To a large measure, this problem can be overcome by properly citing your sources. So does citing your sources mean following some predetermined pattern of citation? I guess my answer would be only if you are planning on publishing the results of your research in some formal genealogical journal or other such publication. Ultimately, what is important is whether or not someone who looks at your citations, which includes you at a later date, can decipher where the materials came from.
Let me give an example of what I am talking about. Here is a perfectly proper citation:
Tanner, Maurice, and George C. Tanner. 1923. Descendants of John Tanner; born August 15, 1778, at Hopkintown, R.I., died April 15, 1850, at South Cottonwood, Salt Lake County, Utah.
This citation conforms to the 15th Edition of the Chicago Manual of Style. See University of Chicago. 2003. The Chicago manual of style. Chicago, Ill: University of Chicago Press. But guess what? I really need a lot more information to find the book. First of all, how about recording where I found the book? Such as which library I was in when I made the citation. For convenience, I may also wish to note the page of the information I used. I also suggest I might want to note if the book is available online. I may have looked at a paper copy of the book in the library where I found the book, but if I look online, I might find the book is readily available in digital format. In fact, the above Tanner genealogy is available both in the HathiTrust.org Digital Library and in HeritageQuestOnline.com.
Keeping a list of sources searched can become really burdensome. If you are like me, I might initially search eight or ten different websites in a matter of a few minutes. I do not become overly concerned about recording these initial searches because I will probably come back to the same sites more than once anyway. But once I begin to address a specific research issue, such as finding a specific record or date, I have a greater incentive to avoid duplicating my searches. This is particularly true if I am visiting a library. Of course, if I do find any information about an individual, either online or in a paper document, I record the information immediately, usually by making a digital copy of the document or linking to the website with a source in my genealogical database.
Having just recently finished another series of presentations to several genealogical societies, I fully realize that recording the information gathered in a genealogical database is far easier said than done. This is where I find the advantage of carrying a copy of my database with me, either on my laptop or smartphone, where ever I go. So I either end up with a digital image of the document and the source where the document came from or a notation on my computer or smartphone in Dropbox.com or Evernote.com about the item.
What I have found is that if I do not transfer that information to a more permanent location in conjunction with my genealogy program, I will often forget the origin of the information and lose the notes I have made. So there is a fairly large measure of discipline involved in successfully creating a research trail that you can follow back to the source.