By providing the organisational context quickly, you can mitigate time, scope and budget creep, improve the quality of outputs developed by consultants and ensure that consequent plans are actionable and genuinely valuable for your business.

However, the ability to provide the needed organisational context quickly and effectively to consultants remains a common organisational challenge, and therefore a pitfall for successful vendor engagement. This paper covers how you can overcome this pitfall.

Conclusion: Despite market hype around the role of data scientists and in-house developers for the successful exploitation of artificial intelligence (AI), organisations are increasingly looking to their vendor partners to provide ready-made solutions. Both business and technology leaders are expecting solutions to be based on the vendor’s ability to leverage their customer base across various industries to create AI features such as machine learning models.

Vendors are responding by increasingly incorporating these features into their offerings, along with a new breed of vendors that are producing pre-trained or baseline machine learning models for common use cases for specific industries.

However, organisations must be prepared to contribute to this AI product development or continuous improvement process which in practical terms means giving major vendors access to data. Without access to good data the result will be sub-optimal for both parties.

Conclusion: While the current artificial intelligence (AI) initiatives are data-driven, there are instances whereby the current data is insufficient to predict the future. For example, answering the following questions might be challenging if the available data is only of a historical nature irrelevant for forecasting purposes:

  • Q1: What will be the effect on sales if the price is increased by 10 % as of the next quarter?
  • Q2: What would have happened to sales had we increased the price by 10 % six months ago?

The purpose of this note is to provide a framework that can be used to derive sales principles to answer the above questions. The same approach can be used to derive other business processes principles such as procurement, customer service and client complaints tracking.

Related Articles:

"Analytics artificial intelligence maturity model" IBRS, 2018-12-03 09:44:43

Conclusion: Increasingly, leaders in the field of AI adoption are calling out the limitations of the current machine learning techniques as they relate to knowledge representation and predictive analysis.

Organisations seeking to adopt machine learning as part of their AI-enabled transformation programs should ensure they fully understand these limitations to avoid unproductive investments driven by hype rather than reality by expanding their definitions of machine learning to include the use of graph networks and social physics solutions.


Conclusion: Artificial intelligence technologies are available in various places such as robotic process automation (RPA), virtual agents and analytics. The purpose of this paper is to provide an AI maturity model in the analytics space. The proposed maturity model can be applied to any type of industry. It provides a roadmap to help improve business performance in the following areas:

  • Running the business (RTB): Provide executives with sufficient information to make informed decisions about running the business and staying competitive.
  • Growing the business (GTB): Provides information about growing the business in various geographies without changing the current services and products.
  • Transforming the business (TTB): Provides information to develop and release new products and services ahead of competitors.

Conclusion: Organisations seeking to ride the new wave of AI-enabled transformation are facing a clear choice when it comes to the adoption of supporting AI capabilities such as machine learning or speech recognition, either:

  1. DIY (Do It Yourself) – By adopting AI early as stand-alone services; or
  2. MODIFY (Make Others Do It For You) – By waiting for AI functionality to be embedded in existing solutions.

Deciding which path to take requires that organisations reflect on their current maturity when it comes to building solutions. Only those organisations that can honestly demonstrate full development lifecycle capabilities and that have contemporary development tools and frameworks should expect anything but proof of concept success with DIY approaches to AI solutions.

Conclusion: In seeking to achieve their vision, goals and objectives, organisations constantly evaluate internal and external factors in order to take action. Although tuned to the unique needs of each enterprise, there have been identifiable waves of factors and responding actions that have occurred since 2000 in the form of business and digital transformation.

Business transformation addressed the changing nature of markets in a connected and globalised world by focusing on delivering cost savings through new models of operation, while the subsequent wave of digital transformation sought to employ technology and exploit pervasive connectivity to increase the efficiency of internal processes and customer-facing interactions.

IBRS has identified a new wave we call “artificial intelligence-enabled (AI-enabled) transformation”, which is focused on optimising business operations through the use of emerging technologies that leverage “self-learning” algorithms to make predictions, respond to real-world objects and events, and possess user interfaces that mimic how humans communicate.

However, in order to successfully exploit this new wave of transformation, organisations must first understand what exactly AI is and how AI-enabled transformation differs from the waves that have come before it.

Conclusion: Due to years of tactical software deployments in response to urgent digital transformation uplifts, organisations have created a jungle of business intelligence (BI) technologies deployed in the absence of a well described and comprehensive approach to the challenges faced; challenges that will continue to increase with the shift to AI-enabled transformation.

Instead the majority of solution paradigms have centred around the application of emerging technologies with little articulation of a coherent architecture traceable to the underlying functional or non-functional requirements required to support a well governed and long lived data analytics platform. Instead, with each new trend in reporting and analytics, e. g. big data, results in a litany of partial solutions.

