How Forage Fits Together

Purpose: see the whole chain, from choosing which sites to read to the record of what happened, and how each step feeds the next, so that every guide that follows makes sense in its place.

1. You choose what Forage reads. On Sites you include the SharePoint sites to read and put them in order (choose which sites Forage reads). On Settings you choose whether Forage reads SharePoint, the archive or both, and which file types it reads (change what Forage reads). A new deployment reads nothing until sites are included.
2. Forage lists the files. For every site you include, Forage builds its list of files in SharePoint Online and in Squirrel's archive from the record Squirrel already keeps. Every file on the list can be found in Search by its name, whether or not its contents can be read (search).
3. Forage reads them. It reads the words in each file it can open, including the words on scanned pages and in images, which it reads with optical character recognition (OCR). It records what happened to each file: read in full, partly read, or not read, and why. The Exceptions page groups the files that could not be read by cause (understand exceptions and unread files), and the File details window explains one file (read a file's details).
4. Forage keeps its picture current. It follows the changes Squirrel records, and reads a document again only when its content has changed. The Dashboard and Coverage show how far reading has got (follow Forage's progress).
5. Reading makes files searchable. The text goes into a search index kept inside your own deployment. Search finds live and archived files together, by their words, their names or their folders. The steps that follow use the text of the files Forage has read.
6. Labels and rules, where content classification is enabled. A label says what a file contains, and a rule tells Forage when to give a file that label. Once a rule is switched on, Forage checks the files it has read against it, and the results appear on the Classification pages and in Search (labels and rules in plain words, create and edit labels, write, test and switch on a content rule, see what a label or rule found).
7. Suggested labels, where content classification is enabled. The AI assistant, which runs inside your own deployment, looks through the files Forage has read in the sites you choose and suggests labels for the kinds of record it finds. You accept or reject each suggestion. Accepting one creates a label and a draft rule, and that rule labels nothing until you test it and switch it on (get label suggestions). The AI assistant never decides which label a file gets.
8. Files to Protect, where content classification is enabled. You choose one of your own Microsoft 365 sensitivity labels, and the AI assistant lists the files Forage has read that look like the ones carrying it but carry no label, for a person to review. Forage changes no label (find files that may need a sensitivity label).
9. Ask AI a Question. A Full Administrator asks a question in plain English, and the AI assistant answers it from the files Forage has read, with each statement giving the number of the file it came from (ask AI a question about your files). The more Forage has read, the more it has to answer from.
10. Forage for Copilot. Microsoft Copilot can search Forage and read the text of the files it holds, archived files included, with each member of staff shown only what they can open in SharePoint (Forage for Copilot).
11. Reports. The Reports page turns what Forage has listed, read and found into reports for a board or an auditor (produce a report), and Settings can email the reports on a schedule (email reports).
12. The record. The Log keeps who changed what, who opened a file, who exported a report and who searched (see who did what). The Activity page shows what has happened recently (see what Forage has been doing).

Good to know: a file that could not be read is still listed, and can be found by its name.

Questions go to sales@smikar.com.