Scoping What a Security Incident Exposed
When a file share, mailbox, OneDrive or other repository is compromised, the first questions come fast. What was in there? How many files? Which ones held personal or other sensitive information, of what kinds, and whose was it? The answers shape the rest of the response, including whether anyone has to be told. Privacy laws can require organizations to report some breaches, so these answers are often needed quickly.
Shinydocs Pro has already indexed and analyzed your repositories. That index lets you work out what an affected location held without opening a single file: you can limit the Catalog to the files involved, see which of them were flagged for personal and sensitive information, look for copies of the same files elsewhere, and keep a report of what you found. This guide shows how, and says plainly where the index can’t answer for you.
Shinydocs Pro helps you scope an incident. It doesn’t decide whether you have to notify anyone or whom. Make that decision with your privacy officer or legal counsel.
Be ready before an incident
What you can find after an incident depends on what Shinydocs Pro analyzed before it. A few settings now save a lot of guesswork later:
- Turn on detection for every source. Personal and sensitive information detection isn’t on by default. Go to Settings > Analysis, choose Default on the General tab and turn on Personal & sensitive information, so every source checks its files. See Managing Sensitive Content to Reduce Reputational Risk.
- Turn on People, places & organizations as well, so you can see whose names appear in the files.
- Turn on File fingerprints, so copies of the same file can be matched across sources.
- Keep every source on a schedule. The index is only as current as the last update. A source updated daily gives you a record that’s at most a day old.
Act before the next update
If the repository was damaged, for example files were encrypted, renamed or deleted, the index may now be the best record of what it held. The next update changes that record to match the repository as it is now. Hold the source’s updates until you have recorded what you need:
- Go to Sources. If the source is updating, select Stop.
- In the source’s ⋮ menu, select Schedule, choose Off and save. The source then updates only when someone runs it.
- Don’t select Run now for that source until you’ve finished the steps in Record what you found.
Shinydocs Pro searches its own index, so you can do everything in this guide even if the repository is offline or unreadable.
Define the affected scope
Start in the Catalog and narrow it to the files the incident could have reached. Filters on different columns combine, and the item count next to the heading shows how many files match.
By source
Open the menu on the Source column heading, choose Filter by values and tick the affected source. If a whole repository was exposed, this may be all the scope you need.
By folder
When only part of a source was exposed, such as one shared folder, filter on the folder’s path:
- Select Columns, the icon at the right end of the column headings. Type
pathand tick it. - Open the menu on the path column heading and choose Filter by condition > Starts with….
- Type the folder’s path, ending with a backslash, for example
\\fs01\shares\HR\. Capitals don’t matter. - Select Apply.
The filter includes every subfolder. The backslash at the end keeps out folders whose names only begin the same way, such as \\fs01\shares\HR-Archive. The quickest way to get a path exactly right is to copy it from a file in that folder: the Catalog shows each file’s folder under its name. Example matches can show No hits found for a path with backslashes. Select Apply anyway and check the item count.
Unknown Attachment
Paths look different for each kind of source:
| Source | What the path starts with | Example |
|---|---|---|
| File share | The share or drive path | \\fs01\shares\HR\ or D:\Shares\HR\ |
| SharePoint Online | The site’s path, without the tenant’s address | /sites/HR/ |
| OneDrive | The web address of the user’s OneDrive | https://contoso-my.sharepoint.com/personal/sam_rivera_contoso_com/ |
| Exchange Online | The mailbox address, then the folder, in lower case | sam.rivera@contoso.com\ |
Different sources can have the same path, for example two SharePoint sources with a /sites/HR/ site, so filter by source as well.
You can also write the folder as an advanced query. Select the </> icon at the start of the search box and type the path with every backslash doubled and * at the end:
path.keyword:\\\\fs01\\shares\\HR\\*
In a query, capitals must match, a space in a folder name needs a backslash in front of it (Human\ Resources), and a drive letter’s colon does too (D\:\\Shares\\HR\\*). See Catalog - Advanced Search.
By date
The Created column, and the Modified column you can add with Columns, filter by date: open the heading’s menu and choose Filter by condition > Is before…, Is after… or Between…. For example, if a share link was removed on a known date, filter Created with Is before… that date to leave out files added after the link stopped working. These are the dates the repository reported at the last update.
Keep the scope
When the filters describe the affected files, select the save icon next to Saved filters and give them a name. You can then reopen exactly the same scope later, from Saved filters.
Check how current the index is
The Catalog shows each file as it was at the source’s last update. Files added, changed or deleted after that aren’t reflected, so check when that was:
- On the Sources page, the Updated column shows how long ago each source last updated.
- On the source’s own page, Source details shows Last analyzed and Next update. Expand the crawl and open its History tab to see each past update. Each analysis card shows when it last finished.
Unknown Attachment
An update finds what changed in the repository since the last one. Files that are no longer there, including files that were renamed, leave the Catalog straight away and are removed from the index about a week later. Files that changed are read again, and their analysis is worked out again from what they hold now. That’s why Act before the next update matters when files were encrypted, renamed or deleted: the record of what they held before is in the index only until the next update. The same goes for Run now > Full crawl and reanalyze, which reads every file again. See Monitoring Sources.
Find sensitive content in the scope
With the scope filtered, open the menu on the PII column heading:
- Filter by condition > Is not empty shows every file in the scope with suspected personal or sensitive information.
- Filter by values lets you tick particular kinds, such as social insurance or social security numbers. The kinds are grouped by sensitivity: High, Medium and Low.
The PII column shows the kinds found in each file. The item count is the number of files in the scope that hold them.
