Our first webinar on “Knowledge Minining for ServiceNow Knowledge Bases” is unfortunately over. You can find the recording here.
In January, we will start the next round and offer another webinar on the topic, where we will answer your questions afterwards. If you already have questions about the topic, you can contact our experts Anna Busch and Wim van Brussel.
Our first webinar on “Knowledge Minining for ServiceNow Knowledge Bases” is unfortunately over. You can find the recording here.
Difficulties keeping your knowledge base up-to-date?
Struggling to keep track of a vast amount of knowledge articles?
How mature and reliable really is your knowledge base?
Let’s find out! Dial-in to our webinar on 30 November, 11:00 (GMT+1)!
Last week I visited tcworld, a conference for technical communication. It took place on-site in Stuttgart for the first time since the covid outbreak, and I could attend in person.
Being at the conference showed me what I had been missing out on throughout the last two years. Not only was I able to hear some great presentations, but also network with other attendees more intensely than is possible online. I learned from industry leaders, met with experts in the field and got their insights on the latest trends.
Especially the presentation “Automatischer Ontologieaufbau – Quality in, Quality out” (Automatically building ontologies – Quality in, Quality out) by Congree Language Technologies GmbH and Mercedes-benz AG showed what is possible in terms of ontologies and knowledge graphs when you have good data quality. What an incentive to keep your data clean.
Sometimes attending from the comfort of one’s home might seem tempting, but you are missing out. It´s time to leave your comfort zone once again, so if you have the opportunity to attend a conference in person, I highly recommend it!
Knowledge and expertise are two of the most valuable assets for essentially each organization. Properly managing these assets directly improves customer experience, workflows process optimization, knowledge retention, and a culture of learning.
What is the ServiceNow Knowledge Management (KM) application?
The ServiceNow KM application is a well-functioning tool that adequately captures, stores, and manages this knowledge and experience.
The application has become widely established through its functionality and usability; some subsequent flaws are however luring around the corner:
We live in times where we are bombarded with data which is continuously changing. What worked well yesterday soon becomes antiquated. The fast pace of day-to-day business forms a challenge to keep an organizations’ knowledge bases up-to-date. With hundreds, if not thousands of knowledge articles — which form the basis of the ServiceNow self-service concept in general — an overview of all know-how is difficult to maintain.
Continuously skimming through all knowledge to detect outdated, redundant or incorrect data is a costly endeavor. One would need to go all lengths to analyze the reliability and maturity of your knowledge base. How can you address this challenge?
What is Knowledge Mining?
In the world of big data, data mining is the process of extracting valuable information and insights from large data sets by using Artificial Intelligence. Knowledge Mining as a subcategory of Data mining focuses on knowledge instead of data. It allows to automatically extract intelligent information from a variety of sources, including unstructured data such as text, images, and videos. Therefore, it uses Natural Language Processing (NLP). NLP is a method of artificial intelligence that can analyze text and understand its meaning. Using NLP, it is possible to improve the quality of Knowledge Bases in an automated way.
Why should I use Knowledge Mining for ServiceNow Knowledge Bases?
Knowledge mining can be used to identify, extract and rate relevant information from ServiceNow. It can flag articles in the KB as reliable or outdated and help with maintaining the KB. This makes it easier and more fun to work with for customer service representatives, IT professionals and other employees.
There are many benefits to using Knowledge Mining for ServiceNow Knowledge Bases. Perhaps the most obvious benefit is that it helps you to find the information you need quickly and easily, like the responsible for a specific topic. This can save you a lot of time and frustration, especially if you are working on a complex project with multiple stakeholders.
Another benefit of Knowledge Mining for ServiceNow is that it boosts the accuracy of your knowledge base. By mining data from multiple sources, information can be cross-checked to ensure it is correct. This helps to prevent misinformation from spreading and ensures that your knowledge base is as accurate and up-to-date as possible.
Finally, Knowledge Mining improves the overall quality of your knowledge base. By constantly mining data and adding new and relevant information, you can make sure that your knowledge base is comprehensive and of the highest quality. This will make it more useful for everyone using it.
How to get started with Knowledge Mining for ServiceNow Knowledge Bases – avato’s Knowledge Analytics App
ServiceNow Knowledge Bases offer a wealth of data that can be extremely valuable for ServiceNow users looking to improve their workflows and optimize their processes. So, how can you get started with Knowledge Mining for ServiceNow Knowledge Bases? Well, luckily, there is no need to reinvent the wheel – avato’s Prodigy Analytics App has you covered! The Prodigy Analytics App is a solution that enables you to analyze ServiceNow Knowledge Bases quickly and easily for insights that help you streamline your workflows and improve the maturity and reliability of your KB.
