The Rise of Customer Experience Application Platforms
Until recently, Customer Experience vendors focused on packaging native capabilities as hosted microservices, and a separate middleware platform combined microservices from multiple vendors into a single workflow. In recent years that’s changed. Vendors are increasingly hosting a “foundry” of third-party REST APIs, in effect reselling partners’ complementary functionality. Vendors are orchestrating calls to both native and external APIs. In so doing, these Customer Experience vendors are morphing, in effect, from Software-as-a-Service, to a bona fide “Application Platform.”
Some examples help to illustrate this extraordinary shift:
- Genesys has transitioned from a suite of hosted Contact Center applications to “…an application platform that manages interaction workflows comprised of microservices created by Genesys and its partners.”
- Salesforce has transitioned from a suite of hosted CRM applications to “…an application platform that manages customer relationship workflows comprised of microservices created by Salesforce and its partners.”
- Qualtrics has transitioned from a suite of hosted survey applications to “…an application platform that manages customer feedback comprised of microservices created by Qualtrics and its partners.”
The list goes on, with vendors in every sub-category of the Customer Experience landscape attempting to become indispensable Application Platforms. This shift will have profound implications for enterprise customers.
Enterprise Customers Stand to Benefit
With an Application Platform that delivers broad and powerful capabilities, enterprise customers have the means to unleash hyper-personalization, dynamic experimentation, and accelerated adjustments to changing conditions. The Holy Grail for Customer Experience management is now within reach.
Financially, enterprises benefit from the fact that orchestration services are bundled in essentially free of charge. Moreover, to the extent that branded, specialized microservices can be purchased from a plethora of vendors (either directly or as part of a foundry), there is downward pressure on pricing.
Too Much of a Good Thing?
Enterprises that embrace Application Platforms will need to overcome several challenges.
Orchestration is easier said than done. A workflow can perform in the aggregate only as well as its individual, underlying API calls perform. An API call to a third party that returns corrupted data may lead to compromised results overall. Latency in an API call to a third party may lead to increased latency overall. And even if workflows are initially designed to mitigate these and other issues, they are intrinsically unreliable, because underlying APIs can themselves change at any time without notice.
Compliance is another impediment for enterprises. Workflows that leverage personal, identifiable information are subject to data privacy regulations every step of the way. Workflows that use AI must be ethical, unbiased, and explainable throughout.
Finally, though each API call is billable at a contracted price, the enterprise customer will need to accept the inherent price volatility for each individual workflow.
And what if there are multiple Applications Platforms in the environment, with some overlap of features and functionality? Challenges are compounded exponentially.
Guidelines for Customer Success
To capitalize on vendors’ transition from SaaS to Application Platforms, enterprise customers should prioritize use cases and then implement in manageable stages. Each stage of implementation should add value to the business. Allocate budget for analytics to depict intra-and inter-platform flows and quantify Key Performance Indicators. Set targets in advance, and then measure results periodically to stay on track.
Most importantly, enterprise customers should hold vendors accountable for results. Only Customer Success can fuel expansion of features and growth in usage of Customer Experience Application Platforms.
About the Author
Michael Sisselman, MBA, M.Sc. is an expert in Customer Experience platforms that delight customers, empower the workforce, and drive business outcomes. He most recently worked at Avaya, where he led a team of data scientists in Conversational AI, Intelligent Automation, and CX Analytics. During the (surreal) summer of 2020, Michael spent time developing methodologies pertaining to Business Value Consulting and Customer Success.
Michael lives in New York and plays classical guitar every day. For more information, click his LinkedIn page.
Categories: Conversational Intelligence, Intelligent Assistants, Articles
Related Articles
-
Exploring Microsoft’s New Magentic-One Framework: A Flexible Agentic System for Complex Enterprise Tasks
-
InstructLab and RAG: A New Oppportunity for Enterprise AI with IBM’s Granite Models
-
Opus Research Welcomes Ian Jacobs as VP and Lead Analyst
-
Navigating a Future of Automation: Strategic Insights from NICE’s Analyst Summit 2024
It is becoming a more complicated world this communications space. It is a high time for more folks to get out and share their knowledge like Mike has so others can learn and benefit from it. There are many pitfalls and dead-ends one can find themselves in but really talking to a skilled, experienced consultant is probably a great and easy start.
