I do not own Elemental at present, but I've gotten a chance to play it. It's something I was very interested in before release, but I have no plans to buy it now in its present form. The reason? My initial impression was that the game wasn't really very satisfying as a strategy game. For me, strategy games are about economy first and combat second, so that was my focus when playing. I did very few quests, recruited very few NPCs, and did not cast many spells, so I haven't had the full experience, but I believe my concerns are still valid.
My proposed changes might not be the best way to fix the problems, but I hope they spark some critical thought about the game, as a lot of what I mention here can be changed without making huge changes to the game itself. And even if these changes don't make it into the core game, they are the kinds of things that should be possible to implement into a mod.
1) Economy
Design Goal: Reduce the number of cities people build so that city spamming is not a particularly good strategy.
Implementation: A remarkably small quantity of resources are placed on the map so that people only build cities by them.
Analysis: There are two values for resources in any given strategy game, the base value and the multiplier (i.e. I make 10 gold per turn, and +200% gold per turn, so I get 30 gold per turn). Several good strategy games do not include a multiplier at all. Multipliers are generally always available, so economic success in pretty much every strategy game involves maximizing your base value and building the requisite multipliers to boost it.
This is also true in Elemental. However, Elemental offers you virtually no control over gaining base value points, since they ALL come from resources on the map. Want to play with a technological focus? Better find a lost library, or otherwise you won't be able to! Your entire success and failure is determined by which resources you have access to.
Is this fun? Not really. If I want to play with a complete technological focus, I want to have more technology than everyone else. I expect to be crushed militarily while I research all kinds of new buildings that I don't really need. I do NOT expect to be out-teched by someone who just happened to get more lost libraries than I did.
Additionally, resources are so few and far between that huge tracts of land are just left barren wasteland filled with monsters, which isn't particularly interesting even if it does fit the setting.
Changes: There need to be more resources, and the resources need to be less effective. A city with a single gold mine should not be able to bankroll your entire empire, but maybe a city with two or three of them should. With more resources, utilizing city placement as a mechanism to increase your base value becomes a strategic element - do I place my city near two gold mines, or near a gold mine, metal mine, and farm? Right now you just see a gold mine, place a city, and rake in the cash.
More buildings to increase your base value in cities so that you can actually specialize your cities properly instead of just building the one of everything that you're limited to. This would make it possible to actually generate the resources you need without having to go out and find resource deposits, so that not having something isn't a death sentence. "But won't people just spam cities to build these buildings?" See next point.
A legitimate penalty for building cities. The design concept of "Well don't provide players any incentive to build cities and they won't build cities" is grounded in a reduction of player options. Strategy games are based on wanting things you can't have, not about not wanting things at all. So the players should want to build new cities but should be unable to do so because the cost is too high or the negatives outweigh the positives. If you spam cities without any resource deposits nearby, then you should be penalized with crippling maintenance that your weak cities can't handle. Or the cost of building these cities should be high enough to force you to build only what you need. Or both.
A focus on city specialization. Buildings are built faster than technologies are researched, and cities have a high number of slots for buildings. The result is that every city can basically build every building, and can do so before the next building is even available. So cities are frequently not building anything. Strategy games are about offering you more buildings to build than you can build with your current time and resources so that you must select the ones you need. In Elemental, you never have to choose, and even if you disagree with everything else I've said, you cannot disagree that strategy games are about making choices.
2) Combat
The most interesting thing about fantasy strategy games is that you can have different races and these different races can each have their own focuses and their own units with their own special abilities. In Elemental, this is sorely lacking. Many people have said this already so I won't rehash what they've already been over, but if you're going to include a tactical battle system, then you're going to need some meaningful tactics in your battles. And meaningful tactics comes from unit variety.
This is sort of the same thing, but good equipment variety would be nice, as well. Designing your own units isn't an interesting gameplay element when the only consideration you have to make is "can I afford to field units that are this good?" Weapons need personality. In Fire Emblem, if you take away the weapon triangle, then the sword users STILL beat the axe users because they hit the axe users with many weak hits and the axe users have low defense. The axe users' incredibly high attack easily overcomes the lance users' defenses, and the lance users' defenses render the weak hits of the sword users ineffective. This is the kind of distinction I want to feel when I'm choosing weapons, and certain weapons/armor that are weaker statistically but have useful special abilities MUST be included. That's the sort of thing people are looking for.
3) Technologies
The technology system is neat, but I want more of a tech tree feel. This could be accomplished by making the probabilities of technology appearance malleable, so that fewer 100% of the time techs appear, and as you research more techs in a given area, "lower level" techs in that area become more likely to appear. That way, blitzing straight down one branch of the techs is somewhat mitigated by the fact that the tech you want might not show up until you try a couple times, but if you're doing a broader research scheme, then it will. This also gives more meaning the the probabilities techs have of showing up, rather than "this tech always shows up 70% of the time in every game."
I was also disappointed in the tech tree as a whole; I felt as though there just wasn't enough there in the Civilization and Warfare trees, and even less in the others. I rarely ever felt as though there were multiple techs that I'd benefit greatly from now and was forced to forgo one in favor of the other. In particular, I found the way the Adventure techs for Exploration work to be very disappointing. They really should reveal hidden resources on the map like they say they do, not spawn new, predefined resources on empty tiles.
4) Variety
I find Age of Wonders 2/SM to be a fascinating game. The reason for that is because each race is unique, has its own units and flavor, and each spellbook has its own spells. Everything is different, yet balanced. That's the kind of thing a strategy gamer lives for: balance beyond pure stats. Trying to figure out if it's better to have Life Stealing or 4 attack. If the speed of a cavalry-based army is worth not including healers. Will I get killed if I research this technology before I field my army? Is the "Raise Skeletons" spell good enough to create a meat shield to leave my archers free from melee, or do I need some spearmen? To really draw people into the world, there needs to be much, much more of this than is there right now. And every bit of variety adds a relevant decision that the player has to make.
Conclusion
Strategy games are about giving players decisions to make, and these decisions should have a meaningful impact on the success of the players who make them. Elemental is lacking in decisions. City placement is basically decided for you (put the cities by the resources, build cities wherever there are resources), building selection is basically decided for you (build pretty much everything in every city), unit design is basically decided for you (build the best equipment available unless you can't afford it), and even technology selection isn't particularly engaging.
Elemental may not be purely a 4X game, but it is still primarily one, and I think a lot of people are put off by the lack of strategic depth, even if they can't pinpoint it as the reason. Strategy gamers like to constantly feel like they're missing out while they save up to buy things they really want, and until Elemental can bring that feeling to the players, I don't think it will ever be a good game. Fleshing out the current systems to provide the depth that people expected from launch won't be enough on its own. There's a lot to be learned from looking at older strategy games, both TBS and RTS (ones that have cities, like Kohan) and asking why they are fun. I'd love to see Elemental take that, expand on it, and one day become one of these games in its own right.