Taking Feedback Into Consideration: Revising Steemboxes.com

Steemboxes.com was launched barely 24 hours ago and the response has been fantastic. While nobody has taken the plunge and purchased a mystery box just yet, some of the feedback received has been extremely helpful in refining the concept and making it into something that Steemians would want to use.

One of the loudest pieces of feedback was that the boxes are too expensive. Because said packages are being shipped out of Australia, it can be prohibitely expensive to ship things overseas (especially Europe). The concept took into consideration that profit margins would be very slim, and at times, boxes would be a loss leader in that a box might lose money after shipping, but the promotion would bring more users onto the site.

At launch, I did try adding a mystery envelope for $30, which I believe some people thought was still too expensive. Understandably, this pain point has been the hardest to reconcile with. I knew the really exorbitant boxes probably wouldn't attract buyers, they were more tongue-in-cheek if anything, but it has made me realise that people are hesitant to spend money on mystery items.

I will get back to this point in a moment, but another piece of feedback which relates is a comment from @shaidon who responded

Do you have at least a range of items that it actually could be?

For example, I roughly know what is going to be in a Lootcrate box. There will be a T-shirt and a few other trinkets, with a monthly theme.

Obviously with this there will be a random element, but is there a guarantee or either quality or authenticity?

There were other pieces of feedback which echoed similar concerns, they want to have an approximate idea of what is in the box. I can completely understand, especially during the prolonged bear market, you want to make sure you're not buying a box of cheap Wish.com items or rubbish lying around someone's house stuffed into the box.

I also received some helpful feedback from everyone's favourite Steemian @ausbitbank who pointed out that people might not feel comfortable with the Steemconnect permissions being asked for. I realised in error that when developing the app, that I had copied and pasted scopes from an existing project using Steemconnect which required voting permissions and so on. Being such a simple app, no permissions were really needed.

Thanks to ausbitbank, I realised this error and removed all of the offending scopes. The only permission it needs is to login, which currently allows you to keep track of your orders you have made through the site.

Another helpful piece of feedback from yours truly ausbitbank was that he and others would find it more convenient if the site utilised the Steem Keychain browser extension, which doesn't require logging in and only granting permission. This feature is coming, so if you have the extension installed, it will use that for purchases instead.

All of this feedback is extremely helpful. Steemboxes.com launched as an MVP of sorts, with the intent to grow and evolve based on honest and helpful feedback like I have received. This is the kind of feedback that no amount of STEEM can buy, so thank you.

Change is coming

This is why I believe a change has to be made to the concept. It will take some time to distil the ideas into a workable concept that is both viable, transparent and comforting for the end user.

I have some initial ideas as to how it could go and definitely will be wanting some feedback from the community as they're fleshed out and then released into the wild. This will be a process that might take a few goes to get right, but ultimately, I think this is a concept that requires community buy-in and if you ignore the community, your app is dead in the water.

One of the ideas I am toying with is boxes have a set list of items that you can possibly get, with percentages attached to them. A $50 box might have a $100 item that could possibly be included, with perhaps a percentage chance of 10%. Unlike sites like mysterybrand where you see what you're getting ahead of time, you don't know what items you are going to get until arrives, but you will know the percentage chance.

The crucial thing about this idea is that it turns the boxes into a game of chance and when dealing with games of chance on the blockchain, I wholeheartedly believe transparency is key. Drawings for boxes will need to be validated, using a pseudo-random number generator (PRNG) will probably be the likely choice if I go down this route.

The idea being you can verify your box contents were generated based on a generated set of numbers equating to products that can go into the box. Each draw will have a generated seed or allow the user to provide their own unique seed attached to the box.

The thing with random number generators based on my experience is poorly written implementations can be manipulated or hacked quite easily. There have been some cases where blockchain based games using random number generators could be exploited, usually because they were using poor implementations based on Math.random.

This idea is not set in stone, but it seems to be a decent candidate for changing the site and showing users what they could be getting in their boxes, as well as some mystery associated with said boxes.

There will also be a redesign. The current site was an MVP to get feedback and now the next step is to design something, possibly based on the future ideas around chance for items in a box.

Thanks again to everyone who has provided feedback, please keep it coming. In the interim, the site will continue to function like it currently does until other changes are released.