Excellent insightful article! It’s no surprise you were by far the most successful AI project leader at Avaya for customers. I can recall CX executives fixated on your presentations on AI models using credit reporting agencies to personalize the customer experience. It was a pleasure working with you. I look forward to reading more of your articles on how to commercialize AI for businesses.
Nice overview with great specifics!
Your explanation of this significant transformation helps many industries. Thank you for informing and communicating so clearly.
Bang on Mike. I definitely have seen the effect with Salesforce. I am also seeing ServiceNow as an orchestration platform attempting to become all singing, all dancing.To quote from their website: “We enable the digital workflows that unlock productivity for employees & enterprise alike. Deliver a faster, smarter & more automated experience.” Google ServiceNow and microservices. I don’t get any royalties or loyalties from them, but it is a SaaS I see a lot.
Here is a thread of comments (edited for clarity) along with responses from Michael. They appeared on the LinkedIn post announcing publication of the article here:
Bernard Gutnick [https://www.linkedin.com/in/bernardgutnick/]
Excellent insight on how AI moves from the concept to reality. It’s no surprise you were by far the most knowledgeable person at Avaya who could present AI models that completed resonated with CX executives. I still recall how your live examples of how businesses could use credit agency data combined with data warehouses results in an incredibly unique experience for a customer. It was a pleasure working with you and learning how to commercialize the possible rather than just talk about it.
Michael Sisselman
Thanks for kind words, Miss those days when we worked together on Real Time Decisioning, which leveraged AI to drive differentiated, optimized CX.
Meanwhile, my article in Opus Research was based on a series of extraordinary conversations with execs interested in Conversation AI, Intelligent Automation, and CX Analytics.
*****
Sascha Poggemann [https://www.linkedin.com/in/saschapoggemann/]
Very interesting read. What is your take on conversational AI playing a role in this?
Michael Sisselman
CX is predicated on effective self-service, and so enterprises need to deploy various chatbots, voicebots, and knowledgebots that comprise Conversational AI. Clearly orchestration and other platform services are imperative, but the question is whether these services will be provided by the bot vendors themselves, or through a stand-alone Conversational AI platform that is built for that purpose—like your recently released Cognigy.AI V4. https://www.cognigy.com/
*****
Noland Bradshaw, Ph.D. [https://www.linkedin.com/in/nolandbradshawphdmphil/]
Interesting piece! As a financial model risk manager, the idea that the workflow can perform in the aggregate only as well as its individual, underlying API calls perform, resonated with me.
Within a model risk management framework, we observe that financial model risks are mitigated only as well as the underlying components of the framework, be they risk policies and procedure, model development and model validation.
Michael Sisselman
Brilliant observation. Like IT reliability and performance, it seems like financial risk is sort of additive, and hence the “weakest-link” argument is appropriate. That said, it’s equally possible that compilation of piece-parts in some cases creates synergies, and hence value. Do you agree?
*****
Dan Wilson [https://www.linkedin.com/in/dwilson1/]
Agreed. “enterprise customers should prioritize use cases and then implement in manageable stages.”
Michael Sisselman
Yup, the “Big Bang” usually fizzles.
*****
Christine Loomis [https://www.linkedin.com/in/christine-loomis-8536671/]
Your explanation of this significant transformation helps many industries. Thank you for clarifying.
Michael Sisselman
Absolutely agree that APPLICATION PLATFORMS can be used to advance CX use cases across many verticals.
*****
Greg Pelton [https://www.linkedin.com/in/greg-pelton-631789/]
Nice article Michael. During this transformation, there are a couple things to worry about: The first is workflow. It is all fine and good that APIs can be leveraged to bring in a plethora of different services, but there needs to be easy ways to wrap this within a workflow. The complexities of programming the workflow can outweigh the benefits of the microservices, especially when each microservice has its own set of assumptions. The second is what you hinted about with respect to data. Each microservice wants access to the core data sets and may retain their own data residue over time. It is very hard to enforce a data retention policy when you can’t see inside each Lego block to inspect what it is collecting. It is very hard to conform to GDPR, CCPA, etc. when the internal data is not visible, or if the APIs do not explicitly allow redaction. AI only adds to this problem because any residual customer data that might be trapped in the models is completely opaque.
Michael Sisselman
You win the prize—a virtual coffee from Starbucks—for extremely thoughtful comments that are closest to what I am writing about in my next article for Opus Research.
*****
Tony Shrader [https://www.linkedin.com/in/tonyshrader/]
You can’t emphasize enough the correlation between micro-services and deep personalization for customers. “consume a little service there, another service here” through a workflow. Nice Job!