Enter Data Vault 2.0 (DV2.0) is the first well described architecture, methodology and modelling approach to emerge from the BI community in the last 5 years. DV2.0 provides a solid basis for organisations wishing to avoid the data sins of the past and adoption should be a top consideration for the inevitable expansion of BI that flows from business application transformation and as part of a clear DataOps strategy.

Conclusion: The development of AI-based solutions is heavily dependent on various types of data input in the form of either:

  • Large data sets used to conduct experiments to develop models and algorithms for predictive analytics, optimisation and decision recommendations; or
  • Enriched and tagged corpuses of images, audio, video and unstructured text used to train neural networks using deep learning techniques.

While at first the data management needs of AI-based solution development might leverage both data scientists and their existing business intelligence platforms to exploit these types of data, the actual lifecycle management needs of AI developers will expand quickly beyond the boundary of the traditional enterprise data warehouse.

Therefore, like the source code and configuration data underpinning transactional business applications, the raw data and algorithms of AI solutions must be managed by evolving DevOps practices towards a comprehensive “DataOps” model.

Conclusion: Although online digital platforms are in ready supply, organisations remain unable to avoid the receipt of critical information in the form of paper documents or scanned images. Whether from government, suppliers or clients, organisations are faced with written correspondence, typed material, completed forms or signed documents that must be consumed. For a variety of reasons, it may be unreasonable or impractical to expect this information to be sent in machine-readable form.

However, machine-readable content from incoming information, both past and future, is emerging as a prerequisite to exploit artificial intelligence and machine learning as part of digital transformation. Therefore, organisations need to re-examine their data ingestion strategies and move proactively to the use of optical character recognition on incoming paper- and scanned image-based information.

Conclusion: Business leaders should convert recent global interest in AI applications, safety and effectiveness into AI governance guidelines in the exercise of their triple bottom line responsibilities (for profit, social responsibility and sustainability) as outlined in IBRS research note, “The emerging need for IT governance in artificial intelligence”1.

AI includes a very broad range of technologies being applied in virtually all industries. This means that the use of AI in both IT and operational technologies2 (OT) requires C-level attention and supervision.

Conclusion: Organisations continue to emphasise their competitive differentiation based on the data they hold, and the insights gained from analysing this valuable resource. The rate at which organisations are shifting from traditional process-based to insight-oriented differentiation is being further accelerated by the adoption of Cloud-based data analytics services.

The combined result is an increasing portion of enterprise project activity that can be classified as extract, transform and load (ETL).

Despite ETL being the mainstay of data integration for decades, the cost of specialised skills and significant manual effort expended on integrating disparate data sources is now coming into sharp focus. In response, organisations are rightly seeking lower-cost solutions for data integration.

Although ETL exists in the form of at least one tool in almost every enterprise, the cost of ETL as a proportion of data analytics projects means organisations must decrease reliance on traditional ETL tools in favour of automated solutions that exploit machine learning techniques to reduce the need for ETL developers.

Conclusion: The release of Amazon’s Echo in 2014 heralded the first of a series of “ambient” technologies1. These new devices are unobtrusive, multiple purpose and capable of responding to conversational input through integration with virtual digital assistants (VDAs) such as Amazon’s Alexa.

A key enabler of these platforms is the ability to implement “skills” or expand the platform’s capability to interpret and respond with appropriate conversational content beyond the basic function of the device itself.

The consistency of information required by organisations under omni-channel delivery models, combined with under-resourced editorial teams, mean organisations must prepare for conversational channels by transforming existing content sooner rather than later.

Failure to do so will see history repeat itself through short-term replication of content to support new channels only to have that content and channel functionality merged back into increasingly sophisticated content management platforms at significant cost.

Conclusion: Abbreviated trialling of RPA platforms is shaping up as a relatively low risk, low cost approach to exploring the use of robotics to aid business process rather than lengthy technical evaluations.

However, business process re-engineering experience shows that just automating existing business processes without addressing inherent inefficiencies and adding a robotic overlay is a total waste of resources.

Basic RPA applications do not need IT coding and can reduce repetitive tasks and improve accuracy.

In more complex situations, use of RPA platforms and tools relies on leveraging IT systems integration in providing robotic aid to assist human intuitive decision-making.

Conclusion: Abbreviated trialling of RPA platforms is shaping up as a relatively low risk, low cost approach to exploring the use of robotics to aid business process rather than lengthy technical evaluations.

However, business process re-engineering experience shows that just automating existing business processes without addressing inherent inefficiencies and adding a robotic overlay is a total waste of resources.

