AI knowledge base in practice: the most common company use cases

In Austrian small and medium-sized companies, the most valuable knowledge rarely sits in records – it is in people’s heads and in scattered documents: how did we solve this problem last time? What settings keep the installation running reliably? An AI knowledge base brings a company’s documents together in one place and answers questions like these in plain language – with a source pointing to the original document. The most common use cases: finding earlier solutions to problems, looking up documented installation settings, having maintenance know-how to hand on the move, onboarding new employees, holiday cover, holding on to experience before retirement and answering phone enquiries quickly. Finding records such as quotes or delivery notes comes as a useful bonus.

In brief

• An AI knowledge base answers everyday questions from your own company documents – and every answer points to its source.
• The strongest use cases are all about know-how: earlier solutions to problems, installation settings, maintenance knowledge, onboarding, cover and holding on to experience before retirement.
• On site or at a customer’s, mobile access is what counts: instructions and settings are on your phone rather than in a folder back at the office.
• Knowledge no longer depends on individual people – which makes a noticeable difference during holidays, sick leave, onboarding and staff changes.
• To be clear: the ERP knows what was sold – the knowledge base knows how things are done. Finding records is a bonus, not the main point.
• You can spot your own top use cases from the questions that keep coming up in the team and from knowledge that only one person has.

What is an AI knowledge base – and how does it differ from storing files in folders?

An AI knowledge base is a central store for company documents that you can ask questions, as you would a well-informed colleague. Instead of clicking through folder structures or trawling through your inbox, you type a question in plain language – for example “How did we solve the condensation problem with the ventilation system at the Bergmann property?” – and get an answer taken straight from your own documents. The article What is an AI knowledge base? gives a detailed introduction to how it works.

It differs from traditional file storage in three ways. First, it searches by meaning, not just by file name. Even if you don’t know the exact wording or where the file is stored, you will still find the right document – even if the service report says “condensation” and your question says “moisture”. Second, text recognition (OCR – the automatic conversion of scanned pages and photos into searchable text) makes paper documents searchable too once they have been scanned – from the old machine logbook to the handwritten maintenance record. Third, every answer names its source. One click opens the original document – you don’t have to trust the AI blindly.

It is also important to distinguish it from general AI chatbots on the internet: a business knowledge base answers only from your own documents. If something isn’t in the documents, the system tells you so. That is precisely what makes it reliable in day-to-day business. Why this difference is so important, and how ChatGPT with your own data works, is a subject in its own right.

Just as important is the distinction from the ERP system (the business software for orders, stock and invoices). The ERP manages structured business data – who ordered what and when, which invoice is outstanding, how much is in stock. The knowledge base unlocks the unstructured knowledge alongside it: instructions, service reports, settings records, meeting notes, email threads. To put it in a sentence: the ERP knows what was sold – the knowledge base knows how things are done. It doesn’t replace an ERP; it closes the gap the ERP leaves.

Which use cases come up most often in everyday business?

The following seven use cases crop up across all industries – from the trades to manufacturing to law firms. Each follows the same pattern: a starting situation, a question in everyday language, an answer backed by a source – and a noticeable difference to the working day.

Use case	Typical question	Who benefits most
Finding earlier solutions	“How did we solve that last time?”	Technicians, project managers
Installation settings	“What values keep the machine running reliably?”	Production, maintenance
Maintenance know-how on the move	“What does error code E42 mean, and how do I fix it?”	Field service, installation
Onboarding and induction	“What is our procedure for a complaint?”	New employees, trainers
Holiday cover	“Where are the documents, and what was agreed?”	Office, administration
Retirement	“What does only the colleague who is leaving soon know?”	Management, successors
Phone enquiries	“When was the last service for customer Huber?”	Back office, front desk

“How did we solve that last time?” – finding earlier solutions

The situation: a plumbing and heating business with fifteen employees is faced with a heating problem at a customer’s that seems oddly familiar. Someone in the team has solved something like this before – two or three years ago, at a different property. But who? And how? The project manager at the time has left the company, and the service report is somewhere on the server. So the problem gets solved all over again, with the same wrong turns and the same effort as the first time.

