Legal teams over-preserve data because “keep everything” feels safer than making a defensible call about what can be released. It rarely is. Over-preservation trades one risk, spoliation, for another. It shows up as ballooning storage costs, slower review, and a data footprint that becomes its own liability the longer it sits.
Why over-preservation happens
Most legal holds start narrow and then widen under pressure. A custodian list grows because nobody wants to be the person who left someone off. A date range extends because narrowing it feels like a judgment call with no clear owner. Data sources pile on, laptops, shared drives, Slack channels, because collecting everything feels lower risk than deciding what’s actually relevant.
None of this is careless. It’s a rational response to an asymmetric risk. Under-preserving and facing a spoliation motion is a visible, career-defining failure. Over-preserving is invisible. It shows up months later as a storage bill, a review budget overrun, or a data map nobody can fully explain in the next matter. The incentive to err toward “more” is built into how legal holds get made, not a training gap.
What over-preservation actually costs
The costs are real even when they’re diffuse:
- Storage scales with volume held, not volume reviewed. Every extra custodian, every extra year of retention, adds to a bill that continues long after the matter that justified it closes. [NEEDS SOURCE: current benchmark for enterprise legal hold storage cost per TB/year]
- Review cost scales with what’s collected, not what’s relevant. Broader collection means more data moving through processing and review before anyone determines it doesn’t matter. [NEEDS SOURCE: benchmark for review cost per GB]
- Unstructured legacy holds compound. Data preserved “just in case” for one matter often never gets released, so it sits, unmanaged, until the next investigation has to account for it too. This is the dynamic behind what UnitedLex’s preservation team calls a Prevent Over Preservation, or POP, Analysis. It identifies data that was preserved defensibly at the time but has no ongoing legal basis to keep.
The point isn’t that preservation is bad. It’s that preservation without a release mechanism is a one-way ratchet. Every matter adds data. Almost nothing takes it away.
The real trade-off isn’t “preserve more or preserve less”
Framing this as a choice between aggressive preservation and cutting corners misses what’s actually happening. The best legal hold programs don’t preserve less to save money. They document why each custodian and data source is in scope, and they release holds on a defined schedule once a matter closes, instead of letting them pile up indefinitely.
That distinction matters more than volume. A general counsel weighing which partner to trust with preservation, collections, processing, review, and production isn’t really asking “who preserves the most.” They’re asking who can explain, in a deposition or an audit, why the scope was what it was, and who has a repeatable process for narrowing it again once the obligation lifts. Defensibility comes from documented judgment, not from breadth.
What a defensible preservation partner makes easier
This is where the practical decision point sits. The right partner doesn’t just execute preservation; it makes the scope decisions faster to make and easier to defend later. In practice, that looks like:
- Scope decisions get documented and revisited on a set cadence, so a legal team never has to reconstruct the reasoning from memory months later, under pressure, in front of opposing counsel.
- Collection targets metadata and custodian relevance from day one, so review starts against a smaller, more relevant set instead of the widest possible net.
- Holds come with a built-in release process, so data comes off the books as soon as the legal basis for keeping it ends, instead of sitting there indefinitely, waiting for someone to remember it’s there.
None of this requires preserving less aggressively when the situation calls for it. It just means the team spends less time reconstructing old decisions and more time acting on new ones.
Related content
For teams working through how preservation obligations intersect with collaboration platform data, or looking to jumpstart a POP Analysis on unstructured legacy holds, UnitedLex’s Litigation & Investigations team covers both in more depth on the Preservation Services page. UnitedLex’s forthcoming Preservation and Forensics report, running through early October, will go deeper on where these obligations are heading next.