Basic RPA applications do not need IT coding and can reduce repetitive tasks and improve accuracy.

In more complex situations, use of RPA platforms and tools relies on leveraging IT systems integration in providing robotic aid to human intuitive decision-making.

Conclusion: As the nature of work is becoming less routine and linear, the most effective collaboration solutions are supporting the ways that teams and individuals want to work.

At the same time, customer service techniques are changing to appeal to individuals in the ways that they like to be treated.

Developments in business work flow and customer service are emerging in four broad generations of deployment:

  • Business process, work flow and customer service have morphed from document and transaction-centricity to
  • augmentation by social networking and mobility applications, followed by
  • increasing support from a conversational (Chat) model aided by interactive robotic speech, and
  • in future, even more personalised and intimate experiences delivered by Artificial Intelligence (AI) and Virtual Digital Assistants (VDA).

Many IT organisations are trying to change their perceived image from high-cost / low quality to value-added service providers. However, many of the adopted approaches revolve around improving just few processes (e.g. problem management). While these processes are important, they are insufficient to produce the desired effect for IT groups to deliver value-added services. 

In this IBRS Master Advisory Presentation (MAP), IBRS outlines the high-level issues, surrounding Running IT as a Service from both business and technology viewpoints.This MAP is designed to guide and stimulate discussions between business and technology groups and point the way for more detailed activity. It also provides links to further reading to support these follow-up activities.

The MAP is provided as a set of presentation slides,  and as a script and executive briefing document.

Many organisations are seeing growing demandand discussion around mobility and mobile ap-plications, in particular in the Networks Group.In theory, mobility can enable significant businessinnovation and optimisation of business process-es. However, few organisations have been able toclarify the benefits of mobility in terms that arealigned to their organisational goals and visionsstatements. This challenge is exacerbated by therapid innovation and changes underway in themobility market.

What is needed to address these problems is aconsistent, repeatable process that embeds mo-bility into the organisation’s overall IT Strategy.At the same time, mobility needs to be treatedslightly differently to many traditional projectsof work, as most mobility initiatives are smaller,with shorter deliver times, than large system de-ployments, but of often intimately interconnectedwith, and enabled by, the traditional larger backend systems.

To meet this challenge, IBRS developed its Mobil-ity Strategy Methodology, which provides a formalframework and process.

Conclusion: Pattern-based and repeatable processes, such as gathering operational data, validating data, and assessing data quality, offer potential for automation. The Web and software-as-a-service technologies offer powerful tools that facilitate automation beyond the simple mechanical pumping of data from one system to the next. Operational management tasks that focus on administration and control can and should be automated, so that managers have time to think about the organisation as a system, and can focus on continuous improvement.

Conclusion: Manually re-implementing application functionality in a new technology every few years is highly uneconomical. Model driven automation offers the potential to eliminate software design degradation and to minimise the cost of technology churn. Yet the model driven approach only works if conceptual models of questionable quality are discarded, and if deep knowledge about the business is used to develop elegant, compact, and tailored specification languages for domain knowledge.

This article is the final in a series of three on technologies and techniques that are leading to fundamental changes in the architectures used to construct software applications and software intensive devices.

Conclusion: Collaboration is not something you can buy. It is not a product. It is not even a solution. It is an approach to doing business. As such, collaboration initiatives must be viewed more as a transformative business project with IT support. Large-scale, monolithic collaborative initiatives run exclusively by IT will prove difficult to justify over time and likely turn out to be white-elephants. Instead, collaboration should be driven first and foremost by a change in company culture fully backed by management, with IT supplying a supportive network and software service architecture.

Conclusion: When selecting Software as a Service (SaaS) solutions, IT managers should demand evidenced from SaaS providers as the levels of service that can be expected using a formal framework. Including IBRS’s SaaSability questionnaire in requests for information will help to ensure that all parties understand their roles and responsibilities.

Three months ago two Google researchers unveiled a project which has wide implications but attracted little attention. They proposed using ambient-audio identification technology to capture TV sound with a laptop PC to identify the TV programme that is the source of the sound and to use that information to produce personalised Internet content to the PC. This technological turnkey is called Mass Personalisation by the researchers because it brings TV and the Web together, harnessing large audiences but which are informed over the Web as individuals.


As-a-Service machine learning (ML) is increasingly affordable, easily accessible and with the introduction of self-learning capabilities that automatically build and test multiple models, able to be leveraged by non-specialists.

As more data moves into Cloud-based storage – either as part of migrating core systems to the Cloud or the use of Cloud data lakes/data warehouses – the use of ML as-a-Service (MLaaS) will grow sharply.

This paper summarises options from four leading Cloud MLaaS providers: IBM, Microsoft, Google and Amazon.