How to Cut Your Sales Content Library in Half Without Reps Noticing

Somewhere in your content management system there's a folder called "Objection Handling — Q3 2019." Nobody has opened it in fourteen months. It isn't there because anyone believes it still works. It's there because deleting it would require someone to be responsible for that decision, and nobody enjoys being the person who binned the one deck a quiet top performer secretly still opens. So the folder stays. Next to it sit another two hundred like it, and the library grows the way a garage grows — not through any single bad call, just the compounding cost of never throwing anything out.

I've now run this audit at four companies this year, ranging from a 35-person SaaS floor to a 200-rep insurance broker. The pattern is identical every time: a library with 250 to 400 assets, of which somewhere between 15 and 25 percent get opened at all in a rolling quarter, and a top 5 percent that accounts for most of the usage that does happen. Reps aren't confused about what's good. They've already voted with their clicks. Enablement just hasn't counted the votes.

Why libraries never shrink on their own

Adding content is visible work. A new battlecard, a refreshed deck, a fresh competitive brief — someone built that, it shows up in a status update, it looks like progress. Removing content is invisible work with visible downside risk: if you delete the wrong thing and a rep needed it next Tuesday, you hear about it immediately. If you leave two hundred dead assets sitting there costing nothing but search time and platform licence fees, nobody complains, because nobody's looking at them anyway. That asymmetry is the entire reason your library only ever grows. Nobody is incentivised to prune it, so it becomes an org chart of every enablement hire's tenure, layered on top of each other, never subtracted.

The fix isn't a better content platform. It's ninety days of data, five honest conversations, and the nerve to act on both.

Step 1: pull ninety days of usage, not opinions

Before you ask a single rep anything, get the numbers. If you're on Highspot, Seismic, Showpad or similar, pull the asset-level usage report: opens, downloads, shares, and — critically — last-touched date. If your library lives in a shared drive with no usage tracking, that's your first finding, and you should say so in the readout: you cannot govern what you cannot measure.

Sort every asset by opens in the last ninety days. You'll get a curve, not a cliff: a small cluster used constantly, a long tail used never, and — this is the part people skip — a middle band used rarely but not zero. That middle band is where the real judgement calls live, and it's exactly why step two exists.

Step 2: five interviews before you delete anything

Don't survey the whole floor. A survey gets you answers reps think enablement wants to hear. Five structured interviews, thirty minutes each, get you the truth, because you're not asking "what content do you use" — you're asking "walk me through the last deal you nearly lost, and what you went looking for." Behaviour, not opinion.

This step exists to catch what the logs can't see. A rep who downloaded a pricing objection sheet once, in March, then saved it to their desktop and has referenced it in every deal since, shows up in your data as one open in ninety days — indistinguishable from something genuinely dead. Delete on logs alone and you'll cut something load-bearing. The five interviews are your check against exactly that failure mode, and they're cheap enough that skipping them to "move faster" is a false economy.

Ask about new hires too, even if you didn't interview any — a piece with low overall opens but heavy use in someone's first ninety days is protecting your ramp, not decaying. That's a different asset than the one nobody has opened since it was published. Worth cross-checking against whatever you use to onboard, like a New-Hire Ramp Content Checklist, before you assume low volume means low value.

Step 3: three buckets, not two

The temptation is "keep" and "delete." Resist it — you need a third bucket for the assets your logs flagged as dead but your interviews flagged as fine, which usually means the content is good but undiscoverable.

That middle bucket usually runs 10 to 15 percent of the library in my experience, and it's the bucket that saves you from the accusation that you're "gutting" the library. You're not deleting good content that's badly filed. You're deleting content nobody, anywhere, is relying on.

Step 4: retire quietly, communicate the keep list loudly

Don't send an email announcing you've deleted 180 assets. That invites every rep to go check whether their personal favourite survived, which is the fastest way to manufacture panic over things that were never actually load-bearing for anyone. Instead, communicate what's staying: a shorter, cleaner, better-tagged library, with a note on where the fixed assets moved to. If someone comes looking for something that's gone and it turns out they genuinely needed it, restore it and log why — that's real signal for your next pass, not a failure of this one.

What "without reps noticing" actually means

It doesn't mean hiding the change. It means the cut was accurate enough that nothing anyone relied on is missing. If you've done the audit properly, the only people who notice are the ones who now find things faster, because there's a third of the clutter between them and what they need.

Score what survives with something like a Sales Content Effectiveness Calculator so "keep" isn't just a vibe, and put a cadence in place — a Sales Content Calendar Template works — so the library doesn't quietly re-bloat back to 400 assets over the next eighteen months. Otherwise you'll be having this exact conversation again, at the same company, with a different enablement lead's name on the old decks.

The tool you were about to buy to manage a library nobody trusts costs more, every year, than the ninety days it takes to find out what's actually being used.