How will AI impact digital experience platforms and the online experience?

We’ve all had bad sales experiences … dealing with a slick seller trying to make his quota. The classic used car salesman comes to mind pretty quickly. They are impatient, motivated only by money, lacking awareness of our needs and values, they are selling a product they know little about, and do not have the knowledge or motivation to steer us into making a decision that will best suit our needs. Depending on the marketing support offered to the seller, they might or might not have support, visual artifacts or samples for me to consume to help make my decision.

Are web experiences today better than bad sellers?

Yes absolutely! I am happy to navigate to the information myself rather than wasting time in a conversation with a bad salesman. But, if the information isn’t ready and accessible, I can only help myself so far before I need to interact with someone. Then I have the added task of reliving my online experience to advance the conversation (again, we’re getting better at this, but still – this is the norm).

Also – the web is much more scalable. Having those conversations take a human, and we’re not terribly good at running thousands of conversations simultaneously.

Within my teams, we often discuss the impacts of AI on our day to day work, but we also try to speculate on where AI will take our industry. I think (and hope) the impact will be nothing short of complete and utter destruction of what we know and experience today.

Revolutionary change happens through convergence

AI will converge into a number of industries, and digital marketing is no exception. But, it’s not just the emergence of AI that has driven this change.

DXP

Digital experience platforms have proven that a personalized experience drives higher rates of conversion and overall consumer satisfaction. Getting someone to do what you want them to when they visit your website means tailoring the experience to the needs and values of the individual. Of course, these tools are only as good as the humans running the experiment, so knowing what data to gather and what attributes define their audience is a decision that must be made. And those who are best at it create an operational “infinite loop” to continually test, measure and execute new theories of how to drive performance. So, the emergence of amazing products and techniques for gathering analytics and compiling and leveraging user data have made huge advancements in the past few years.

AI

We’ve all seen the demos. Let’s just move on.

How will it happen?

So the question is, how will the integration of AI impact digital experience platforms and the overall digital marketplace? I think this will happen in 4 phases that have already begun.

Step 1 – Co-pilot

AI is already applied to a number of use cases that can be leveraged by both the technical and marketing teams. Content generation is here today, and with some small tweaks, adding in the “brand” of an organization into the algorithms will ensure that content created by AI in images and copy will meet and exceed the brand guidelines of the organization. Leaving Adobe Summit, I was not surprised to see offerings from Adobe focusing on the integration of Conversational AI into the product suite. Image generation and copy creation by AI is becoming commonplace – and to see Adobe bring products like FireFly to the market makes perfect sense. 

Developers are not left out either – code analysis by AI helps improve and simplify the applications that are being created by engineering teams, and have immediately shortened the time needed to create efficient implementations of the most elaborate user experiences (by those who have embraced it).  This is happening today.

Step 2 – Observant Advisor

Next on the evolutionary ladder is the use of AI to continually study and make recommendations on either tests that need to be conducted, personalization or more significant changes that need to made to the user experience in order to increase and optimize conversions. Again, there are a number of products that tout the use of machine learning and AI to perform these tasks. These products will become increasingly common, and remarkably more capable until we trust them enough to run on their own.

Step 3 – Auto-pilot

Once we have built a site and customer attribution, personas etc. – would we dare to turn this over to an AI to run and optimize the site on a continual basis? I believe so.

We already see strategies using AI like dynamic landing pages, auto-optimization and auto-personalization being delivered by a number of platforms. These are arguably “under the management of AI”, but only in a very rudimentary, limited way.

The AI is only allowed to change pre-defined, limited elements of content and user experience. Any significant change still requires human intervention in the form of design and development at the very least – but for how long?  Yes, this too will change.  Although I think it will be hard for us to map and explain the values and effectiveness of various presentation elements to the AI.  That part will be interesting – mapping this weird world of user experience to and engagement strategy that an AI will understand.

Step 4 – Revolution

A good sales or support experience includes a meaningful, minimal conversation about your needs, wants and wishes, followed by thoughtful response suggesting a highly refined set of products or services that would best fit your needs once enough information is communicated. And, this is delivered to you from a place of trust – not one motivated by quotas or even influenced by the budget of time and attention to your needs.

This perfect experience doesn’t really exist – either in person or online – but it will soon.

Conversational AI will create a literally personal experience for each visitor. With the ability to listen and understand what you’re looking for, and with the right queues – the ability to truly assess what drives your decision to move ahead and the best path to get you where you need to be. Carousels, accordions, tabbed data – all gone and replaced with things like a real-time rendering of the product based on your specifications or a complete how-to video compiled on demand to answer your product support questions or step you through a difficult process.

What will also be interesting is what vendors will do in response to this revolution … will the traditional DXP vendors be able to make this change?  Will it be a disruptor that leads the charge?  Will they market personality modules based on data models and use cases  specific to things like commerce and customer service?  What will the marketplace for DXP morph into?

Content Integrity and Blockchain

In order for this future to exist, we have to address a horrible elephant in the room – content and information integrity.  AI is like anything else – garbage in means garbage out. In order to feed the AI proper information, we must have the ability to ensure the data is good, and when it’s not good, determine the source of the error so that it can be easily corrected and not repeated in the future. This is a much more immediate problem we have yet to solve within social media, but this should be much more easily addressed within corporate content supply lines.

Blockchain will provide an end to end capability to understand the origins and subsequent mutations of content as it progresses throughout various channels – both within an organization and in the wild. It is essential this immutable record of content is established for the sake of the AI to determine the quality and any potential bias introduced in the supply chain.

An Exciting Future … I think

What will really happen? I don’t think we will wait long to find out. It will also be interesting to see what vendors have the investment and innovation to make the future come true. Platforms will introduce new and exiting “personas” of AI targeting B2B and B2C channels – becoming online personalities meant to built trust with visitors once key traits can be determined. Is it more insidious? Well, this is marketing, so my feeling is yes. But, I think it’s what will happen, and now I am on the record. Thanks for reading!