Best of all, it is simple to use. The App will automatically generate a report that includes all the insights gleaned from the data mining process. This report can be exported to Excel for further analysis or displayed as a Dashboard.
If you are looking for a quick and easy way to get started with Knowledge Mining for ServiceNow Knowledge Bases, avato’s Prodigy Analytics App is the perfect solution!
There are a lot of general tips and standards you can follow to help make your knowledge articles support your readers’ needs. Use simple language, good structures, helpful graphics, a style guide… All of these are great steps towards user-friendly articles. But there are some user needs that are hard to anticipate when you do not know your users well. What menu structure is most intuitive to them? What information do they need to look up most often? What expert vocabulary are they familiar with?
You can make assumptions on these things based on the content and your own role in the team. But there is a chance that you miss the mark. Then you run the risk of spending needless time making something unintuitive for your readers, which will cause frustration. Frustration you could easily avoid by getting in touch with your readers and learning more about their needs. Let’s take a look at how to gather insights on such needs.
Gather insights passively
The most low-effort way for you to gather insights on your users is to let them approach you with feedback. There are multiple ways you can do that, but not all will yield the results you might be looking for.
What stops users from giving you feedback?
First, just because you offer ways of gathering insights, you might still not get the input you are hoping for. This is because different hurdles are holding your users back:
Have you ever found information that seemed wrong or worth optimizing to you, but not reached out to someone that could make the adjustment? Subconsciously, the effort to go to the length of reaching out to the content creator can seem so big – and having to put yourself forward can seem so exhausting – that it holds us back from giving input. Often, the issue has to be truly relevant to us personally or have a very big impact on others for us to consider actually addressing it.
It can therefor help if you give incentive for reaching out. For example, gamification can help by awarding points and expert titles to users that participate a lot. Or public acknowledgement of valuable contributors. Think about how you can make it worth the contributors’ while and make their input feel appreciated.
On the technology side, any tiny extra step on the way to giving input might be the point where a user feels it is too much effort after all. For example, if a contact form asks for more than basic information, users can get frustrated with how many fields they have to fill in. They might even worry for the security of their data and close the form. But even a minor inconvenience like having to open another page can be too much if the user is not strongly determined to get that feedback to you.
That’s why it’s important to make it as fast and easy as possible to give the input. Include as few steps as possible and make them intuitive and straightforward.
Options to passively gather insights
There are a few popular ways to let your users approach you with feedback:
Many knowledge bases offer a central contact point and ways to directly contact the authors of knowledge articles. But during a full workday, an extra email can still be too much work, if it does not seem to offer benefits. To make it more likely for people to contact you via email, try offering a pre-filled mail (e.g., via an adjusted “mailto” link or a mail template), that users only have to insert their actual input into. Though also consider offering a contact form:
A contact form can feel like a lower hurdle than composing an email. That’s because it already gives a structure and limited input options and you don’t have to explain circumstances, add a subject and a greeting, and so on. But make sure to keep the number of mandatory fields as low as possible, both to lower effort and not make your users feel like you want to collect their data for questionable purposes.
Many applications or pages also offer page rating functions, asking users whether they found the content helpful or not. While this can help identify content that need changes, you otherwise will not learn as much from such ratings as you might hope. Even knowing a piece of content got a lot of bad reviews doesn’t necessarily help fix it, as you first have to identify what exactly makes it so bad.
Of course, you could include a short questionnaire to clear up such questions, but this comes at a time expense not many readers are willing to give. (We’ll hear more about questionnaires and how they can be more helpful when we talk about actively gathering insights.)
The comment section offers easier and faster feedback options than email and contact forms. It is less formal than an email and typically right there, underneath the content. A detriment, however, is that (at least with non-anonymous comments) you are putting your comment out there for other readers to see and potentially judge. That can contribute to the mental hurdle for this feedback option.
Out of these options, a good basic combination is to have a comment section and choosing at least one between email or contact form. That way you give options both for speed and ease, as well as for more private ways of contact.
Rating options or optional questionnaires can help gather some overarching user input but tend to give less helpful input and often also a lower return rate overall. Treat this more as a nice-to-have than a must-have.
