Well, with that context, I'm not convinced that idea has much meaning. Unless I'm missing something, I could rephrase it as "poorly thought out 1-off outreach programs are not worth it, but they can probably be tweaked"
Besides that, your response reminded me of the importance of the target audience, and with this new shiny forum, why not bring it up.
My idea of how to think about the target audience flipped on its head when I heard Matthew Nisbet give a talk when he was touring with Chris Mooney about 4 stages of science engagement (I can't find the reference and I'm paraphrasing).
Group 1: People, like us, who are always deeply embedded in the scientific process.
Group 2: Citizens, who seek out science by bringing their children to science museums, subscribing to science magazines, or visiting planetariums.
Group 3: Americans who are passively interested in science, they tune into television shows with science themes like PBS Nova.
Group 4: Those who aren't interested in science and won't seek it out. The only science they are exposed to is from the meteorologist. This group is often associated with the most hostile views on science and believe that science threatens their ideologies.
My holy grail is to go after group 4 when designing outreach programs, but in reality, the most "effective" outreach I think I've done (and I've done a quite a bit) has reached group 2 and only a smattering of group 4's children (who happen to be in the classroom). Thoughts?