A few years back, I inherited a contract folder from someone who’d left the company three months earlier. It was on a shared drive called “Legal_OLD” and contained 2,400 files. Some were PDFs. Some were Word docs with track changes still on. There was a folder called “FINAL_FINAL_v3” that contained two contracts I’d never heard of and one signed agreement that had expired in 2019.

My boss asked me to “evaluate contract repository software.”

I spent the first week trying to schedule demos. By Friday I realized I was about to spend a budget cycle picking a tool to organize a folder I didn’t even understand yet. I canceled the demos and spent the next three weeks cleaning the folder instead.

That order matters. Clean the folder, then buy the software. Not the other way around. Here’s why, and what I’d actually look for once the mess is mapped.

Why we all reach for software first

The instinct is understandable. The folder is a disaster, the renewals are sneaking up, somebody on the exec team said “we need a system,” and there are forty vendors with demo buttons that all promise to fix it.

But contract repository software does not clean your folder. It moves your folder. If you move a mess into a fancier container, you now have a mess in a fancier container and an invoice for $18,000 a year.

I’ve watched this play out at three different companies. The pattern is the same:

1. Team buys repository software because the shared drive is chaos.
2. Implementation kicks off. Vendor asks for “your contracts.”
3. Team realizes they don’t know which contracts are current, which are superseded, which are drafts that were never signed, and who owns the relationship.
4. Implementation stalls for six months while somebody tries to figure that out.
5. Half the contracts get uploaded with the wrong metadata. The other half never get uploaded at all.
6. Two years later, somebody inherits the tool and writes a blog post like this one.

The software is not the problem. The problem is that nobody knew what they had before they tried to organize it.

What “cleaning the folder” actually means

I’m not talking about a deep cleanse where you read every contract. You will never do that and you don’t need to. What you need is a working map of three things:

• What contracts are active right now.
• Who owns each one.
• When each one renews or terminates.

That’s it. If you have those three things on a spreadsheet, you are ahead of about 80% of companies I’ve worked with.

Here’s the pass I run, and you can do it in a couple of weeks of focused time:

First pass — sort by file type and date. Anything older than the longest contract term you typically sign (say, five years) goes into an “archive” folder. Don’t delete it. Just move it out of the way. You can come back to it.

Second pass — group by counterparty. All the Acme Corp contracts go together. This is where you’ll find duplicates, superseded versions, and the SOW that everyone forgot was attached to a master agreement.

Third pass — flag the live ones. A live contract is signed, in effect today, and either auto-renewing or running to a future end date. Put these on a spreadsheet. Columns: counterparty, contract type, signed date, end date, renewal terms, owner, value if you know it, file path.

Fourth pass — find the owners. For each live contract, figure out who in the company actually deals with the relationship. That’s the owner. If you can’t find one, that’s a finding — write it down.

At the end of this, you have a spreadsheet of your real contract portfolio. Maybe it’s 80 contracts. Maybe it’s 800. Either way, you now know what you’re trying to organize, which means you can finally pick a tool that fits it.

What the cleanup teaches you

This is the part nobody warns you about. The cleanup is also a requirements-gathering exercise. By the time you finish, you’ll know things you couldn’t have known from a demo:

• You’ll know how many contracts you actually have. (It’s almost always fewer than people guess, and the active ones are fewer still.)
• You’ll know what types dominate. NDAs? Vendor agreements? Customer MSAs? Real estate? This shapes which features matter.
• You’ll know where the renewal landmines are — the auto-renew clauses, the 90-day notice windows, the ones that quietly escalate.
• You’ll know who the real stakeholders are. Not the org chart stakeholders. The people who actually get pulled in when a contract issue comes up.
• You’ll know which contracts are missing entirely. There’s always a few.

Now you can shop for software with a real shopping list, not a vague feeling that “we need a system.”

Practical buying criteria, in the order they actually matter

Once the folder is mapped, here’s what I look for. Not in the order vendors put it on their pricing page — in the order it actually matters when you’re using the thing every day.

