Why doesn’t filing email attachments by hand work in the long run?

Because it depends on everyone’s discipline – and that invariably breaks down in day-to-day business. As soon as things get busy, attachments stay in the inbox, everyone files them differently, and after a few months the inbox has become the real archive. Automatic import gets round this discipline problem.

Which email attachments are imported automatically?

Rules decide this: for example, attachments from certain senders, certain file types such as PDF, or everything from a specified team mailbox. Signature images, logos and obviously irrelevant files are filtered out.

What happens if the same attachment is sent several times?

Duplicate detection checks whether a file with identical content has already been captured and, if so, doesn’t import it again. An email thread in which the same PDF goes back and forth five times still produces only one entry in the knowledge base.

Is the text of the email captured too, or only the attachment?

The email text can be captured as well, because the crucial information – a commitment, an appointment, a price quote – is often in the body of the email rather than in the attachment. Whether this makes sense, and for which mailboxes, is a deliberate decision rather than something switched on across the board.

Can employees’ personal mailboxes be connected?

Technically, often yes, but in practice it’s wise to hold back: personal mailboxes also contain private and confidential emails. A good approach is to connect only shared team mailboxes such as office@ or invoices@ at first, and personal mailboxes only with clear rules and the consent of the people concerned.

Who sees the imported documents in the knowledge base?

This is controlled by user and role management: documents from the accounting mailbox can, for example, be restricted to accounting, while project documents are open to the whole team. Because the document type helps decide who gets to see a document, a person confirms it in the review area before the attachment is released. It’s important to set these permissions before connecting, not afterwards.

Connecting Outlook and Microsoft 365

You can capture email attachments from Outlook automatically by setting up a connection to Microsoft 365 that continuously monitors specified mailboxes and imports relevant attachments into the central knowledge base according to fixed rules. Rules by sender and file type, duplicate detection and automatic assignment to customer or project replace manual filing, which invariably fails in everyday work. A person still confirms the document type in one click, because it decides who gets to see the attachment. The inbox stops being a hidden document archive and goes back to being what it should be: a way of communicating.

In brief

In many SMEs, the email inbox is the company’s largest unorganised document archive – order confirmations, delivery notes and commitments often exist nowhere else.
Manual filing doesn’t fail because of people but because of the approach itself: it depends on a level of discipline that can’t be sustained in day-to-day business.
Automatic import works with rules (sender, file type, mailbox), duplicate detection and automatic assignment to customer or project.
The system suggests field values such as customer, project, installation, date and topic and fills them in itself; a person confirms the document type in the review area, because it determines who can see the document.
The email text can be captured too, because the crucial information is often in the body of the email rather than in the attachment.
Connect team mailboxes first; with personal mailboxes, data protection comes first and restraint is advisable.

Why is the inbox a hidden document archive?

Ask anyone in your company where last autumn’s order confirmation from a supplier is, and the honest answer is usually: “Somewhere in Outlook.” Over the years, thousands of attachments pile up in mailboxes: quotes, delivery notes, inspection records, signed contracts, technical data sheets. Many of these documents exist nowhere else.

The problem isn’t the volume but access. A mailbox belongs to one person: if that colleague is on holiday or has left, what they know is practically out of reach. Outlook search does find emails, but not the content of a scanned PDF attachment. And anyone looking for something has to know who received the email at the time – which is precisely what nobody remembers two years later.

Why does filing attachments by hand fail?

The usual answer to this problem is: “From now on, everyone saves important attachments in the project folder.” Almost every company agrees on this rule at some point – and it rarely lasts more than a few weeks. That isn’t down to unwilling employees but to the approach itself: manual filing is an extra step with no immediate benefit to the person doing it, and steps like that are the first to go when time is short.

On top of that comes inconsistency. Even if everyone does file, everyone files differently: different folders, different file names, a different idea of what counts as “important”. The result is file storage that nobody trusts – so everyone searches in Outlook after all. A solution that relies on constant discipline is no solution; capture only becomes reliable once it happens automatically.

How does the automatic import of attachments work?

