Tips on how to analyze economic time series data

Be it stock prices, index valuations, budgetary spending, deficits, exchange rates; virtually any significant economic data is analyzed using time series data. In these days of Excel 2007, it is a great temptation to just plug the numbers in, select a graph and use it for whatever purpose one wants. But such a simplistic approach would be doing an injustice to the analysis, and also would reduce the chance that the data is analyzed, understood and appreciated in the way it should be. So what to do? Based on my not insignificant experience of this, I add some guidelines below. I don't claim this to be an exhaustive list, and comments are welcome

• Adjust for Inflation:-Specially for monetary values and interest rates, this is crucial. And sometimes it gives very interesting insights. For example, the real interest rates in India have been negative for quite some time, and that makes us appreciate why the RBI is chary about reducing it.
• Try per capital/per unit numbers:-It allows for adjustments for population/volume growth, and yields interesting results. Thanks to its population, India lags behind on such data
• Put the numbers in context:-Expressing as market share, budget share, relative data etc puts the numbers in context, like in per capita.
• Use Indexing if possible:-That makes the growth trend easier to understand
• Use secondary axis if useful:-For example, you want to show that FDI inflows fell when the exchange rate appreciated. Instead of using 2 graphs, you can use 'Plot on Secondary Axis' option in Excel, and plot them on the same graph
• Try to cover an entire economic cycle:-Otherwise, the result may just be because we are in a boom or a burst. This is true especially for stock price/valuation data
• More the better:-We often tend to forget history, and that is a reason why bubbles repeat, whether it be across countries or across time periods. Hence, a longer data set with more sources, helps to triangulate trends and have a broader perspective. 
• But be careful of underlying structural changes:-With Excel 2007, one can use callout and other options to depict the reasons for inflection points. Use those with care where warranted. For example, the 1991 liberalization would have lead to inflection point in most Indian data.
• Strike a balance between aggregates and subclasses:-Given the high Gini Index/Digital divide etc, one should be careful about what the data signifies for India as a whole. Often, the divide between states/castes/gender/urban-rural population parameters, would be large enough to vitiate any extrapolating of the conclusions. For example, India telecom boom should be read in the context of rural teledensity just being 32%(versus urban of 140% or so).
• Focus on outcomes not just on spending:- Especially in case of data on spending, avoid the tendency to mistake spending for outcomes. Try matching the spending to the outcome(say education and pass rates, health spend and mortality rate/life expectancy). For share price data, that may mean investor participation, retail folios etc