I think the only solution if you want complete transparency is to publish details of the outliers too. If someone gets 30 pips negative slippage, then customers need to be aware that they could experience similar. Likewise, if someone gets 30 pips positive slippage then that should be published too.
Of course if 3 customers trading at $1000 a point are slipped by -5, that obviously isnt balanced by 3 customers slipped by + 5 who are trading chump change. FXCM have quite rightly been pulled on this particular point when they tried to pull the wool over customers eyes with their slippage stats claiming the NUMBER of instances of positive slippage is balanced by an equal NUMBER of instances of negative slippage.
Technically I suspect that there's no reason why you couldnt provide the daily distribution in spread (although I suspect commercially you probably wouldnt want too, and even if you did, I supect people would not trust the source of the data)
At the end of the day, its up to the individual trader to monitor this stuff, and either they get what they need or if they dont, then they need to seek alternatives. To be blunt, I dont care what typical spreads are, or average spreads, I only care about the fills that I get, not the average of a quarter of a million other punters.