With a connection to Outlook or Microsoft 365, the knowledge base continuously monitors specified mailboxes and imports relevant attachments by itself. What counts as “relevant” is set by rules that are defined once during set-up. The system rests on four building blocks:

Rules by sender, mailbox and file type. Examples: “Import all PDF attachments from supplier domains”, “Capture everything from the invoices@ mailbox”, “Ignore signature graphics and logos”. That way, what comes in is substance, not clutter.
Duplicate detection. The same file often goes back and forth several times in an email thread. Duplicate detection checks whether a document with identical content has already been captured and, if so, doesn’t file it again – one entry instead of five copies.
Assignment to customer or project. Automatic tagging uses the sender, subject line and document content to work out which customer, project or topic an attachment belongs to and fills in these fields itself – so a delivery note from the Berger company ends up with the Berger case without anyone having to move it there.
Confirmed document type. The system doesn’t decide on its own what kind of document it is – invoice, delivery note, contract, record: it makes a suggestion, and a person confirms it in the review area. The reason is permissions, because the document type determines who may see the attachment later. For clear-cut attachments, this takes seconds.

Scanned or photographed attachments also go through text recognition (OCR, converting an image into searchable text), so the content of a photographed invoice can be found too. The mailbox itself is left untouched: the emails stay in Outlook, and the knowledge base works with copies.

What happens to the email text itself?

Often the crucial information isn’t in the attachment at all but in the body of the email: “Fine, we’ll hold the price until the end of the year”, “Delivery has been pushed back to calendar week 41”, “The complaint is accepted”. Commitments like these matter to the business – and if only attachments are captured, no file storage in the world will help you find them.

That’s why you can specify, when connecting, that the text of certain emails is imported as well – for example, the correspondence from a team mailbox. The email is then treated like a document: tagged, assigned to the customer and searchable. Ask “What did the Berger company commit to on the price guarantee?” and you get an answer that points to the specific email. A deliberate selection makes more sense here than capturing everything – not every three-line internal note belongs in the company’s knowledge.

How do you handle permissions and data protection in the team mailbox?

As soon as emails and attachments can be searched centrally, the question arises: who may see what? The answer lies in the knowledge base’s user and role management. Documents from the accounting mailbox can be restricted to accounting and management, while delivery notes and project documents are open to the whole team. Because the document type is the key to these roles, it is confirmed in the review rather than assigned by the machine. The order matters: think through permissions before connecting rather than patching them up afterwards.

Data protection needs to be on the table from the start, because emails contain personal data about customers, suppliers and employees. In broad terms, this means: connect only mailboxes with a clear business purpose, keep staff openly informed, limit access through roles and sign a data processing agreement (DPA – the contract setting out how the service provider handles your data) with the provider. Solutions like junova are built for this with EU servers, role management and a DPA, and don’t use your content to train AI models; responsibility for sensible rules within your own company, however, stays with you.

Where does automatic capture reach its limits?

To be honest, automation doesn’t solve everything. The biggest limit is personal mailboxes: business, private and confidential emails are mixed up there, and no rule can separate them reliably. A good approach is therefore to start with shared team mailboxes such as office@, purchasing@ or invoices@ and either leave personal mailboxes out altogether or connect them only with tight rules and the explicit consent of the people concerned.

Besides the approval in the review, some manual work remains on the content side too. Rules catch the typical, not the unusual: the one important attachment from an unknown sender slips through the net and has to be added by hand – for example by forwarding it to a capture mailbox or uploading it in the browser. And the rules themselves need looking after: new suppliers, new mailboxes and changed procedures should be reflected in them from time to time, or the filter will go out of date.

How do you get the connection up and running?

A practical start is a small one: a single team mailbox, a manageable set of rules, then a few weeks of watching what gets imported and what’s missing. Based on that, you refine the rules and add further mailboxes. This step-by-step approach matches the general advice to connect sources one at a time rather than all at once – the article Connecting documents and systems gives an overview of all sources, from the network folder to the paper archive, along with the recommended order.

Realistically, the Outlook connection is one of the most rewarding steps in the whole digitalisation process: the set-up effort is manageable, the ongoing effort comes down to briefly confirming the document type, and the difference in everyday work is felt straight away – because searching other people’s mailboxes, asking the colleague who’s on holiday and “Can you send me that again?” simply become a thing of the past.

Connecting Outlook and Microsoft 365: capturing email attachments automatically