1. What to call the game.
Elemental: Fallen Enchantress. That way the expansion would be called Elemental: something else.
I am always in favor of more summoning spells to offer that as an optional strategy. The most powerful summoning spells could be granted from quests.
I'm not sure how current quests function, but I would like to see quests with multiple choices to choose from, but the result of your choices is random with mostly favorable, and sometimes terrible, results. This way, quests cannot be memorized by players who would naturally select the choices with the best reward, or avoid a hard choice path.
I'd also like new quests to spawn automatically, without requiring to purchase the quest spawner item from the shop,
Never build a scout/explorer. It's a waste of a city's queue. Instead, use the main army to scout with while killing monsters and doing quests at the same time.
Always attack other armies/cities with a full army size. Never fight with 3 units per army or less. Never fight with a pioneer in the army. In fact never include a pioneer in the army for the AI. Just make the AI protect its pioneers by moving the pioneer into the same tile as another army, but keeping them separate armies.
I personally favor the economic game: focus on building city improvements, and building units only when required.
As a general rule of thumb, a great way to start off the game is to research leather armor and spears. Fully equip the sovereign with leather armor, and give him a spearman or two for meatshields and extra damage. Then the sovereign is ready to go monster hunting and questing fairly safely.
Generally, always buy armor for champions/sovereigns to greatly increase survivability. With a good weapon, they become tanky damage dealers, even with lower initiative. Armor is sooooo important for survivability that its painful to see AI champons carrying only their initial armor. If the AI cannot purchase armor, or find some from quests and monsters, I wouldn't mind seeing the AI teach in regard by getting it all for free.
Personally, I've never used healing items on my champions: they are so tanky with armor I've never had to. But I suppose an AI could make good use of them with their champions if the healing done is about 30+ in a turn.
In my experience, the AI never won a tactical battle against me. The least the AI could do is to gather 3+ armies and wait for all of them to be within striking distance of target army or city. When they are in position, they should all attack in the same turn so that the defending army does not recover its health before the next battle.
7. Game Impressions
I voted good.
To make it good+, it could do with some balancing and pacing.
As an example for balancing, I feel that at the beginning of the game, I'm much better off using my champions to do physical damage. Magic damage spells just don't compare to a good weapon early on. Especially Burning Hands. It's 1 damage at level 1 is it?
As for pacing, units with armor take too long to build, and I usually prefer building an improvement instead. I would like to see unit train times capped at 15 turns for low production cities, and 10 turns for higher production.
There are more city improvements, but it feels like there are a third of them that mainly do -unrest. I tend to build everything else first and keep tax rate low.
To make it excellent, it needs some more features to flesh out the experience and to make it stand apart from comparable games. Of course this would be the expansion for the game.
I'm thinking special city siege tactical battles with unique game mechanics and city improvements to fully develop the idea.
Another thing would be dynasties. To bring out it's full potential, children should inherit the genes (traits) of its parents. The Total War series could be used for inspiration. In addition, marrying children to other factions would have real effect on relations, and rules of inheritance could determine the ownership of cities. It is just something to make Elemental to stand apart from other 4x games.
More Kingdoms vs Empires features would add to the lore. Having a type of UN council for Kingdoms, and a UN council for Empires with resolutions that affect both like factions and opposing factions. E.g. a vote on resolving intra-Kingdom disputes, or how to deal with a powerful faction on the opposing side, or how to deal with a threatening random event.
One feature I currently hate is diplomacy. It's just ordinary, and I'd compare it to Civ 5's diplomacy, which I also hate.
I think that currently there is nothing meaningful to trade. If I ignored diplomacy and only declared wars or signed peace treaties, what would I would be missing?
Treaties are rather dull. +% to guildar or research just are not interesting. Something more interesting would be a treaty that shares the ZoC for strategic spellcasting purposes, two factions sharing their mana pools for spellcasting, or two factions sharing their shards to power their spells. Useful treaties for allies fighting stronger opponents.
I am completely against relation modifiers to determine relations.
I would rather see two factions discussing issues relating to them mostly, and also other players. Agreeing on issues, successfully fulfilling the goal of issues, increment the relation meter . Refusing issues decrements the relation meter. Going to war or killing someone's mother resets relations to 0.
The idea is to get factions to talk in the form of a conversation with proposals and counterproposals using words rather than balancing the perceived value of a trade with guildar and influence as it is now.
12. The Campaign
Makes good sense to separate the sandbox game from the campaign, It felt weird in GC2 to keep researching the same techs over and over again after every mission. All I'm really looking for is a good story, exposure to the Elemental lore. Maybe some in-joke references to how War of Magic was the worst war in the history of mankind. And of course an alternative tutorial that introduces various features used in the sandbox.