Jump to Section
Yes, it's there. But can you find it? A construction Q&A for IT leaders, on records that actually hold up
Construction records are there but can you find them and does retrieval hold up?
Every construction firm says its records are backed up. Almost none have tested whether those records are usable when it actually matters, such as for the auditors, or a FOI request. This series answers real questions from construction IT and operations leaders about the gap between "we have it somewhere" and "we can produce it right now."
1. Our lead project engineer just retired. Is there anything only she knew how to find?
Reality check:
On most projects, one or two people become the unofficial index for where things are actually saved. When they leave, that knowledge leaves with them, and everyone else is left searching folder by folder for records that were never properly indexed in the first place. This is one of the most common and least discussed risks in construction records management.
The workaround:
The usual response is a handover document: the departing engineer writes down where things are, or spends a few weeks walking a replacement through it. That transfers some of what they knew, but only what they thought to write down, and only if there was enough notice before they left.
2. We've got 15 years of project files in SharePoint. If an owner disputes a closed-out job, could we actually find the right version?
Reality check:
Files from a 2011 project are technically still there. Whether the CAD file still opens, whether it's the version that was actually built to, and whether anyone remembers which folder holds the final approved revision, is a different question. Most firms only test this the day a claim arrives, which is the worst possible time to find out the answer is no.
The workaround:
Right now, that means someone calls the project engineer who worked the job - if they're still at the firm - or manually digs through shared drives comparing file dates and folder names, hoping the "FINAL_v3_reallyfinal" file is actually the one that got built. It works, eventually, if you have weeks and the right person still answers the phone, but unfortunately litigation and audits don't give you either.
3. Our project engineers say everything's backed up. Legal says that's not the same as defensible. Who's right?
Reality check:
Both are right, which is the problem. Backed up means the file exists somewhere. Defensible means you can produce the exact version, prove it hasn't been altered since handover, and show the chain of custody, all inside a timeframe legal must react to. Most firms have the first and assume it covers the second.
The workaround:
Teams bridge the gap with process: a naming convention, a sign-off email trail, someone's memory of what happened at handover. That's not nothing, but it's not evidence either. It depends entirely on the people who did it still being around, and remembering it the same way legal needs it explained.
4. Do we need to rip out SharePoint to fix this, or can it stay where our teams already work?
Reality check:
This is usually an early objection, and understandable. Nobody wants to ask site teams and project managers to learn a new system on top of everything else, and nobody wants to justify a new procurement line for a problem most people don't believe exists yet.
The workaround:
Most firms respond with discipline instead of a system: tighter folder structures, naming conventions, maybe an annual push to get IT to archive aging projects before they fall out of active storage. That holds up for a while, until the convention breaks down at scale, someone stops following it, or the person who set it up moves on. It's a rule standing in for infrastructure.
5. We're piloting Copilot across project teams. Can it actually see records from five years ago, or just this year's SharePoint?
Reality check:
Copilot and other AI tools are only as good as what they can read. If older project records sit in obsolete formats, disconnected file shares, or legacy systems outside SharePoint, AI tools can't surface them, which means five years of institutional knowledge is invisible to the exact tools meant to make use of it.
The workaround:
Some teams try to close this gap by hand, migrating old project files into current SharePoint sites project by project so Copilot has a chance of indexing them. It surfaces a handful of legacy jobs, slowly, and only the ones someone remembered to move. It doesn't scale past a pilot.
6. We're acquiring a subcontractor. How fast could we actually produce their full project history if we needed to?
Reality check:
Due diligence and post-acquisition integration both depend on being able to produce a clean, complete project record on short notice. If that record lives across someone else's legacy systems, personal drives, and inconsistent folder structures, the honest timeline is weeks, not days, and gaps in the history become your risk once the deal closes.
The workaround:
Most firms handle this by parachuting someone in during due diligence to manually pull together whatever project files they can find across the target company's drives, emails, and legacy tools. It surfaces some of it, on a deadline, with no real way to confirm the parts that matter most weren't missed.
7. Doesn't Microsoft 365 already have this covered?
Reality check:
Microsoft 365 is genuinely good at retention policies, compliance holds, and short-to-medium term storage. What it isn't built to do, is actively preserve content over decades: migrating files forward as formats go obsolete, verifying that a record hasn't changed since it was filed, or keeping a 2011 CAD file openable in 2031. Retention tells you a file must be kept. It doesn't tell you the file will still be usable when someone needs it.
The workaround:
Most IT teams lean on retention policies as if they were a preservation plan, or have someone occasionally sweep for aging file formats and manually re-save them before they might stop opening. It's not scheduled, it's not comprehensive, and it depends on someone remembering to do it before a format breaks, not after.
So what actually closes the gap?
Look back across these seven questions and they're not seven different problems. They're one problem showing up seven times: something exists, but nobody has truly tested whether it can be found, trusted, or produced under pressure.
Everything above happens to describe SharePoint, because that's where most of these conversations start. But SharePoint is usually the easy version of this problem. The same questions get harder, not easier, once records are sitting in a legacy document management system nobody's licensed anymore, a file share nobody quite owns, or specialist engineering software half the current team can't even log into. The gap doesn't belong to one platform. It belongs to whatever your firm has accumulated over the last fifteen, twenty, thirty years, wherever that history happens to live.
That's the case for intelligent archiving: not a bigger SharePoint, not a better folder structure, but a layer that sits across all of it - wherever "it" happens to be - and does four things none of the workarounds above can do at scale:
- Keeps records readable. Formats are migrated forward as the software that created them goes obsolete, so a 2011 file still opens in 2031, regardless of which system it started in.
- Proves integrity. Version history and fixity checks are captured automatically, so "this hasn't been altered since handover" is something you can demonstrate, not something you have to ask someone to remember.
- Makes retrieval systemic, not personal. Consistent metadata and search mean records stay findable no matter who's on the team, or which system they were originally filed in.
- Extends to whatever tools come next. Copilot today, something else in five years - content that's genuinely archived, not just stored, is content those tools can really use.
For firms already living in Microsoft 365, Preservica's Preserve365® applies this directly inside the environment you've already deployed, with no rip-and-replace. For records living elsewhere, in legacy platforms, file shares, or systems a recent acquisition brought with it, the same principles apply; the starting point just looks different.
We'll go deeper on what intelligent archiving looks like in practice in a future post. For now, if you want to talk through where your own records actually stand, contact us.
Have a records question of your own? Get in touch and we'll answer it in a future post.