Example question: “Have we ever had underfloor heating that stays cold in just a few rooms, and what was the cause?”

What the answer gives you: the relevant service reports and project notes from the past, summarised in a few sentences – for example, that in two earlier cases a sludged-up manifold was to blame, and how it was flushed out. Every statement refers back to the original document.

What changes day to day: the company stops solving the same problems over and over. Experience becomes a tool everyone can use rather than the private property of a few individuals. And people are more willing to jot down solutions – because recording them now pays off: they will be found again.

Looking up documented installation settings and machine parameters

The situation: at a manufacturing company with thirty employees, an older press brake only runs cleanly when pressure, speed and tooling are set in the right combination. These values aren’t in any manual – they live in the heads of two machine operators and on notes in the tool cabinet. When an order comes in for a material that was last processed a year ago, the guesswork starts all over again.

Example question: “What settings did we use last time to bend 3-millimetre stainless steel on the press brake?”

What the answer gives you: the documented settings from the last comparable order – from the photographed settings note, the set-up record or the project note, with a reference to the source and the date.

What changes day to day: set-up times no longer depend on who happens to be on shift. Less material is scrapped on failed first attempts. And the notes in the tool cabinet finally have a home where people actually find them: a photo of the handwritten settings note is enough, and text recognition makes it searchable.

Maintenance know-how on the move: on your phone in front of the installation

The situation: a service technician from a refrigeration and air-conditioning business is at a customer’s, standing in front of a unit showing an error message. The operating manual is a PDF on the server back at the office, the printed version is in a folder in the workshop, and the vital piece of experience – that on this model the fault is almost always the pressure sensor – is in a colleague’s head. Ringing the colleague in the back office interrupts her work, and his own too.

Example question (on the phone, right in front of the unit): “What does error code E42 mean on the Model X cooling unit, and how do I fix it?”

What the answer gives you: the relevant section of the stored manual, summarised in clear steps, with a link to the original page – and, if it has been documented, the note from earlier service reports that on this model you should check the pressure sensor first.

What changes day to day: the technician carries on working – no waiting for a call back, no trip back to the office, no second appointment. The office gets interrupted less. And for the first time, manufacturer information and the company’s own experience sit side by side in one place – available to everyone in the team, not just the longest-serving technician.

Induction and onboarding: new employees find answers themselves

The situation: a joinery takes on a new work planner. In his first few weeks he asks dozens of questions a day: how are orders created? What is the procedure for a complaint? Who orders fittings, and from whom? Every question is perfectly reasonable – and every one interrupts an experienced colleague. After a few weeks, the newcomer stops daring to ask some questions and starts improvising instead. That is exactly when mistakes happen.

Example question: “What is our procedure when a customer complains about a cabinet front we’ve delivered?”

What the answer gives you: the documented procedure, step by step – who is informed, which form is used, how photos are taken and everything is recorded – with a reference to the stored work instruction.

What changes day to day: new employees answer most of their everyday questions themselves and only ask when real experience is needed. Onboarding becomes easier to plan, experienced staff are interrupted less often, and the answers are consistent – rather than three versions from three colleagues.

Holiday cover and sick leave: always able to answer

The situation: the office manager at an electrical business is on holiday for two weeks. She is the only one who knows where the registration documents for the company lorry are and what was agreed with the tax adviser. The colleague covering for her sits in front of the same folders – and still can’t find anything, because she doesn’t know how they are organised. Sick leave is even trickier: there is no handover, and nobody likes to ring someone who is ill.

Example question: “What do we do when someone calls in sick, and where do I find the form?”

What the answer gives you: the documented procedure, including the form – without disturbing anyone on holiday and without searching through someone else’s folder structure.

What changes day to day: the person covering can still get things done, and customers and suppliers don’t even notice that someone is away. A holiday really is a holiday – and the company’s ability to answer questions no longer depends on the rota.

Retirement: the knowledge doesn’t retire too

The situation: at a metalworking company, the workshop manager is retiring after thirty years. He knows every machine and its quirks and keeps his own records for recurring custom jobs. What traditionally happens: a few weeks of handover, a thick folder, lots of good intentions – and a year later his successor is still ringing him regularly.