Limited types of feedback in passive gathering
Now that we’ve considered some ways of passively gathering insights, another issue becomes apparent with this passive approach: Not only are there many things holding users back from giving input in the first place. It is also likely that the input you do get is limited to certain types, as you are depending on a strong enough motivator for a reader to give feedback. Which means something about the content causes a strong enough reaction in the reader to have that motivation. That makes it likely the cause is errors or other inconveniences in the content.
This can of course yield vital input, such as updates, corrections and structural improvement suggestions. And sometimes you might even get information about user behavior. But the latter are likely to be rare cases. That’s because you are missing out on feedback on things that don’t happen to trigger a strong reaction in the user as they are reading.
So, while you should not ignore the value of comments, emails and contact form submissions you might get, you can get a lot more insights by also using active ways to get input.
Gather insights actively
Instead of only relying on your readers approaching you with feedback, why not actively check into their needs? Here are some options to consider for getting to know your readers better:
Questionnaires somewhat fall in the area between passively and actively gathering insights, depending on how you use them. In the most basic form, a page rating is a questionnaire. On the more active end, you might approach your readers directly, asking them to fill in a questionnaire that asks more specific questions on their needs and work habits. Used with more nuance, they can be a valuable tool to gather insights.
If you get many people to contribute, you can even use this to make a representative survey of your users’ needs. But don’t be afraid to start small. Even very few submissions can give you interesting insights. Just be careful to consider that some of it might be outliers and not the needs of the majority of your readers.
Getting participants for questionnaires can be tricky, because they suffer from the same mental and technical hurdles as the methods above. If you can, consider offering an incentive for participating. Also, make sure the questionnaire doesn’t take up too much time, is easy to navigate, and gives participants a good overview on how far into the questionnaire they have progressed. That way you help limit frustration.
User interviews are a great way to get in touch with your users and gather insights, especially when you’re just getting started with looking into their needs. You can ask questions on any number of subjects you’re curious about, with the freedom of flexible follow-ups and allowing the users to bring up topics of their own. Additionally, it can get you in touch with readers you wouldn’t otherwise interact with personally. If you leave a positive impression, it’s a great promotion for your work as a knowledge manager and you’re more likely to get input again from these readers in the future.
One big concern with interviews is that they take a lot of time. But even within fifteen or just five minutes, you can easily gain valuable insights. Plus, you are more likely to get buy-in from the readers if you do not take up a lot of their time or make it feel like a big commitment.
Another concern, like with surveys, is that you need a certain number of participants to get representative results. But you don’t have to conduct a quantitative usability research. Every single participant is someone you can learn from. Again, just keep in mind that some statements might be outliers and not the norm.
Another great way to get and stay in touch with your readers is to set up or join a regular meeting with them. This doesn’t have to be a dedicated meeting for knowledge base decisions but could be any regular meeting you can join.
Ideally, go for meetings that include frequent readers of the knowledge base as well as management, perhaps even owners of process changes. That allows you to gain insights both on the daily work and needs of your readers as well as management needs and process changes that might affect the knowledge base. You will learn a lot that’ll help you improve the documentation, and you have a space to ask questions in if necessary.
Additionally, you are building relations with the other participants of the meeting, giving you reliable points of contact for any future questions. You are also building trust with them, meaning they will be far more likely to reach out to you for any changes needed in the knowledge base, lowering the hurdles to communication we saw with some feedback mechanisms.
Keep gathering insights
Once you have started some of these ways of gathering insights, keep going. The more you learn, the wider your understanding of the processes and user needs will get, revealing more and more opportunities to improve the knowledge base and make it more helpful for your readers. It might completely change the way you approach doing knowledge management in the future.
Drafts, working copies, old versions. Over time, numerous variants of a document can accumulate. For example, in marketing, the same text appears in a flyer, a brochure and a newsletter. Duplicates also occur in documentation, for example when content for very similar products is maintained separately. In addition, sometimes information is copied from one knowledge silo to another so that more users can access it.
Why is duplicate content bad?
At first glance, all these copies seem justified. After all, they serve a purpose. Over time, however, duplicate content leads to problems. If something changes, the information must be updated in several places. If a copy is overlooked or a small mistake happens, contradictions arise. If work then continues on several versions, users may struggle to determine which variant is current. Or is any version really correct? This confuses users who wonder which one applies to their case.
