Protecting Data Throughout Its Lifecycle
Identify whether data is at rest, in transit or in use at each lifecycle stage and choose the protective control that fits, through to secure deletion.
One cash box, four different dangers
At closing time at the Kothrud store, the day's cash lives through four moments. Farida Shaikh counts it on the counter, with the shutter half down and customers still leaving. It goes into the cash box in the store safe overnight. Next morning it rides in a locked van to the bank. Old bill books, years later, go into the shredder. The money is the same in all four moments, but the danger is different each time, and so is the protection: a safe for storage, a locked van for the journey, a quiet counter and few eyes while counting, a shredder at the end. Data moves through the same kinds of moments, and protecting it well means knowing which moment it is in.
The same money, four moments, four different protections. Data at rest, in transit, in use and at the end of its life needs the same kind of thinking.
The data lifecycle
Every piece of data has a life. The data lifecycle is the series of stages it passes through: it is collected or created, stored, used, shared or moved, sometimes archived (kept but rarely touched), and finally destroyed. A prescription in the Tulsi Health app follows this path exactly. Mrs. Lata Gokhale photographs it on her phone, the app sends it to Tulsi Health's servers, it is stored in the database, a pharmacist opens it to prepare the order, it is kept for as long as the law and the business require, and then it should be deleted.
Each stage puts the data in a different place, in front of different people. A protection that works at one stage may do nothing at the next.
Seen through the cash box, these stages collapse into three states plus an ending. Data is at rest when it is stored and waiting, in transit when it is moving from one place to another, and in use when it is open and being read or changed. The ending, disposal, is its own problem. Each state has its own typical dangers and its own appropriate protective controls, meaning the safeguards that actually fit that moment.
At rest: the box in the safe
Data at rest sits on a storage device: the app's database on a cloud server, MedLedger's files on a counter laptop's disk, the nightly backup, a pen drive in a drawer. The main dangers are a device being lost or stolen, someone copying files they should not reach, and old copies nobody remembers.
The central control here is encryption: scrambling data with a secret value called a key, so that without the key it is unreadable gibberish. When a counter laptop's whole disk is encrypted, a thief who walks off with it gets a box of noise; when it is not, they can pull the disk out and read every file without ever knowing the login password. The same goes for the database and the backups. Encryption is only as good as the care of its key, so the key is kept separate from the data, just as a shop does not tape the safe's key to the safe door. Alongside encryption, access control limits which accounts can open each folder or table at all.
In transit: the locked van
Data is in transit whenever it travels across a network, the connections that link phones, laptops and servers (a server is simply a computer that holds data or runs a service for others). When Mrs. Gokhale's prescription leaves her phone, it passes through her home Wi-Fi, her internet provider and several other networks before reaching Tulsi Health. Anyone positioned along that path, such as someone on the same café Wi-Fi, could try to read or alter it.
The control is to encrypt the data for the journey. Websites and apps do this with HTTPS, the secure version of the language browsers and servers use to talk; the padlock in the browser's address bar means the connection is encrypted and the site has proved its name. Email between Tulsi Health staff should travel over encrypted connections too. In transit also has a human side: an encrypted email to the wrong person is still a leak, so checking the recipient is as much a control as the padlock.
In use: open on the counter
Data in use is data someone has opened to work with, like a prescription on Farida's screen. To be read, it must be decrypted, so the encryption that protected it at rest and in transit no longer helps. This is the cash spread out on the counter: the moment it is most useful is the moment it is most exposed.
Controls for data in use are about people and screens. Give access only to those who need it, so only pharmacists can open prescriptions. Turn screens away from the customer side, or fit a privacy filter that makes them unreadable from an angle. Lock the screen when stepping away. Show only what the task needs, such as the last four digits of a phone number on a delivery slip, a technique called masking. And avoid making new copies: a screenshot or a printout of an open record turns data in use into a fresh piece of data at rest that nobody is protecting.
Sort the controls below by the state of data they protect. Some are technical, some are habits; both count.
The end of the story: retention and secure deletion
Data should not live forever. A retention period is how long a type of data is kept before it is destroyed, set by what the law requires and what the business genuinely needs. Pharmacies must keep certain prescription records for a period the law sets; old app logs or marketing lists may be needed for a few months at most. Keeping data beyond its period adds nothing but risk: every extra year of records is extra harm if it is ever exposed.
Destroying data properly is harder than it sounds. Pressing Delete usually just moves a file to the bin, and emptying the bin often only marks the space as free; the contents stay on the disk until something new overwrites them, and recovery tools can bring them back. Secure deletion means making the data truly unrecoverable: wiping a disk with a tool that overwrites it, destroying the encryption key of an encrypted disk so the remaining scrambled data can never be read, or physically shredding paper and destroying old disks. It also means finding every copy, including backups, email attachments and that export on a developer's laptop.
Deleting removes the data from view; secure deletion removes it from existence. Old laptops sold or donated with an ordinary delete are a common source of leaks.
Follow one prescription through its whole life below. At each stage, choose the control that fits that moment. A wrong first choice counts as an incident, so think about where the data is and who could reach it.