Editorial
The Gist
- Self-service struggles. Many self-service tools fail to meet rising customer expectations due to gaps in leveraging unstructured data effectively.
- Knowledge management evolution. Modern knowledge management systems can bridge data silos and make sure bots provide accurate and helpful responses.
- CX that adapts. Meeting customer expectations requires continuous adaptation of technologies and strategies to deliver frictionless experiences.
There is a great divide between what customers expect and what brands deliver. It’s a divide that’s growing as customer expectations climb even higher.
What’s happening? In many cases, the actual customer experience (CX) is improving. Think about how easy it is to place an order on Amazon and have the item arrive at your home the next day. However, as prices rise, brands that cater to discretionary spending (i.e., retail, hospitality and travel) are upping their CX game. The result of this is that customers come to expect that level of service from other industries like banking, health care and even government.
Generative AI is certainly raising the stakes as well. Customers now expect a ChatGPT-style answer from every brand’s bot. When they don’t get that, the divide just grows.
As a CX leader, there’s a lot you can do to close the gap. But if you’re looking for low-hanging fruit, then you should focus on making your self-service channels as useful and frictionless as possible. To do that, you’ll have to get a handle on your data.
Structured vs. Unstructured Data: What's the Difference?
Two main types of data exist: structured data and unstructured data.
Structured data is the information that typically fits neatly into a spreadsheet or a relational database. It’s information like customer names, account balances, product details and sales data — anything specific and, well, structured. It’s easily organized, searchable and accessible, and it’s therefore quite usable.
Unstructured data, on the other hand, isn’t easily placed in a spreadsheet. It includes PDFs, emails, documentation, policies, support tickets, chats, voice recordings and free flowing text. This type of data is where you get the information that will truly help you understand your consumers better and build the systems and channels to support them with a personalized customer experience. However, because it lacks clear structure, this data has traditionally been harder to interpret and therefore harder to harness.
However, with AI, machine learning (ML), natural language processing (NLP) and generative AI, we finally have the tools to put that unstructured data to good use.
Related Article: How Artificial Intelligence Can Break Through Data Silos
Where Data Falls Short for Customer Expectations
We know that customers have long preferred to solve their issues independently, without speaking to customer service agents. In fact, a Harvard study from as far back as 2017 found that 81% of customers attempted to take care of an issue before reaching out to a live representative. More recent surveys and studies have shown that this sentiment has only grown.
Tools like chatbots, knowledge bases and FAQs can help customers with self-service, but unfortunately, they’re still falling short. A recent Gartner survey found that only 14% of customer service issues are fully resolved in self-service.
Connecting Self-Service to Unstructured Data for Better CX
Self-service channels fail because they can’t resolve most customer queries. And they can’t resolve those queries because the large language models (LLMs) they rely on aren’t trained on the wealth of unstructured data specifically available within the company.
Unfortunately, the answer isn’t as simple as pointing the LLM at every available PDF or call transcription. Harnessing that unstructured data means prioritizing and elevating a discipline that most enterprises would rather not think about: knowledge management.
Let me assure you that knowledge management today is not what it used to be. At the risk of dating myself, I’d even say, “This is not your father’s Oldsmobile knowledge management.”
The Role of Knowledge Management Today
Traditionally, within the customer experience realm, knowledge management has meant making sure contact center agents can access relevant and accurate company and customer information to better serve their clientele. As more companies enable customer self-service channels, knowledge management systems are also employed to feed those automated tools.
New knowledge management systems integrate with databases, customer relationship management platforms and anywhere else structured data is stored. The system pulls data from these areas, categorizes it and makes it easily accessible. For unstructured data, AI technologies like NLP and ML tag, classify and organize PDFs, emails and multimedia files to make unstructured data accessible. These systems then pull both sets of data together at the point of conversation.
Modern knowledge management can also help mitigate the issues of decaying, conflicting or just plain bad data. That’s important because the axiom “garbage in, garbage out,” is especially true when it comes to LLMs. Humans and bots need clean, accurate unstructured data to do their jobs. If out of date, customer data becomes more harmful than helpful, which leads to hallucinations, potential compliance violations and customer dissatisfaction.
Modern knowledge management tools can deploy something called content intelligence in which they have the ability to scan thousands of documents very quickly and identify where there are contradictions or missing descriptions. They assess the likelihood that the content can be consumed by a bot, and they can then provide an assessment of the quality of the content or conversely the risk that the content could cause a hallucination.
Related Article: 9 Principles to Improve Your Customer Data Management
Webinar
Nov
6
Fix the Content Bottleneck: Build a Better WebOps Strategy
Content stalled? Dev overloaded? You’re not the only one. Learn how streamlined WebOps bridges the publishing gap.
Register
Webinar
Nov
13
Know Your Caller Reputation: How to Protect Your Brand and Get More Calls Answered
80% of unidentified calls go unanswered. See why your calls aren’t getting through.
Register
Webinar
Nov
20
How to Build a Solid Knowledge Foundation for AI Success
See how leading brands keep their AI honest, compliant and actually helpful.
Register
Webinar
On demand
Beyond Storage: Smarter Content, Bigger Impact with DAM + AI
Discover how the DAM + AI duo makes content smarter, stronger and more accessible.
Watch Now
Webinar
On demand
Agentic AI Playbook: Real-World Customer Service Use Cases You Can Deploy Now
Boost self-service by 30% and slash call volume by 63% with agentic AI.
Watch Now
Webinar
On demand
Call Spoofing Trends Financial Institutions Can't Afford to Ignore
Don't let robocalls sabotage your customer relationships. Learn how to secure your voice channel.
Watch Now
Bots Backed by Better Knowledge Management
Customers want a frictionless, self-service experience. They want a bot to understand what they’re saying as well or better than a human could. They want fast resolution. And they expect a ChatGPT-style answer from every brand’s bot.
From a business perspective, the answers that a bot provides must be vetted, compliant and free from hallucinations. With new knowledge management systems and the human capital to power those systems, we can now leverage the structured and unstructured data needed to do both and finally narrow the customer expectation gap.
Of course, customer expectations will continue to rise and evolve, so as CX leaders, it’s up to us to keep evolving right along with them. What constitutes a great customer experience this year probably won’t cut it next year. The key is continuous evaluation of customer expectations, as well as the technologies and strategies needed to deliver their next best experience.
Learn how you can join our contributor community.
