Starting Simply

Developing an Adventure

Stage 1: Paint with a broad stroke - This is more about the world around the adventure. Is it a dungeon crawl? An outdoor adventure? Something urban? What is threatening the world? Pirates? Orcs?

Stage 2: Identify key plot points you want to develop further - River piracy, a missing artifact, rivals in a power struggle

Stage 3: Connect the dots to create structure - The Drow are behind the piracy on the river, and have stepped up activity of late. Merchants are in an uproar. Now an attack in the market has brought forces into the open. You need to find the Drow hideout and battle their network of evil. Meanwhile, the dead are stirring in the outlying towns. It has riled the Drow, but what is the connection.

Stage 4: Add non-combat details for color and skill checks - gather information at the market, maybe a bridge is out, etc.

Stage 5: Add “random” elements for variety - a Cave Bear wanders across your path, a Blue Dragon swoops down to threaten you. This is good in particular because it shows characters there are times it is better to avoid an encounter than embrace it.

Its in the details

I finally picked up a copy of The Hoard of the Dragon Queen, and have been reading it and play testing the opening encounters. I'm glad I waited until now to do so. Two further quick points before I elaborate. We've been able to play several sessions over the holidays, with my brother serving as DM. I also finished putting together a first level adventure that I'm eager to run.
So what is the connection? It's in the details. The details of DMing that is. I like the elements that are in place with The Hoard of the Dragon Queen, but it is not very DM-friendly. The first encounter is set up with a read-aloud paragraph followed by a few lines of text identifying the running woman, suggesting that if you don't interfere, they'll assume you are with them and kill the family, then adding that if you are successful, you should go in "x" direction. It then says you need to defeat 3 more groups of raiders to reach the keep safely. The next paragraph starts referring to future missions.
I feel like they are missing something, and something big. I know they want to reduce dependence on grid-play, but this is extreme. Spells still have range, the first fight is against Kobolds, who get pack attack features when an ally is within 5' of their enemy, and characters still get move actions with fixed values for movement, so you can't remove these things entirely and still play the game.
I like the fact that they've created a quick-playing combat system. It is fun to use, and the quick resolution keeps everyone at the table focused on the matter at hand. In all honesty, it would have been enough for me if they'd said, "the alley is 50' ahead of you, and you can see the keep looming over you on the right several blocks away." Instead, they say "you see 8 Kobolds attacking 5 villagers, if you do nothing the villagers die, if you kill the Kobolds, go to the Keep."
The Lost Mine of Phandelver introductory module has the same issue. They write the encounters directly into the flow of descriptive text for the module. This forces you to constantly pick at columns of text to glean the nuggets you need at any given time. I would prefer a callout box listing the encounter information for easy reference (# monsters, attack bonuses, damage, defenses, special abilities or actions they may take, location).
With our home-spun campaign, the biggest hurdle to jump has been introducing and resolving non-combat situations. We've been playing for nearly 35 years, but without much regularity for the last 20. Part of this has come from getting busy with other things in life, part of it the decline at periods over that time of the game itself, and partly the evolution of the game into something that leant itself well to one-off battles. We stopped playing non-combat scenarios, and forgot how to have them flow well. There is a good deal of rust on our cogs, and it showed.
You could tell, however, that he had done a lot of testing and tinkering with battle groups to find an interesting balance that could be met well. But as the adventure rolled on, it began to feel too crafted, too one-dimensional. My brother has always been adept at creating parties of monsters that make for interesting and dynamic combat, but they hold the line too closely. Every battle starts to look the same, every encounter is a slight variation of the same 3 or 4 monsters mix.
Comparing them to my own dungeon, I am that much more eager to run it. Hopefully, I have found a good balance between detail and story. I've created a crypt, with several undead types in it, as well as various underground vermin and detritus. For each room I've included not just what inhabits it, but how they might react or respond to intrusion, how they might interact with rooms around them, as well as triggering elements for traps or ambush, etc. There are places that reveal more info to characters making successful skill checks.
When I finished, I was unsure if what I had was really worth running. I threw much of it together before the DMG came out, maybe a third before the Monster Manual. I chose a crypt as a throw-back to first edition because I started before the Players' Handbook came out. It was a hodgepodge of elements, cobbled together to fit a game system I could only hint at parts of. But in retrospect, comparing actual play with another DM to official "direction" from Wizards, I think it is actually pretty good.
And I think the value for it lies in the details.