I do not know the format for the records from 1947. Earlier marriages are recorded in register books. Marriages up to 1920 are online (you can view 1916-1920 at AmericanAncestors.org).
As a test case, I chose a marriage that was early enough that I could look at the image at FamilySearch. The register snippet below is taken from Film 2315512, DGS 004329363 image 204:

As my test case, I used the 2nd marriage on the page, for Blanche Turcotte. There is a glitch at FamilySearch which caused her name to be repeated in the spouse's name line on the 'index card' page, but his name appears in the search results:
spouse:
J Alberic Mcclure
FamilySearch's information about this record says:
Record Collection:
Massachusetts Marriages, 1841-1915
Document Information: Line Number 344
Reference ID p 122 no 344
GS Film Number 002315512
Digital Folder Number 004329363
Image Number 00204
To get the volume number, we have to retrieve it from the film note. Holyoke is in Hampden county, so this is Volume 595:
[1910]: vol. 595 - Franklin to Middlesex counties
This 1910 Register page has 18 entries on the page (entries 343 through 360, the stamped numbers in the left-most column).
Ancestry's image of the 5-year index book looks like this:

Her entry reads (supplying the surname from the first entry -- I generally put information taken from elsewhere on the page in brackets):
Name Town Year Volume Page
[Turcotte] Blanche Holyoke 1910 595 122
On Ancestry's 'Record page' they show:
Index Volume Number: 65
Reference Number: F63.M36 v.65
By looking at the register and comparing the information, it becomes clear that:
- The Volume number in the index matches the Volume numbers in the FamilySearch Film Notes.
The Page number in the index is the stamped number in the top left corner of the page, not the handwritten page number in the center where it says "Page [blank]"
The entry number in the left column is not indexed in these printed volumes.
- The Ancestry Index Number and Volume Reference number is for the printed 5-year index book, not the original register.
Ancestry offers to let us search. Can we find J Alberic Mcclure?
- Search for Marriages in 1910, in Holyoke (exact search): 1,474 results
- Narrow search to Volume 595 (exact): 97 results
- Narrow search to Page 122 (exact): three (!) results:
Results 1–3 of 3
Name Marriage Date Name Marriage City
J McClure Vlberic 1910 Holyoke
Annie 1910 Holyoke
'alter Cox 1910 Holyoke
1–3 of 3
I would expect 36 results (18 couples x 2) but here we only see three people, none of whom are the person whose entry we used to start the search!
So -- while this may work in principle, my experience with this database is that you can't depend on it in practice. As we can see from our test case, the OCR from this collection is horrible.
My experience is that the names are often disassociated from their index information, so that Ancestry has the name pointing to the Volume and Page of the person above or below them on the page. You can't even depend on there being one entry in the database for one line in the 5-year index book.
Re-running the search again with 1910 (exact) and leaving Volume 595 and page 122 as exact, but taking out the town name, resulted in the same three names.
What you can do is:
- Find the name of a groom by browsing
- Find the name of a bride by browsing
- Compare the two entries to see if they are on the same Volume and Page.
Having done so, you still need to confirm the information the old-fashioned way, by ordering a 'certificate' (but what you really want is an image copy -- otherwise you'll just get a clerk's transcription of what's in the register book) and/or finding out information about the marriage by using other sources.
Very important reminder: In Massachusetts, there may be registrations for the same marriage in up to three different towns -- one for the town in which the marriage took place, and one each for the bride and groom's town of residence. That is, if the wedding was a destination wedding and the bride and groom lived in two different towns, you may find the same wedding in THREE different registers.
Be very careful, especially when using digital images where the header "Marriages Registered in the [town name]" has been cut off. Not all registers list the place of marriage. Use the name of the person by whom married as a clue to further records.
It is not uncommon in my experience to find marriages registered in both Massachusetts and Vermont or New York if the couple crossed the border to get married. Similarly, you may find deaths and births in the Vermont Vital Records indexes also registered in Massachusetts when the subject is a Massachusetts resident.
Given the state of Ancestry's computer index, it's entirely possible that you might find two references from these index volumes that don't match which actually refer to the same marriage -- if by mischance, you only find the groom's index reference from his town of residence and the bride's index reference from hers.
For these index books, it's especially important to keep in mind J. Mark Lowe's advice to "slap your hand" and browse the books instead of doing a search.