Suggestions for the optimal settings for the Fujitsu iX500 ScanSnap Scanner for scanning to Evernote?

[arose10 42]

I'm getting closer to beginning the substantial project of going paperless using my new ScanSnap scanner and Evernote.  Most of the documents I'll be scanning are two-sided black-and-white documents that were highlighted by me.  I have some questions on what the optimal ScanSnap settings might be.  Looking at ScanSnap Manager...

1) The "Scan to Evernote (Document)" profile seems like it would be the logical starting point, right?  What's the difference between this and the "Scan to Evernote (Note)" profile?  It seems like the former creates a PDF and the latter creates a JPG?  For purposes of Evernote, why would someone ever want to create a JPG instead of a PDF?  Are there any advantages?

2) Under "Application > Application Settings..." there is a checkbox for "Convert to Searchable PDF".  See...http://d.pr/i/bQumJ6.  I assume selecting this option has it so that Evernote performs OCR instead of ScanSnap's software?  Is such a setup preferable?  (I see a separate "Convert to Searchable PDF" option under "File option > Select OCR".  See...http://d.pr/i/ezk4HH.  I assume selecting this option has it so that ScanSnap performs OCR instead of Evernote?  Or am I just confused here?)

3) ScanSnap allows you to select an "Image saving folder" on your hard drive.  See...http://d.pr/i/E7M5tq.  If the document is being saved in Evernote, what reason is there to have an additional copy taking up space on your hard drive?  Is there any way to turn this off?

4) Anyone have any tips for best practices for "File name format"?  Here are the options that are presented...http://d.pr/i/bZ3n3T.

5) What "Image quality" level do folks recommend?  Or do folks just go with automatic resolution?  Here are the options that are presented...http://d.pr/i/DYiIxr.

6) Given that most of my documents are highlighted, I'm leaning towards setting the "Color mode" to "Color".  Will this be okay?  It's important to me that the highlighting I did in these documents, in yellow, is captured in these scans.

7) Is there any reason to select the "Reduce bleed-through", "Brightness", "Increase text contrast", or "Deskew by text on document" options?  See...http://d.pr/i/NnPIsE.

8) What do folks recommend for "Compression rate"?  Given the massive number of documents I'll be scanning, I want to be at least somewhat conscious of space, but I also don't want to have any issues with legibility.  See...http://d.pr/i/6YYLqr.

9) Any other key settings that I didn't mention in the above that are worth paying mind to, including any related settings on the Evernote side of things?  I'm all ears.

Thanks in advance for your guidance and insights!

[gazumped 11,637]

Hi.  Much of the information you request depends a great deal on how (and whether) you intend to access and use the documents you're processing.  The best advice in most cases is to try the various options and see which works most effectively for you.  I can tell you how my options are set (I have the older version of the iX,  the S1500),  but YMMV.

1. Yes, the difference is PDF vs JPG.  In general,  PDFs are better: they are multi-page,  annotatable,  and -in the most recent versions of Evernote- easy(ier) to manipulate.  Only main difference I can think of:  on a mobile device,  PDF files show as an icon.  JPG files show in full.  But JPGs are 1 file = 1 page;  multi-page documents require several files.
2. If you choose 'searchable PDF' your scan time is very slightly longer and you scan locally.  Otherwise Evernote will,  in its own time,  OCR your file and save a text version.  The OCR does not (AFAIK) export with the original scan.  Local scanning means you know it's done,  and have control over when.
3. If you are totally confident that a scan to Evernote will work,  scanning to a folder may be superfluous.  However you may wish to change your mind,  or you may have problems.  A contingency copy is a good backup.  You can always delete files in your folder from time to time. (But see also 'scan to folder' later...)
4. Personally I use yyyymmdd - document type - originator - keywords,  with the date set to the source document date.  If you're dealing with a lot of history,  setting the date might be unnecessary extra work.  It's occasionally useful (to me) in that case to have a 'scanned' date - again,  YMMV.
5. Automatic causes me no issues
6. If you want yellow,  go with color.
7. Unless you have problems with the quality of your scans,  it's not necessary to mess with the settings.
8. My compression is set to 'normal' (3)
9. Because I use my fancy titling scheme,  I use 'scan to folder' and save to my hard drive before using an Import Folder (Windows feature - Macs have scripting) to move the scan(s) into Evernote.  Scanning to folder allows me to edit the title,  edit the scan,  delete unnecessary pages and crop the view,  split and merge files...  It works for me.

