Organize contracts in bulk with AI

If you have a lot of contracts scattered across different tools and folders created by different people with different naming systems, and now you want to organize them, but you don't want to spend too much time, this video is for you. I'm going to show you how you can use the fynk Assistant to organize documents at scale just by chatting. Let's get started.

I'll start with this batch of contracts here.

I simply grab and drop them into fynk.

I'll give this import a name, something like all contracts, and then choose Signed Documents. That doesn't mean all of these contracts are already signed.

Actually, some are drafts, some are still in process, and some are completed.

We'll organize the stages properly later with the AI assistant.

Once I start the import, I'll be redirected to the import page, and the AI analysis starts scanning all contracts automatically.

AI analysis reads each document, extracts the important information, like deadlines, party names, clauses, and any relevant piece of info, and stores it in fynk. Once that information is extracted, it becomes much easier for the assistant to understand the documents and help us organize them.

It can take a few minutes for the analysis to finish on a larger batch.

And already you can see one of the problems we're going to fix.

The document names are inconsistent and pretty messy.

That's normal when different people have worked on the contracts over time, because everyone tends to use a slightly different naming system.

Once the analysis is finished, you could review and confirm the results for every document individually. For this example, I'm going to apply all of the results at once. So I'll click here to apply all AI analysis results.

Let's go back to the Documents tab, where we can see all the imported documents.

And now the analysis has been applied to the whole batch. We're ready to start organizing.

Now I'll open a new AI chat, and we can start with something simple. I'll ask, how many contracts did I just import?

And it tells me that I imported33 new documents.

Next, I'll ask what types of contracts they are.

The assistant searches across those33 documents and gives me a breakdown.

Most of them are service agreements. I also have a couple of SAS agreements, NDAs, employment contracts, a policy, a rental agreement, a partnership agreement, and a license agreement.

Now let's start cleaning them up.

The first thing I want to fix is the document names. So I'll ask the assistant to rename all of these contracts using the document type first, followed by the counterparty name.

I'll send the request.

One important thing here is that the AI assistant doesn't only read and answer questions about your documents, it can also make changes to them.

But before it makes an important change, it asks for confirmation.

Here, it gives me a preview of how the new document names will look and asks whether I want to rename all33 documents.

I'll confirm and click Rename.

After a few seconds, it's done.

Let's go back to the document list and check.

Now the document names are much cleaner and more consistent. I can also keep the AI assistant open here and continue the same conversation while I'm working in the document hub.

The next thing I want to do is organize some of these documents with tags. I'll ask it to create a tag called Software and apply it to all SaaS agreements and license agreements. Again, because this changes the documents, it asks me to confirm. It tells me that the tag will be applied to seven documents. I'll click Create and Tag.

Done.

After a refresh, you can see the new software tag applied to those agreements.

Now I want to identify some of the higher value contracts. I'll ask, can you tag all contracts with a value of 15,000 euros or more as key contracts?

The assistant first finds the documents that meet that threshold. In this case, it finds eight.

I don't actually have a tag called Key Contracts, but I do already have one called Key Account, so the assistant suggests using that existing tag instead.

That's useful because I don't have to remember the exact name of every tag in my account. I can just describe what I mean in natural language.

I could still create a new Key Contracts tag if I wanted, but here I'll use the existing Key Account tag.

It asks whether I want to apply it to all eight documents. I'll choose Yes, tag all eight.

And now those higher value contracts are tagged as key account.

You can also take this a step further.

If you already have a set of tags in your account, you can ask the assistant to use those existing tags to organize the rest of the documents.

So I'll ask it to use the current tags in my account and tag the remaining contracts.

The assistant groups the documents and suggests relevant tags for each group.

For example, it finds NDAs and suggests confidentiality.

For service agreements, it suggests professional services.

For employment contracts, it suggests HR internal. And for a few other documents, like the rental agreement, insurance policy, and partnership agreement, it suggests internal.

It asks me to review the suggestions and apply them.

I'll submit them.

If anything is still untagged, the assistant tells me.

Here, there's one NDA left without a tag, so I'll just tell it to use confidentiality for that one as well.

Now all of the contracts are tagged.

Next, I want to assign each document to the relevant team, so the right people have access to the right contracts. I'll simply ask the assistant to assign each document to the relevant team based on its document type. Again, instead of making assumptions, it asks me a series of questions.

For each group of contracts, I can choose which team should be responsible.

It goes through the document groups in batches, and I select the relevant team for each one.

At the end, I confirm the assignments.

After reloading the page, the Teams column is now filled as well.

So just by chatting with the assistant, this document list is already much more structured than it was when we started.

Now let's organize the stages.

As I mentioned earlier, some of these imported contracts are already signed, while others are still drafts or need review.

I want to move the completed ones to done.

So I'll ask, move all contracts that have a start date before today to the done stage.

The assistant notices that these contracts contain more than one relevant date.

For example, there's a start date and an effective date.

Instead of guessing which one I mean, it asks me to clarify.

I'll choose effective date.

It then finds17 agreements where the effective date is already in the past.

I'll confirm that I want to move all of them to done.

And now, in the stage column, you can see those17 contracts have been moved to done.

There's one last thing I want to do.

I want to identify contracts that are coming up for renewal soon. So I'll ask the assistant to tag all contracts that will renew within the next six months as important.

I don't currently have an important tag in my account, so the assistant asks whether I want to create one.

I can also choose the color.

I'll make it red.

The assistant finds eight documents where the relevant cancellation or renewal date falls within the next six months and asks whether I want to tag them. I'll confirm. And now those contracts are marked as important. You could keep going from here and ask similar questions, depending on how you want to organize your own contract library. But for this example, you can already see the difference. The document list is much more structured. The contracts have consistent names, relevant tags,the right team assignments, and clearer stages. I can also see the parties and other extracted information which makes the whole document list much easier to navigate. And all of this was done across a large batch of contracts by using the AI Assistant instead of updating every document one by one. I hope you like this video. See you in the next one!