In other words, duplicate (or even triplicate and quadruplicate) content
- doubles the maintenance effort
- leads to errors and follow-up costs because the reliability of the information is reduced
- impairs the user experience by making users feel confused
What can I do about duplicate content?
Content management prevents problems with duplicate content. It ensures that everything that belongs together can be found easily. To do this, the information is stored centrally and provided with the necessary metadata. A monitored process with regular checks according to the 4-eyes principle is also part of this.
But what can you do if the duplicates are already there? Searching large collections of documents by hand takes a long time and achieves little. It is quicker and easier to automatically detect duplicate content. This is done by measuring the similarity of two documents. The result is a list of duplicates. Afterwards, one version can be discarded or two documents can be made into one.
How can I automatically detect duplicate content?
This requires the following 4 steps:
Step 1: Collect documents
First you need to know where the information is located and in what format. Wikis, shared drives and data shares are the usual suspects. In terms of formats, you will mainly be dealing with Word, PDF, HTML and, in marketing, InDesign. Pay attention to how often which formats are used. This will save you work in the next step.
Step 2: Extract content
This is the most technically complex step. The text of all documents must be brought into a uniform form. Pure text without markup or layout is best. This can be done automatically. There are tools that support this. If there is no ready-made solution for one of your formats, you have to make a choice. Either you (or the developer of your choice) write a small programme that extracts the text; or you ignore the format. Which is better depends on
- how often the format occurs
- how likely it is that there are duplicates in this format
Step 3: Vectorise
This step does the magic but comes with minimal effort. Computers do have a hard time processing text. But there are ready-made solutions for this obstacle. In Python, for example, packages like scikit-learn or gensim provide everything you need. They make it possible to turn your documents into vectors with just a few lines of code. And computers can work very well with vectors.
Put simply what is happening here is that a list of all the words that appear in your documents is created. Then it is counted how often each word occurs in the respective document. So the document becomes a series of numbers. These numbers can be understood as a point or vector in a coordinate system. Similar documents (i.e. those in which the same words occur similarly often) are close to each other.
Tip: Before you convert the documents, you should
- Remove stopwords. These are words that occur often but have little meaning. These include articles, linking words and auxiliary verbs. There are ready-made lists for most languages that you can use to filter out the stopwords automatically.
- Remove numbers. If two documents are the same except for the date, a phone number or the product version, they are still duplicate content. Therefore, replace numbers with a placeholder.
Step 4: Measure similarity
Now you only have to measure the distance between your documents. There are different methods of measurement. Common methods are:
- Euclidean distance (a straight line between the points)
- Taxicab geometry (sum of the differences for each coordinate)
- Cosine similarity (angle between the vectors)
Regardless of how you measure, you get a value for each pair of documents. You can easily find out which values indicate duplicates by taking samples.
In our projects we use the cosine similarity. It lies between 0 (documents without shared properties) and 1 (identical documents). Experience shows that from a similarity of 0.95, documents are duplicates. Mostly, a couple of short sentences are missing in one version or individual names have been exchanged. With values between 0.9 and 0.95, the documents are still very similar, but with important differences, such as an additional work step.
Let the programme you use for measuring create a list for each document, naming all documents particularly similar to it. This gives you an overview of all duplicates in your collection.
What do I do now with the duplicate content?
That depends on the case:
- If the media are different (e.g. flyer and brochure), you probably still need both versions. Make sure that all users know that there are multiple copies. For example, store all variants in one place or use shortcuts. Tip: The automatic detection of duplicates will alert you if the copies are unintentionally different.
- You should archive or remove old versions. This way you prevent someone from accidentally using outdated information.
- Drafts should be clearly marked as such and possibly kept separately. End users (whether internal or external) should only have access to final released versions. This way, only verified information is circulated.
- You can merge similar documents into one. This relates both to documents that describe similar things or processes; and to cases where work has continued on several copies of a document. This has several advantages:
- If anything changes, you only have to change the content in one place. You save time and no contradictions can arise.
- There is no danger of anyone confusing the cases, as the differences are clearly visible. This prevents mistakes.
- As you reduce the number of documents, it becomes easier to find the document you need. This saves time.
Duplicate content costs time and leads to errors. It creates additional work and can cause confusion. The only way to identify duplicates reliably and efficiently is to use automation. The most complex step here is extracting the text from the documents. Once this is done, you can quickly and easily create an overview of all duplicates. With this list, it is then easy to identify and eliminate problems and risks.