Michael Sisselman
Sounds like you are a big fan of contextual, dynamic, hyper-personalization. But while the upside is huge, many enterprises will likely struggle with manageability, etc. My guess is that skilled people and upgraded processes are no less important in CX than robust technology.
*****
Valerie Peck [https://www.linkedin.com/in/valerie-peck-4b143/]
Really interesting article and true. The need for some of the survey driven vendors to add value past that role is becoming more and more important to keep them competitive with each other. They should be adding in journey mapping and segment messaging/orchestration best practices and guidelines as well.
Michael Sisselman
In addition to survey responses, comprehensive VOC requires systematic inspection of interaction content and contextual metadata. And while the communications and processes need to be orchestrated in real time, capturing this as part of the historical customer journey is imperative.
*****
John Orton [https://www.linkedin.com/in/john-orton-92b820124/]
Hyper-personalization goes hand in hand with the Hyper-segmentation in this case. We are currently seeing a race to deliver real-time decisioning which is invisible, if not imperceptible to customers and delivers valuable outcomes.
Nicely done. Have your bot call my bot … wait – they already did!
Michael Sisselman
Real-time decisioning is a perfect example of a platform that reaches out to assorted repositories, applications, and scoring engines, and then executes rules to drive actions. This is the indispensable component that enables 1-1 sales and service interactions.
*****
David Urbanic [https://www.linkedin.com/in/david-urbanic-5b43/]
Important insights and info.
Michael Sisselman
The platform paradigm extends to NLU as well. Rather than a single NLU engine, we’re starting to see orchestration by a platform across multiple NLU engines, and this improves the overall comprehension and efficiency. Appreciate all of the great work that PAT is doing vis-a-vis “meaning”.
*****
Laurent Philonenko [https://www.linkedin.com/in/laurent-philonenko-3bab5/]
Very true Michael. I would add that getting the right data is the number-one success factor for a truly contextual experience. The evolution towards platforms is ineluctable but will meet expectations only when benefits are delivered with real-world use cases fed by appropriate data. I know I am preaching the choir here!
Dan Miller [https://www.linkedin.com/in/danmiller/]
I couldn’t agree more. Plus, I like that you used the word “ineluctable.”
Michael Sisselman
Didn’t we put a man on the moon? Overcome Y2K? Switch from TDM to VoIP? Agree that acquisition of quality data from disparate sources is of paramount importance, but don’t think that this is insurmountable. Meanwhile, the real question is how much enterprises are willing to pay for true Customer Success. That’s something that I’ll be writing about soon.
*****
Pinku Reen [https://www.linkedin.com/in/pinku-reen-614aa11/]
Heading is so much ‘you’ Michael, you make things interesting! Orchestration should hide the underneath complexity and that is only possible when the underling platform complements it. It shows your rich experience when you say that vendors should focus on useful use cases and implement in manageable stages. Quite insightful!!
Michael Sisselman
Big Bang never works. But to the extent adoption is slow and steady, we need much better analytics to discern meaningful change and attainment of business value.
*****
Michael Pawlak [https://www.linkedin.com/in/michaelpawlak44/]
Interesting to see how the CX industry has evolved over the years. Seeing major players like Genesys and Salesforce move to platforms clearly illustrates the importance of managing all customer touch-points. The key remains in seamless integration and orchestration, which continues to be elusive even for the major players in this space.
Michael Sisselman
Excellent point. large vendors are clearly moving beyond core competencies into complimentary functionality. It’ll be interesting to see how this impacts the smaller, more specialized vendors.
*****
Anthony Uliano [https://www.linkedin.com/in/anthonyxuliano/]
Great insights! Thanks for sharing. We built DaVinci, an iPaaS for contact center and CRM integration because we saw this trend coming. We’re excited to see where all this goes.
Michael Sisselman
Seems like vendors are coming at the enterprise market from both sides: There are iPaaS vendors that are adding CC functionality to their portfolio. And there are CC vendors—like AMC Technology—that are adding iPaaS functionality to the mix. Meanwhile, I love your recent rebranding to DaVinci!
Anthony Uliano
Thanks! We’ve seen for a few years now that customers don’t want a fixed feature solution. They want flexibility to change and grow their agent and customer experiences, without having to wait months for a feature from their vendor. We’re also seeing companies, wanting to combine capabilities like AI and voice, but not from the same vendor. An iPaaS like DaVinci supports those scenarios.
*****