Cities usually fail because you spend too much. Careful expansion and safe investment is the key to success. Start small and keep your hands off. Just like real politics you can't force decisions every year, and you shouldn't adjust policies to counter immediate real time problems. Treat every decision you make as something that will truly come into effect no sooner than a couple of years later, not a way to appease the public the very same morning your advisors point it out to you.
Laying down new zones before previous ones have been properly established is a common mistake which will only cause your population and business to move around and never really settle in which slows down business and prevents population growth. All those condemned buildings that seem to crop up every few years is a result of this migration within your city and this is revenue you're missing out on which you probably need to recoup from the investment that caused the problem to begin with.
I'm no min-maxer who knows the secret formulas to successful cities, but I can at least keep my cities from going bankrupt and grow by doing what I explained above. Time flies in the simulation on your computer and there is almost no limit how fast you can expand, but you have to imagine your city is a real place with real people. If a mayor or other type of politician was pushing lots of decisions at the same time while drawing up new ones every month he'd be called insane and disorderly. Developing an area is more than just laying down zones and drawing up a road network, you need to wait for construction and population numbers to catch up which can take years. The supply-demand meters are not meant to be satisfied IMMEDIATELY, ie don't start laying down lots of industrial areas just because the meter says you need it, but start making plans for expanding the industrial base over the coming years in at a sane pace.