Example question (months after he has left): “What do we need to watch out for when setting up the old lathe for brass parts?”

What the answer gives you: the relevant notes from the former workshop manager’s records that were taken over – scanned notebooks, photographed settings notes, short records from the handover period – with a source pointing to each document.

What changes day to day: the handover gets a goal and a structure. In his last few months, the company deliberately collects the questions that keep landing on the departing colleague’s desk – and writes down the answers informally. You can check whether the knowledge has been captured before he leaves: ask the knowledge base the typical questions and see whether useful answers come back. If they don’t, you can still ask – after all, he is still there.

Phone enquiries: answering customers’ questions straight away

The situation: a customer rings to ask whether their appliance is still under warranty or when it was last serviced. The usual answer: “I’ll need to check and call you back.” The call back takes time – and sometimes it is forgotten. The customer is left with the impression that nobody really knows what’s going on.

Example question (during the call): “When was the last service for customer Huber, and what was done?”

What the answer gives you: the information while the customer is still on the line, backed by the service report.

What changes day to day: “I’ll call you back” becomes “I can tell you now” – a noticeable difference for the customer. At the same time, anyone in the office can answer, not just the person who looks after that customer. Call-back lists get shorter.

A bonus: finding records, quotes and contracts faster

Besides know-how, the knowledge base also makes the usual business documents easier to get at. An old quote for a similar heating renovation can serve as a template for costing. The notice period in the maintenance contract, along with conditions agreed later by email, is just one question away. And a scanned delivery note signed by hand can be matched to the right invoice before it is paid.

It is important to keep this in perspective: for orders, stock levels and outstanding items, the ERP remains the system of record – the knowledge base doesn’t replace stock management and doesn’t post anything. Its contribution starts where the ERP stops: with the wording of the contract, the reasoning in the email thread, the scanned record. If you judge its value by how well it handles records, you underestimate it – its real value lies in knowing how things are done.

A day with a knowledge base: before and after

Comparing an ordinary working day at a fictitious trades business with twenty employees shows how the use cases add up over the day – not a special case, just everyday life.

Situation	Without a knowledge base	With a knowledge base
07:30 – technician needs the settings for an installation at a customer’s	Rings the workshop, a colleague looks for the note, calls back	Asks on the phone and gets the values with their source, right in front of the installation
09:15 – new employee asks about the complaints procedure	Experienced colleague stops what he is doing to explain – for the third time this week	New employee asks the knowledge base; the colleague isn’t disturbed
10:40 – customer rings to ask about the last service	“I’ll call you back” – search through the folder, call back in the afternoon	Answer during the call, backed by the service report
13:00 – a problem with a ventilation system seems familiar	More guesswork, because nobody knows who solved it last time	Old service reports found: cause and solution are documented and ready
15:20 – office manager off sick, colleague covering looks for an agreement	Ring her while she’s ill, or decide based on guesswork	The colleague finds the documented agreement herself
16:45 – quote for a project similar to one two years ago	Costed from scratch, because nobody can find the old quote	Old quote opened as a template, line items reused

None of these situations is dramatic on its own – a quick question here, leafing through a folder there, an answer put off until later. Because the effort is spread across several people and over the whole day, nobody sees it as a cost in its own right – but it still shapes the working day. It is this very invisibility that makes time spent searching so insidious; the article Time lost searching for documents takes a detailed look.

How do I find the top 3 use cases for my company?

You won’t find the most worthwhile use cases on the drawing board but in your own day-to-day work. The following self-check helps you narrow them down within a week. Go through the checklist and count how many times you answer yes:

• Are there questions that come up in the team several times a week – among colleagues or from customers?
• Do you regularly hear “Only Franz knows that” (or Maria)?
• In the last few months, did someone solve a problem that had been solved before – without anyone finding the old solution?
• Are important settings or work steps written on scraps of paper, in private notebooks or only in people’s heads?
• Do technicians or field staff regularly ring the office to ask for documents or values?
• Do new employees ask the same colleagues the same questions for weeks?
• Does the company struggle when individual people are on holiday or off sick, because information is missing?
• Will an experienced employee whose knowledge isn’t documented anywhere retire in the next few years?
• Do people often say on the phone, “I’ll need to check and call you back”?
• Are the same documents stored in several places – server, inbox, paper folders – with nobody sure which version is current?

