Weeks of frustration passed before I finally decided to contact the support team at MyRegistry.com, a website that allows users to combine items from all over the Internet into one gift registry. Most of my time on the site is spent in the morning, at night or on weekends, when support wasn’t available. Plus, I assumed that the development team already knew about the issues I was facing. I didn’t feel like being one more annoyed person repeating the issue they’d heard again and again.

But here’s what was happening: Every time I’d sign into the tool, I’d use the clunky drag and drop functionality to put various items into an order that would make sense for our wedding guests. Every baking-relate item from Crate and Barrel would go together, the handmade serving spoons on Etsy would accompany serving platters from a local shop, bar tools would sit just above the Old Fashioned glasses, and so on.

After spending too much time organizing, I’d log back in a couple days later to find many items out of order. I’d (somewhat) patiently reorder them, finding myself wondering about whether the organization is agile — and whether this time my work would pay off. I always left the site frustrated, knowing still that my threat to quit the site was empty.

This morning my mother emailed me to find out whether I planned to add any other pieces that go with the one set of dinner plates. But the rest of it was there, somewhere. It was time to talk to support. At 8am, a nice rep named Danielle answered my chat request and quickly pointed out a “Visitor View” dropdown with an “As Displayed Below” option. That was all I’d need to do to solve the problem, she said. It wasn’t a bug after all. It was a misunderstanding. Oh, my.

I breathed a sigh of relief, slapped my palm to my forehead, thanked Danielle, considered the ridiculous amount of time I’d wasted up until that moment, and wondered how many other users had asked the same thing.

The reordering wasn’t my only issue with the product. (I know what you’re thinking: First-world problems. I’m thinking that too.) The drag and drop feature is also very user-unfriendly, though I feel the pain, as we still haven’t quite figured out the best solution for Sensei. The iPhone app is pretty useless. The site is not pretty at all, and is in an industry where branding is key.

But here’s the thing: None of this kept me away from the tool. The folks behind MyRegistry satisfied a need significant enough that I’m willing to clench my teeth through all kinds of usability issues. And if they had waited to perfect the drag and drop or ordering before launching, it might not have launched until long after my wedding and many others.

See, the concept is brilliant. I can add items from anywhere, including small local businesses and Etsy dealers. I can sync registries from other stores so that my guests don’t have to search through different ones. I can add funds for our honeymoon or favorite nonprofits. The website even allows visitors to price shop on the web for certain items.

Because of all of this, I’m even willing to force my friends and family to suffer through the same confusing user experience. I feel confident that as its traffic grows, MyRegistry will use input like mine (I ended up telling them about the drag/drop) to keep improving their delivery and product. But even if it doesn’t, others like me will probably keep using it until something better comes along. Why? Because at its core, it’s much different and way better than its competition.

MyRegistry is proof that we can make an app as pretty or defect-free as we want, but a solid idea trumps all.

When Sensei users say to me that they won’t use the tool for retrospectives if there are any bugs, I question the need we’re satisfying first. I believe in Sensei. I want to add features that will make it even better, but is the need strong enough? Are we too early in the agile movement for a retrospective tool? Do we have to wait for organizations to fully understand and value continuous improvement before we’ll get a large group of consistent users?

There are awesome agile teams who’ve been using Sensei for a while. They get the value. They were willing to stick it out when we were dealing with some IE-related defects, and they’ve given great feedback because they feel passionately about the idea. They’ve seen their feedback result in new features. But will there be enough of them? And for those who say they won’t use it without X, Y, or Z, will adding that one feature really make the difference?

My goal is to do everything we can to find out.