The classical interpretation is of a quantum particle spinning on its axis, which gives it an intrinsic angular momentum. At the same time, quantum particles that are orbiting (not a good word to use for atoms, but again we're in a classical frame of mind) the nucleus have an associated orbital angular momentum. Like most other energy-related quantum properties, both angular momentum values are quantized (can take on only specific values). In the quantum world, we use a 'unit value' that we usually then multiply by a 'quantum number' to indicate the total observable magnitude - the reason relates to the Schrodinger equation, which is probably unimportant to go into right now. Fermions, which include electrons, have half integer spin quantum numbers (1/2, 3/2, 5/2, etc.). The total amount of spin angular momentum is determined via the simple equation SQRT[s(s+1)]ħ. Here ħ is your 'unit value' and the quantum number is based on s. For electrons and other elementary fermions, s can only have a value of 1/2, so the total allowed spin angular momentum for an electron is SQRT[3/4]ħ.
An important thing to understand is that momentum is a vector quantity. For reasons well beyond the scope of general chemistry, we are usually concerned about the portion of the angular momentum that is directed along a single spatial axis, which (for one of the three potential axes) is also quantized. For convenience, we pick the z-axis. A total spin angular momentum value of SQRT[3/4]ħ has z-axis projection of (1/2)ħ. However, while the total spin momentum magnitude is fixed at SQRT[3/4]ħ, we note that there is no reason why it has to be oriented up or down (along the positive z or negative z axes). In the absence of any other particles that might influence it, there is a statistically equal probability of both possibilities. For this reason, we say that the z-axis spin angular momentum is ±(1/2)ħ. The associated quantum number is simply ±(1/2).
(This is, by the way, mostly analogous to the orbital momentum quantum numbers you may have learned about, L and mL, the former representing the total orbital angular momentum and the latter representing the component of that momentum that is directed along an arbitrary [usually z] axis. The difference being that the orbital angular momentum isn't confined to a single half integer value. Notice that whatever the positive L value, the mL values have both positive and negative integers, reflecting the fact that the total momentum L is just an absolute magnitude, but mL represents that magnitude along the z-axis, and there is not preference for it to be directed in one direction or the other. The total spin momentum quantum number, s, is usually not of much importance, but the quantum number that indicates the amount of momentum directed along the same z axis has a lot of importance in chemistry and physics and is one of the four you usually learn about in general chemistry.)
How does this relate to anything important? Well, electrons are charged particles. A moving or rotating charge creates a magnetic field. If a nearby charged particle also generates its own magnetic field, then there becomes an energetic interaction between the two particles. Or, in many cases, between the particle and itself - if an electron generates a magnetic field due to its rotation around a nucleus, and it also generates a magnetic field due to its internal spin, then those fields can interact either positively or negatively, depending on their mutual vector orientation. These kinds of interactions are responsible for the way electrons fill into the orbitals of multi-electron atoms. Also, they impact the energy states that an electron may occupy around an atom. In spectroscopy, we see finely divided spectral bands because of these interactions between the spin and orbital angular momentum, something we call fine structure. Basically, the orbital momentum and spin momentum, even for a single electron, can be aligned in parallel or antiparallel fashion; one of these orientations costs a little more energy than another, which impacts the potential spacings between electron levels that we probe with spectroscopic techniques.
In multielectron atoms, the spin impacts the electron configuration in a number of ways. For one thing, fermions obey the Pauli-Exclusion principle, which means that electrons sharing the same orbital have to line up in antiparallel fashion. The implication of this is that there are only two electrons allowed per orbital. This pretty much determines the physical nature of all matter, including the way substances interact, their chemical and physical properties... everything! Actually, though, spins prefer not to line up in antiparallel fashion if they can help it, which means that when you are filling orbitals that have otherwise the same energy, it is the usual rule to fill them each half empty first. For example, in nitrogen, you have three p orbitals and three electrons. Each electron technically can take on z-directed angular momentum values of ±(1/2). There are a number of ways you can fill the orbitals, but the lowest energy state (all other things equal, which they often aren't) is to put one electron into each of these orbitals, with their spins aligned. Each of these unpaired spins has the potential to combine with unpaired spins from nearby atoms... so now you see why nitrogen often likes to bond three times. (The 4-fold bonding preference of carbon, which has two unpaired spins, is not as simple to explain, but this is a start for you I hope to see why spins have such enormous importance in chemistry.)