Just to clarify.
I'm not saying Plone or the plone organisation is completely broken.
I'm also not trying to say people are doing the wrong thing. Everyone is doing what they think is right and the effort is truely awesome.
I'm trying to say that for any organisation UX is really hard for engineers to do. It's hard first to decide who the users and then understand them. It's easy to solve their problems once you do know who they are however.
I'm not saying that I'm unhappy with the direction of Plone, but I am saying that we still haven't decided enough who our users are or aren't or created a way that developers really know who they are building for. It's all implicit and mostly made up on the spot. In an ideal world, every sprint would have a UX expert on every team and some sample users they could bounce ideas off. And those sample users would include large and small sites and everyone would be on the same page that this is really important.
I personally think small sites are important, not because they are my customers (they aren't) or because I think Plone should "only" be about small sites, but because strategically it is very good marketing. Good marketing helps us all. Plone has ignored small sites and it has shrunk because of that. I can't prove that, but I believe it to be true. Big sites should continue to be important to Plone, but small sites and the problems they face are under represented (not deliberately).
As you know I'm a UXer (designer, architect... whatever is trendy these days) but I'm also a Plone fan since the version 1.0. Over the time I've been to four Plone conferences (Vienna, Budapest, Napoli, Bristol) and I used Plone to build many websites. That lasted until Plone 2.5 then hacking Plone through the ZMI + some sporadic external help wasn't enough anymore as things got too complicated for my limited knowledge. I managed to make two more big projects based on Plone 3 with the professional help of external (hard to find) Plone consultants and then I moved to the UK because I wanted to focus my career on what I hoped I could do well: user experience design and research.
Once I left, my ex-colleagues in Italy quickly turned to WordPress for new projects because of the easy hackability of it + the availability of themes, plugins and php developers on the market was much higher.
At the beginning of this year I even made my first WP website for my brother in law's new business: I bought a theme, customised it a bit, installed some plugins (among the many available) and in a few days I was done.
I thought to use Plone because I know its value but I had little time (as it was for free) and the lack of professional themes + the cost of the hosting discouraged me in the end.
So I'm not saying that this is the primary persona/scenario Plone should focus on and I'm aware a lot will improve in this sense anyway with Plone 5 but keeping the threshold low allows new users to approach Plone and then who knows what can happen?!
Personas and scenarios are important because they make you focus on the user needs and force you to prioritise them based on the percentage of a the personas in your user base. It's not easy to get to know what those personas are unless you have a huge budget for market research (and sometimes that's not even enough). We also need to keep in mind that the user doesn't always fit the same persona all the time but often he put differents hats on, permanently or multiple times a day.
E.g. I want my blog in Plone so I need to install it (somewhere) and customise it first as a developer-ish, then I'll be most exclusively the end user of it. Or I'm the admin of a section of an intranet but I'm also a normal employee on the rest of the intranet (and I'm not just talking about different permissions).
So, to conclude my already too long rant, I believe that the future Plone roadmap should also be drawn as much by UX people as by engineers. The latter tend to think about tech solutions rather than user needs and usable solutions to them. It's not a bad thing, I wish I could do more my self without the help of developers sometimes but we need to accept that we are complementary and we need each other.
Most of the cutting edge corporations have already migrated or are in the process of migrating towards a lean UX agile design-led way of working. We should try to do the same with Plone too using all the UX people that one way or another have been involved in projects based on Plone or, even better, use it every day for some reason.
In addition we could use remote tools to ask e.g. web editors about how easy (or not) is to use Plone and how they would improve it. That would require an effort shared by many Plone providers and supported by the clients.