How to read the result: three or four yeses mean you have clear use cases – the questions where you nodded are your candidates. Five or more yeses mean the company is very probably losing time every day to searching and asking around; a structured review is worthwhile.

Three criteria help you rank your top 3. First, frequency: a question that comes up every day beats one that comes up once a month. Second, risk: knowledge that only one person has counts for more than knowledge that is tedious to find but findable – especially if that person is leaving soon. Third, the documents available: a use case for which documents already exist can be tackled immediately; one that still needs documentation creating goes into the second round.

It also helps to look at how your files are stored: file names such as “Instructions_final_v3_NEW” mean nobody is sure which version is current – a clear sign. As a rule of thumb, start with one or two use cases where the pressure is greatest. A small start that works well wins the team over more than a big project that tries to do everything at once.

Which expectations are unrealistic?

Being honest also means saying what an AI knowledge base doesn’t do. If you start with false expectations, you will be disappointed after four weeks – usually on points that could have been cleared up in advance.

It only knows what is documented. The knowledge base makes existing documents searchable – it doesn’t create knowledge. The workshop manager’s experience can only be looked up once it has been recorded, at least informally: as a note, a photo or a short record. The good news is that this becomes much easier, because even unstructured snippets get found. Nobody has to write a manual.

It doesn’t tidy up your files. Duplicate files and outdated versions stay duplicate and outdated – the AI just finds them faster. If you keep an old and a new work instruction side by side, you may get answers from both. Some basic upkeep is part of the deal: archive or mark outdated material.

It doesn’t make decisions. The knowledge base provides the documented facts – a person still decides whether to settle a claim as a goodwill gesture. That is deliberate: you can’t delegate responsibility, but you can delegate finding good information.

Text recognition has its limits. Clean scans work well; handwriting is the hardest case and you will only see how well it works on your own documents; faded carbon copies or very hurried handwriting may be recognised only in part. For critical old documents, spot checks are a good idea.

It replaces neither the ERP nor training. Orders, stock and invoices remain the business software’s job; teaching hands-on skills remains the job of experienced colleagues. The knowledge base takes the burden of answering questions off both – nothing more, and that is enough.

There are no perfect answers from day one. The first few weeks show which questions the team actually asks and which documents are still missing. If you plan this phase in as fine-tuning, you will be further on after a month than someone who gives up at the first “I couldn’t find anything on this” – because that honest answer is precisely the hallmark of a reputable system.

How do these use cases make it into your company?

Getting from idea to everyday practice is quicker than many people expect – as long as you go about it in a structured way. This order works well: first an intro call to identify the specific use cases – ideally using the results of the self-check in this article; then an analysis of your existing documents; then the set-up with the most important documents; after that, we go through the answers to real questions from your day-to-day work; and finally the introduction for the team. The How it works page shows what this looks like in detail. The article Connecting documents and systems covers which storage locations and systems can be connected as sources.

What to look for when choosing a solution: answers must come only from your own documents and always name the source; text recognition should handle scanned documents; access should work on mobile devices; and role management should control who can see which documents – the payroll documents are not everyone’s business. For companies in Austria, servers in the EU and a data processing agreement (DPA – the data protection agreement with the provider) are also important criteria; the article AI and the GDPR in Austria explains the details. One solution developed for these requirements of Austrian SMEs is junova, the AI knowledge base from Goma-it. If you come across a technical term you don’t know – from OCR to role management – you will find it explained in the glossary.

One last piece of advice, and the most important: start with the questions, not the technology. Once you know the ten most common questions in your company – and whose knowledge is at risk of disappearing with the next retirement – you know exactly which documents belong in the knowledge base first. The benefit then shows in a very tangible way: the first time nobody has to answer the same question twice.