Piecemeal Experience

I've experienced a lot of friction lately over a piecemeal approach to delivering features, functions, and services to customers - driven by a development methodology that is based on the notion that individual components should be implemented in production as soon as it is developed, rather than waiting for an entire application to be developed.

This is not unheard of: the notion of "versioning" software necessarily entails delivering a product that has a defined set of features and functions, with the intention of shipping a product to market and earning the revenue to release an improved version that contains a bit more of what the firm had hoped to deliver in the first place, but hadn't the time.

Over the last several months - possibly for the past several years - I have seen sites and software products that roll out functionality in a piecemeal fashion. Rather than putting together a site that delivers its full value to the consumer, bits of functionality are incrementally revealed and the experience is slowly delivered. Version 1.0 is next to useless, version 1.0.1 is little better, but it's not until version 1.17.82 that it provides anything of value.

In theory, this could be done in a manner in which the core value is delivered in the first release and the "bells and whistles" are added later - which used to be common - but in lately, it seems to me that a crippled and useless version is released, and the missing elements are not needless features, but functionality that is required to deliver the core value of the product that is absent. I can't think of a single instance in which it was a good user experience.

The problem with doing this is that it's detrimental to the reputation of the product, and especially when it comes to Web site experiences, that can be deadly: if users visit a site and has a horrible experience, the association to the brand is made, and it decreases the chance users will return later, when the problems have been fixed and appreciate what the provide intended to deliver.

An analogy: imagine receiving an advertisement for a restaurant that promises the best cheeseburgers in town ... but when you show up, you are offered a slice of cheese and a wad of raw meat. "We'll have buns next week," the waiter tells you, "and we'll get a stove next quarter so we can cook that for you."

It's a completely ludicrous scenario for any physical product - nobody would do such an ill-conceived thing - but for user experience in the digital channels, it's becoming standard procedure to hand the customer a few disjointed fragments, with the promise that the rest of the features will be rolled out at a later time.

Some customers will tolerate this behavior, and that some portion of them actually will come back in future in hope of getting what they were promised in the first place - and this does little to discourage the practice. When site owners tally up the numbers, they count the users they are getting, but don't seem to consider those whom they have disappointed and will harder to get to return. Their first experience was that the site was useless, and that's likely to remain their impression even months later, after it's been rendered useful.

That's not to say that it's impossible to roll out partially useful sites and add capabilities later, but the initial roll-out must offer a valid benefit to the user. Consider the development of many e-commerce sites by catalog merchants: the initial roll-out was simply a product inventory; a second phase was adding the ability to compile a list of items; the third enabled the user to print the list and mail it along with a check; and the last phase added the ability to enter a credit card number and order online.

That worked, but only because the steps were in a certain order. If the version 1.0 offered only the ability to fill out an order form - without the ability to view and create a list of products - it would have been regarded as pointless, and it would have been highly unlikely anyone would have used it, or come back a second time to see if it had been made useful by the addition of other features.

It's also worth considering the tense: it worked in the past - not necessarily that it works today or could work in future. When there was a lack of competition, people were easily impressed. If a catalog merchant who sold shoes wanted to open an online store in 1995, they could have gotten away with this incremental approach because no-one offered anything better. Attempting to do the same in the present day would be foolish: there are already many options to the consumer that offer the full package and more. You might get a passing glance from people who are curious to see the new shop, but they would quickly recognize the lack of value, return to their current providers, and likely never drop in to see if you've managed to get your act together.

Even where the service a site offers is unique, it has to offer value. In the present competitive environment, if you roll out something half-baked, chances are a competitor that is more nimble or has more resources can recognize the value of the idea if it were done right, and roll out a better version in short order.

I have a sense there's much more to be considered on this topic, but I've rambled myself into a corner again.