But again - best advice:  start scanning and see what works best!  Good luck...

[arose10 42]

Thanks so much for these responses, @gazumped.  Super helpful.  A couple follow-up questions...

2) "The OCR does not (AFAIK) export with the original scan."  I'm not clear on what you meant by this.  Which OCR?  Export?

9) "Personally I use yyyymmdd - document type - originator - keywords,  with the date set to the source document date."  What's the advantage of this titling scheme for purposes of Evernote?  Does the file title carry over to Evernote in some manner?

Interested in suggestions that other folks have, as well!

[DTLow 5,736]

28 minutes ago, arose10 said:

2) "The OCR does not (AFAIK) export with the original scan."  I'm not clear on what you meant by this.  Which OCR?  Export?

If you OCR via the scanner, the pdf includes the converted text
If you use the Evernote OCR process, the pdf does not include the converted text.  Evernote's process is only used to build a search index.

A  practical example of this is you can not select text in an Evernote ocr'd pdf

>>Interested in suggestions that other folks have, as well!

As per @gazumped, I prefer the pdf format.  Be aware that Evernote's handwriting ocr process only runs on image files

I use scan to import folder on my Mac and delete the copy after I verify the note in Evernote.
The Mac scripting allows me to do some automatic processing on the note

Note titles are a personal choice.  You should consider the sort sequence the notes will be displayed

• I store my receipt scans in Evernote, and display in various views; by budget, by vendor, by date period.  I use yyyymmdd <vendor> <item> $nn.nn
• My project notes are titled <sort prefix> aaaaaaa
• My reference notes are titled yyyymmdd <reference type> aaaaaaa

[arose10 42]

Thanks, @DTLow!  I think I see what you're saying.  Basically, Evernote's OCR process has no impact on the original PDF file.  Based on what you're saying, e.g. "you can not select text in an Evernote ocr'd PDF," it sounds like having ScanSnap perform the OCR is definitely the way to go.

Thanks for the suggestions on note titles, but what I was getting at was the file title for the underlying PDF.  @gazumped was mentioning his process for this, and I was wondering whether the file title has any impact on the resulting note.

[DTLow 5,736]

24 minutes ago, arose10 said:

I was wondering whether the file title has any impact on the resulting note.

On my Mac import, the pdf filename becomes the note's title by default
For example  xxxScanDocument.pdf creates note titled xxxScanDocument

[CalS 5,277]

My method in summary

1. I created 3 custom profiles with my ScanSnap - Simplex, Duplex and Separate (each simplex page to its own PDF file). 
2. ScanSnap does the OCR for all three and places the PDF into my EN import folder so they are immediately imported to my EN Scan notebook. 
3. I re-title and add tags and move to the final notebook from the Scan notebook.  My Scan notebook is like an Inbox for scans. 
4. I only rename the PDF if I think I will ever share it with someone, or when I share it with them.  Otherwise, the names are the ScanSnap originals - 2017_10_19_17_45_42_002.pdf for example.  It doesn't impact search at all.
5. Everything is a color scan and I use the defaults for density, etc.

YMMV, doesn't always work for the OCD among us.  

[gazumped 11,637]

2 hours ago, arose10 said:

What's the advantage of this titling scheme for purposes of Evernote?  Does the file title carry over to Evernote in some manner?

Ah.  I see that others answered your query anyways - I tend to use an "intitle:<term>" search to find stuff,  with date qualifiers if there are lots of hits.  In my context I might be looking for a receipt from a particular supplier,  so "intitle:receipt intitle:<supplier>" filters out everything else,  and since my note titles include a date (courtesy of the PDF file name),  I can sort the hits by title to see them in date order.

For any PDF filename that I might use more than once,  I have a note that lists all the variations in alpha order so that I don't (forinstance) have to type "receipt - <suppliername> - printer consumables" more than once,  I just update the date (using a PhraseExpress keyboard shortcut) and then copy and paste the line into ScanSnap.  This ensures reasonable consistency in my naming strategy and makes it feasible to use long descriptive filenames / note titles.

[arose10 42]

Thanks everyone for their guidance!  I'm sure a couple more questions will arise once I commence this project, but this is super helpful in advance of getting started.