Regarding process, when you choose a stock, you have generally limited your exposure to 2-3%, with sector exposure weighed by your themes or projections for where we are in the market cycle.( This is my understanding based on reading your site for several years.) Since an ETF, even a narrow etf range selection, represents a pretty broad selection of holdings, are you willing to use one etf for a 5-10% position in a portfolio?
This might be a follow up to a comment I made yesterday about combining a stock and an ETF to build a sector. I've written many posts about building a portfolio at the sector level. Building a sector within a portfolio (the way I prefer to do it anyway) starts with the decision to overweight, underweight or equalweight that sector against the benchmark index--we use the S&P 500.
Taking the industrial sector as an example; it is currently about 10% of the index. For the sake of our example we want to equalweight the sector and allocate 10% to it. The best way to build the 10% will depend on what we think is going on in the world and what is the right thing for the particulars of the client. If 10% works out to $14,000 it probably doesn't make sense to buy six different things for the sector due to commission drag. If the client has a lower tolerance for volatility then loading up on something like the EG Shares China Infrastructure ETF (CHXX) or the Claymore Global Timer ETF (CUT) may not be the best thing for them.
The most basic way to build the sector would be with the Industrial Sector SPDR (XLI) or the equivalent domestic sector fund from any of the other providers. Another way could be to buy a fund with a little foreign exposure along with domestic like the iShares S&P Global Industrial ETF (EXI). That one is 49% US with the rest obviously being foreign. Another broad fund would be the SPDR International Industrial ETF (IPN) which has no US exposure, it has a lot of Japanese and European exposure.
For an account where one broad (relative to the sector) fund was the best way to go I would have no problem with a fund like those serving as the entire 10%. Obviously picking a fund requires looking under the hood and being comfortable with what you see.
Another way could be a core and explore type of thing with the 10%. In this case some large portion of the 10% could go into a broad (relative to the sector) ETF and then some small portion into a theme or a country or something else. This could include things like wind, defense contractors or the GlobalX China Industrial Sector ETF (CHII).
Or a broad fund can be bypassed altogether in favor of some combo of themes, countries and industries within the sector. In the case of 10% putting 3% into two different things and then 2% into two different things would get it done. At that point the decision becomes what the best way to build it might be. For example this sector would be a good way to add Germany with maybe a stock like Siemens (SI)--this is just an example I want no part of Germany or alternative energy (most of the companies or ETFs focus on industrial companies that make equipment) or infrastructure (ETF or stock) or do some domestic stock picking with something like 3M (MMM)--that is just an example we don't own 3M.
How far you go with countries, themes or the rest is largely dependent on the time available to spend on the portfolio. It is obviously a lot easier to track one very broad sector fund than a Japanese nuclear power plant maker, Korean shipbuilder, Spanish solar company and Brazilian steel company.