This could get kind of long....I'll try and be concise:
First, you must measure the losses of your mash-tun, and your brew-kettle.
Measuring your kettle is easy. Fill it with water, and drain it out however you normally do so. Then, tip it over and pour out anything that's left and measure it. This is your trub-loss.
Measuring your mash-tun deadspace is trickier, because unless you have a false-bottom you don't have a "true" deadspace. You still have a portion of the mash-tun (maybe a quart of VOLUME) where you can't pickup wort, but the space is filled with grain. So, its only the space between the grain that is truly dead wort volume...which might only be half of the total volume. I measured mine after I finished a batch of beer by simply closing the lid of my cooler and turning it upside down into my brew kettle. The excess wort drained out into my kettle where I could then measure it. I have a round stainless false bottom in a 50 quart cooler, and my excess wort was something like 2 cups.
Then you put those numbers into your equipment profile.
Next, when you get a recipe from someone else, you need to know what efficiency it was prepared for. 70% is a common number, but there's no guarantee. You just have to ask. Then you have to scale the recipe to YOUR efficiency. BS2 can do this, but I'm not at home and I forget the clickety-clickety off the top of my head. Something like "Scale recipe" and then select "by efficiency", enter from and to data, click OK.
Next you need to choose a sparging method: batch or fly. I do batch, and I recommend that new AG brewers do batch, also. Its more predictable and less susceptible to error. It will let you focus on the rest of the AG processes, which are more important than sparging method. You can always go back and switch to fly later.
Once you pick a method, work on doing it the same every single time. Use the same water/grist ratio every time. for batch sparging use the same number of batches every time. Stir each batch the same....let it sit for the same amount of time between runoffs, vorlauf the same. For fly sparging, use the same run-off rate every time.
Since most people want to boil for a fixed about of time, target OG has a big impact on actual efficiency. This is because you will stop sparging once you hit your target pre-boil volume. With a larger grain bill, you will be leaving sugars behind in the mash-tun. That's just the way it is. To get around that, you would have to sparge until all the sugar was removed from the MLT, and then boil a lot longer for a high-gravity beer because you'd have 50% more water in the kettle. So, expect to see a 5% loss for each 10 points of increase in OG. For me I see the following:
Post-Boil-OG Efficiency
1.030 90%
1.040 85%
1.050 80%
1.060 75%
1.070 70%
and so on....So, when learning to AG, and dial in a system (and your technique)...its best to stick with a narrow OG range. It doesn't matter where you start...just stay there until you can repeat the results. 1.050 is a good starting point because there is a plethora of recipes in that range.
The biggest drivers in AG efficiency are grain crush, dough-balls, and sparging technique. Since you are new to AG, I assume you have your grain milled at the LHBS. If not, then DO THAT. Their crush may not be ideal, but it is probably consistent.
Dough-balls are caused by adding water to grain, and pour mixing (often caused by a small MLT). Its much easier to get a good dough-in by putting the strike water in the MLT first and then adding the grain slowly while stirring constantly. As the mash starts to get thick, slow down the rate of new grain. Make sure its all mixed before adding more. Many new AG brewers start with a small MLT, usually the same size as their batch (5g MLT for a 5g batch, etc). This means that a 1.050 beer fills 75% or so of the MLT. Any vigorous stiring is going to cause splashing and overflowing. So, because they are being careful not to splash they don't mix it well enough.
Ideally your MLT should be twice the size of your batch size: 10g for 5g batch. This lets you make anything up to about 1.090. If you want to make bigger beers than that with all grains...then you probably need a second mash tun for those beers, because a 20g cooler would be a little small for a 1.036 beer.
Finally, when you are troubleshooting efficiency it is important to collect a LOT of ACCURATE data. Volume measurement is the most sloppy, but is critical in assessing any problems. You MUST calibrate your volume measurements. I use ONE vessel for measuring all volumes. I have a graduated 1 gallon jug, and I use it for everything. I use it to make markings on my sight glass on my brew kettle, and on my fermenters. I use it to measure and transfer wort from the MLT to the kettle. You need to measure volume that is accurate to the single CUP---1/4th of a quart. If your kettle is only graduated in gallons, you cannot be this accurate. You must graduate in QUARTS, then you can eyeball the fraction of a quart.
For SG, you need to cool your sample below 100F in order to get an accurate SG reading. The correction tables don't work very well above 100F.
Ensure that your thermometer is calibrated. Check it in ICE water and boiling water. use one thermometer for all measurements until you are certain others can be trusted. When you have a problem to sort out...revert to one trusted instrument for everything.
For batch sparging, you should measure the SG, temperature, and volume of each of your runnings. Calculate the amount of sugar in each running, and add it all together. Then make the same measurements pre-boil of the whole wort in the kettle. Everything should add up. If not, something is wrong somewhere.
I'm not a fly sparger...so I don't have much advice other than the instrumentation advice above. I know that manifold design is KEY, so is crush (even more than batch), and so is runoff rate.
See these links for a thorough treatise on the topic:
http://braukaiser.com/wiki/index.php?title=Understanding_Efficiency
http://braukaiser.com/wiki/index.php?title=Troubleshooting_Brewhouse_Efficiency
there....I think I kept this pretty short.