One row per book: date, title, who read it, and whether it landed. Three sheets, 60 rows, sized for a bedtime routine rather than a school assignment.
No email, no account. The PDF is for printing and taping inside a cupboard door; the CSV opens in Excel, Numbers or Google Sheets.
They get confused because both involve writing down a title, and that is where the similarity ends. A reading log answers “what have we read?”, which is the same book logged again every time you reread The Gruffalo at bedtime. A bookshelf inventory answers “what do we own?”, which is each book logged exactly once, the moment it arrives.
Keep them separate. A combined sheet ends up bad at both jobs: too repetitive to use as an inventory, too sparse to use as a reading history. If you also want the ownership side of things, there is a separate inventory template built for exactly that.
A year of a filled-in reading log becomes a genuinely nice thing to have: it is a record of what your child liked at each age, which age-appropriate authors landed, and which books got requested on repeat. It also tends to surface something useful for gift-buying season: a fast skim shows which authors or series to hunt for more of.
What it will not do is stop a duplicate gift, because it tracks what has been read, not what is currently on the shelf, and it is not something a grandparent standing in a bookshop can check. That is a different problem, and it is what Shelfmark and its share link are for.