I put off setting a retention policy for two years because I assumed the hard part was money. Somewhere in my head was a picture of a bill that grew every month until it became a real line item, and I did not want to look at it. Then I finally sat down and worked out my inspection video storage cost, and found out I had been avoiding a problem the size of a phone bill.
The money is not the reason to have a policy. But you should still run the number, because knowing how small it is tells you which arguments you are allowed to make.
Run the arithmetic for your own shop
Start with one inspection. A roof walk that is genuinely useful runs somewhere around ten to fifteen minutes of recording. At ordinary phone streaming quality, call it fifteen to twenty megabytes per minute. That puts a single walk around two hundred to two hundred fifty megabytes. Round it to a quarter of a gigabyte and move on.
Now scale it. Eight inspections a week, fifty weeks, is four hundred jobs a year. Four hundred quarter-gigabyte recordings is roughly a hundred gigabytes a year.
Commodity cloud storage runs in the neighborhood of a couple of cents per gigabyte per month. A hundred gigabytes is therefore about two dollars a month in the first year. It stacks, because retention means you are holding last year's too. Year three you are sitting on three hundred gigabytes, call it seven dollars a month. Year ten, if you never delete anything, you are at a terabyte, which is roughly twenty-five dollars a month.
That is the whole frightening number. Ten years of never deleting anything costs less than the fuel in one return trip across town.
Run it with your own volume. A crew doing thirty inspections a week gets to that terabyte a lot faster, and if you record at higher quality or leave the camera running during ladder setup you can double the file sizes without noticing. But the shape holds. For most residential shops, inspection video storage cost is not a budget conversation.
The costs that never show up on the invoice
This is where it actually gets expensive.
Retrieval and labor. Finding one specific walk inside twelve hundred recordings is a person's afternoon if the archive is not organized by address and date. At a loaded office rate, two of those afternoons a year cost more than the entire decade of storage. The pile tax is real and it is paid in hours, not dollars per gigabyte.
Duplication. Most shops are not storing one copy. They are storing the original on a phone, a copy in the job software, a copy in a cloud drive, and a backup of the cloud drive. Four copies of a terabyte is four terabytes, and only one of them is under any kind of policy.
Migration. The day you change software, everything you kept has to move or be abandoned. A hundred gigabytes moves over a weekend. A terabyte spread across three systems and a departed employee's account is a project with a budget, and I have watched shops simply orphan the data instead, which is deletion by neglect wearing a nicer shirt.
Exposure. This is the one that matters. Every recording is footage of somebody's property, and a growing archive is a growing thing to protect, for as long as you hold it. That obligation does not get cheaper as the storage does. It is why our approach to storage and access is written out on the security page, and why I think about the size of the archive as a risk surface before I think about it as a bill.
Keeping everything forever is a decision, not the absence of one
Because storage is cheap, the default in most shops is to keep everything until something breaks. That feels like neutrality. It is not.
You have decided to hold every frame you have ever recorded, of every house you have ever been on, indefinitely, without writing down why or telling anyone. If someone later asks what your retention practice is, the honest answer is that you do not have one, and the pile is whatever survived phones dying and employees leaving. That is not neutral either. It is arbitrary, which is worse.
And it is not my place, or your software vendor's, to tell you that a shorter window is fine. That is a question for a lawyer licensed where you work, because retention obligations vary enormously by state and by contract. I am a contractor. This is not legal advice. The point here is narrower: do not let the smallness of the storage bill make the decision for you, because the bill is the least important input.
What actually brings inspection video storage cost down
If you do want a smaller archive, and there are good reasons to want one that have nothing to do with money, four things move it.
Record the walk, not the setup. Half the footage in my early archive was a ladder and a lot of breathing. Start the session when you start narrating.
Stop keeping four copies. Pick one system of record and let the others be temporary. This does more than any compression setting.
Set a window on purpose, with counsel, and let the software enforce it rather than a person. Ours is configurable per company for exactly this reason, media is deleted after the configured window plus a grace period, and a record that the inspection happened and that its media was purged is kept afterward, so the archive shrinks without becoming mysterious.
Control who can reach it. A smaller number of people with access, and links that stop working on their own, means the archive gets less dangerous as it gets older rather than more. That is the secure sharing side of the same problem.
The honest conclusion is that inspection video storage cost is not why you should set a retention policy. It just means cost is not a reason to avoid setting one. The reasons are the ones in the retention overview: knowing what you have, being able to find it, being able to explain your practice, and not discovering your policy by accident on a bad day.
Set the window with a lawyer, then set it in the tool once. Two dollars a month was never the problem.
InspectStream records what an inspector observed. It is not a public adjusting service and does not prepare, negotiate, or advise on insurance claims.