I've been researching my family history since the year 2000, not as long as many, but longer than some. When I started, the internet was a fairly minor resource for genealogy, and I took regular trips up to the Family Records Centre in Islington and the National Archives at Kew in order to make my earliest discoveries. Over the years I've made my fair share of mistakes and errors (and no doubt I'll keep doing so), but the one good thing about mistakes is that you - hopefully - learn from them!
Here are a few hints and tips based on things I've learned along the way.
It might seem like extra work, but it's so important to record your sources. One day you might look at a record and wonder how on earth you came up with that birth date. If you know where something came from, you know how reliable it might be, and the source could be useful to return to when new information comes your way.
Remember - one piece of information will lead to another. Make sure the foundations are good - strengthened with good data - then you can build a solid case rather than a precarious house of cards.
Sometimes you have a hunch that something is right and you follow that line of thought, looking for information that confirms the idea. This is where you should be careful - and while it might hurt to do it, you should do as much as you can to falsify the theory. This is what a good scientist does - they don't try and prove a theory right, they try everything they can to disprove it - if the concept survives, the likelihood is that it's right.
Remember - it's easy to fall into the trap of confirmation bias, gathering only data that supports a theory. Research other possibilities, eliminate other contenders. Truth is always best.
Sometimes a record you access online will have multiple pages - for instance a military record - and it's the convention that the link takes you to the first page, then you click forward to read the entire document. But it's also worth clicking backwards from that first page - and, although it's rare, you may find more pages related to the person you're researching.
This has been successful for me on three or four occasions. With one, I clicked backwards from a simple Banns announcement and discovered a three-page letter written by my 5x-great grandmother in 1833 - it mentioned her husband's military service and death, her son's death abroad, and some details about a brother and sister.
This goes for census returns too - when you've found the family you're looking for, take a peek at the neighbours, extended family can often be found living close by.
Remember - when searching online, you're not searching through original records, you're searching indexes made by people who don't necessarily understand the document they're transcribing.
Ancestry Trees (public or private) and other online trees can be a great resource, but take any information found there with a large grain of salt. While multiple trees might seem to validate a particular family connection, many of them have entire branches merely imported from other Ancestry trees, so a single mistake ends up being replicated again and again, giving it a false credibility. By all means use them, but do your own research to confirm the information.
Although not common, some people aren't genealogists, they're collectors of data - it doesn't really matter if the data is true, as long as it fits, in it goes!
Remember - confirm with primary sources, confirm with multiple data points, confirm by eliminating other possibilities.
Often our earliest research is the most basic and quickly becomes consolidated as 'ground floor facts'. Sometimes they might be letters from relatives answering your first basic enquiries, or an old family story that seemed to get disproved somewhere along the line. It's always worth looking at these again - a fuller family tree and new avenues of research that are now available can change the context of old information, or make sense of a new mystery.
Remember - old facts might take on new meaning when the light is shone on them again. Keep everything!
Don't get stuck on names. Even if you've done just a little amount of family history research then you know you have to search different spellings of names - for instance my Ewings are sometimes recorded as Ewen, and my Hodgkins can be Hodgkinson, Hoskin or Hodgkiss, etc. Some search forms allow you to use wildcards where you can replace a letter with an asterisk (*). When searching for the name Levell I often use L*v*l, which catches all the Levils, Levills, Lovells and Leavels, etc. The earlier back in time you are, the more likely you are to have inconsistency in spellings.
This goes for the names of things like army regiments too. If you discover an ancestor was in the 72nd Foot, find out what else the regiment was called - most changed their name after 1881 (the 72nd became the 1st Battalion Seaforth Highlanders). Newspapers may use different terms, 'Foot' or 'Regiment', and if you search "72nd" you may not find a lot, but try "72d" and you'll be on the right path. Some regiments went by several different names - so when I was researching the Dumfries Light Dragoons I also searched for The Dumfriesshire Dragoons, the Dumfriesshire Fencibles and the Dumfries Fencilble Cavalry, among others variations.
Remember - a hard-to-find ancestor can sometimes be found by looking around them rather than for them - search for their children, their friends, their address, their occupation ... footprints come in all shapes and sizes.
You may not be ready to write your grand family history, but it can still be helpful to write out what you know in narrative form - it gets you thinking about the story of your ancestors and helps to highlight gaps. Often we focus so much on isolated facts that's it's only when we bring them together to fulfil a narrative that we uncover the real story. To help in the telling you'll want to add context, either historical, geographical, or family, all of which helps to complete the picture and humanise the tale of your family history.
Remember - your narrative can focus on an individual, a family group, or even a geographical location ... write purely for yourself and see where it takes you.
If you can't quite get a definite answer to a genealogical conumdrum, it might be best to mark it down as a 'don't know' rather than put in the nearest fit. Certainly, record the possibilities - but seperately, not in the official fields (name, date of birth or death, etc.). Once you fill in that data field it can become accepted fact, and if you then base further research on that 'fact', and it's wrong, you've sent yourself down the wrong alleyway. Look at it again a year or two down the line when more new records have been made available - the answer might come. Or maybe the answer was never recorded, or is permanently lost - you might have to make do with a question mark.
Remember - sometimes not knowing the answer can be just as interesting, record all the possible theories - that's a story too.