1. Search that works on the inside of the document.

Full-text search across every file, including signed PDFs. If I’m looking for “limitation of liability” or “Acme Corp” or “auto-renew,” I want to type it in one box and find every contract that contains it. This is the single feature that makes a repository worth the money. Without it, you’re still opening files one by one.

2. A clear, editable metadata layer.

I want fields I can define: counterparty, contract type, owner, signed date, end date, renewal notice period, value, status. I want to add custom fields without calling support. I want to bulk edit them when I realize I’ve been spelling “Acme” three different ways.

3. Renewal reminders that go to a human, not a queue.

The renewal date is the single most expensive piece of metadata in your repository. The tool needs to send a notification to a real person, by email, with enough lead time to do something about it. I want to set this per contract. 30 days, 60, 90, 180 — whatever the notice period is. If the tool only does “30 days before end date” globally, that’s a problem.

4. Permissions that fit how your company actually works.

Not everyone should see every contract. Sales shouldn’t see the CFO’s employment agreement. The board shouldn’t have to log in to find the bylaws. I want roles I can configure, and the ability to share a single document with someone who doesn’t have a login. Bonus points if I can do that without creating a security incident.

5. Easy bulk upload.

You just cleaned 600 files. You need to put them in. If the upload process is one-at-a-time, or requires perfect metadata before a file will save, you’ll hate your life by week two. I want drag-and-drop folders, OCR running quietly in the background, and a “fix it later” mode for metadata.

6. Export that doesn’t trap you.

I want to be able to export every contract and every metadata field, as actual files and as a spreadsheet, without calling the vendor. If a tool makes this hard, that tells you everything about how they think about the relationship.

7. A price that scales with what you have, not what they hope you’ll have.

Be skeptical of pricing that bundles in “AI clause extraction” and “negotiation workflow” and “e-signature” when all you need is a searchable, secure, monitored repository. Those features are useful at certain scales, but you can add them later if they earn their keep. Pay for the thing you’ll actually use this year.

What I’d skip in year one

Some of the most pitched features in this category are not necessary on day one for most companies:

• Workflow automation for approvals. Useful when you have a high volume of new contracts moving through review. Not useful if you sign 30 contracts a year.
• AI clause extraction. Genuinely cool, sometimes accurate, often oversold. Worth piloting in year two when you have a clean base of contracts to point it at.
• Integration with every system you own. Pick one or two integrations that actually save time. Skip the rest.
• Custom reporting dashboards. A spreadsheet export and a pivot table will get you 90% of the way there for a long time.

A pragmatic order of operations

If I were starting over at a new company tomorrow, here’s the order:

1. Spend two to four weeks cleaning the folder and building the contract spreadsheet I described.
2. Use the spreadsheet to run renewals for a quarter. Calendar reminders. Plain email follow-ups. Old school.
3. Once I know my real volume, my real types, and my real headaches, demo three repository tools.
4. Pick the one with the search and metadata I trust. Negotiate the price.
5. Migrate the cleaned-up contracts in, not the original shared drive.
6. Set a date six months out to review what’s working and what isn’t.

The whole point is that the software is supposed to make a clean process faster. It is not supposed to be the clean process.

One practical next action

Open your contract folder right now. Sort by date modified. Count how many of the top 50 files you can identify in under 10 seconds — who the counterparty is, whether it’s active, and when it renews.

If that number is under 40, don’t buy software yet. You’re not ready to evaluate it. You’re ready to spend a week with a spreadsheet first. Then come back to the demos with a list of what you actually have.

You’ll save yourself a budget cycle, a bad implementation, and the slow embarrassment of paying $18,000 a year for a fancier version of the mess you started with.

---

I’m Dave, and I write about contract management the way it actually works. No jargon, no sales pitch, just what I’ve learned from 15+ years of doing this job. If this was useful, stick around.