Unknown Attachment
For counts across sources, the Personally identifiable information card on the Overview shows files by sensitivity By source or By policy, and Insights shows the most common kinds and how the counts changed over time. See Insights and Content Score.
No PII found isn’t the same as no PII. A source where detection never ran also shows no suspected PII, on the Overview card too. Before you rely on an empty result, open the source’s page and check that Personal & sensitive information is on and Up to date.
If detection wasn’t on for the source
Detection doesn’t open the files. It checks the text Shinydocs Pro kept when it read each file, the step shown as Reading file contents on the source’s page. If that step had run before the incident, the text is already in the index, and turning on Personal & sensitive information can still check it. Files that weren’t read, shown as excluded or gave no text, can’t be checked. See Analysis Settings - General and Extraction for which files are read.
Detection runs when the source updates, and every update starts by checking the repository for changes:
- If the files are unchanged, for example a link was shared too widely or an account was misused, turn on Personal & sensitive information for the source and select Run now.
- If files were encrypted, renamed or deleted, the update also changes the record, as Check how current the index is explains. Finish Record what you found first, then contact Shinydocs Support before you run the source.
Detection looks for patterns, such as the format of a card or ID number. It can flag numbers that only look like one, and it misses information written in a form its rules don’t expect. Review the files before you report on them. The rules are listed in Default PII Rules for Shinydocs Pro, and Detecting Potential PII and PSI using Shinydocs Pro explains how detection works.
Work out whose information it is
Knowing that 15 files hold social insurance numbers is a start. Knowing whose numbers they are is what lets you contact the right people.
- The values that matched. Each file’s
pii_matchesfield holds the text each rule matched, such as the number itself. It goes into a report when you export all fields (see Record what you found), and you can compare it with your own records. - Names found in the files. If People, places & organizations ran, add the entities-person column with Columns to see the names found in each file, and entities-location for places. Both go into reports.
- One person’s files. Type a name in quotes in the Catalog search box, for example
"Sam Rivera". The search box looks in file names, paths and properties and in the names People, places & organizations found. It doesn’t search the text of files, so it can’t find a name that analysis didn’t pick out.
Names are picked out of the text automatically, so some will be missed and some words will be taken for names. Treat them as leads to check, not as a list of affected people.
Copies in other places
The same files may also live somewhere that wasn’t affected, or somewhere that was. With File fingerprints on, Shinydocs Pro matches identical files across your sources:
- The Duplicates column marks files that have an identical copy: Has duplicate on the oldest copy, Duplicate on the others.
- Add the Copies in column with Columns to see which sources hold another copy of each file.
- Select a file’s Duplicate or Has duplicate badge to show every file with the same fingerprint. Remove the source and folder filters to see copies outside the affected scope.
Record what you found
A report lists every file in the scope, so you have a dated record of what the index held, even after the repository changes. Anyone who can use Control Center can create one.
- Show the columns you want in the report, for example path, Modified, PII, entities-person and Copies in.
- Select Action at the top right. With no files ticked, the action covers every file that matches your filters.
- Choose Report and open Additional options.
- Give the report a name, such as the incident’s reference.
- Choose what goes in it. With Export all fields ticked, the report has every field Shinydocs Pro holds for each file except its text, including
pii_matches. Clear it to export only the columns you can see: the PII column gives the kinds found, not the matched text. - Select Next and confirm.
Unknown Attachment
The report is a ZIP of CSV files. Report row limit sets how many rows go in each CSV file, and a larger report is split across several files. Download it from View activity > Remediation once it’s complete. See Getting to Know Content Catalog & Creating Reports in Shinydocs Pro.
A report withpii_matchesholds the personal information itself. Store and share it as carefully as the files it lists, and only with the people handling the incident.
Other records worth keeping:
- Settings > Audit records each report: who started it and when, the number of items, the filters used and the report’s name. Only Control Center administrators can open Settings. See Audit, Logs and Resources.
- Your saved filter, so you can rerun the same scope, for example after detection has run.
- A printout of Insights: include only the affected sources and select Print for a dated summary of their counts, including PII by kind.
What Shinydocs Pro doesn’t tell you
Shinydocs Pro knows what your repositories held. It doesn’t know what happened to them:
- No access history. It doesn’t record who opened, copied or downloaded a file, or when. It can’t show what an attacker actually took. Use the repository’s own audit logs and your security tools for that.
- No detection of the incident. It doesn’t detect intrusions, ransomware or data leaving your network.
- Only what was indexed. Files added after the last update, in locations a source doesn’t cover, or left out by a source’s settings aren’t in the Catalog.
- Only what could be read. Detection and name extraction work on the text Shinydocs Pro kept. Files it couldn’t read, such as encrypted or password-protected files, aren’t checked.
- Suspected, not confirmed. PII results and names are found automatically. Expect some false matches and some misses.
- Files, not people. Counts are of files. One file can hold many people’s information, and one person can appear in many files.
- No notification workflow. It doesn’t decide whether to notify, or track who was notified.
Let’s review
- Before an incident, turn on detection, People, places & organizations and File fingerprints for every source, and keep sources on a schedule.
- After an incident, hold the affected source’s updates until you’ve recorded what the index holds.
- Limit the Catalog by source, folder path and date, and save the filters.
- Check when each source last updated, and whether detection ran on it.
- Filter on PII to find the files with sensitive content and the kinds they hold, then use the matched values and names to work out whose information it is.
- Look for copies of the files in other sources.
- Create a report, keep it secure, and note what Shinydocs Pro can’t tell you.


