VCR: MVP vs. MMF vs. MMP vs. MBI

"Men follow their sentiments and their self-interest, but it pleases them to imagine that they follow reason. And so look for, and always find, some theory which, a posteriori, makes their actions appear to be logical."

--Vilfredo Pareto

Minimum Viable Product (MVP), answers the question would this product be workable or feasible in my intended market. A minimum viable product can be comprised of static (or fake) data. You’re checking to see if the concept as a while would comprise a viable product in the market place. If you have a customer already and the product is defined then you don’t need to waste time on MVP. If you’re testing the waters—would my potential customer pool want this product—then by all means create an MVP as quickly and as cheaply as you can. You can’t sell an MVP because its smoke and mirrors, but you can decide whether to a build a product based on the interest in your MVP.

Minimum Marketable Feature (MMF) is the smallest number of features that would comprise customer value. For example, given a property and casualty (P&C) rating system the ability to provide insurance quotes let’s say with a rating engine for auto insurance—with a REST or GraphQL service—would be a minimum marketable feature. You have one line of business and a feature—rate auto insurance—and you could sell that as a SaaS product.

Minimal Marketable Product (MMP) is a product developed with minimal effort to make money. This might be the approach you take to begin generating revenue as cheaply and as quickly as possible. An MMP might be comprised of the smallest number of MMFs necessary to make money. For example, (re Property and Casualty rating) General Liability (GL) would be cheaper to build than Auto. So, GL rating as a service might be MMF1, but without policy generation GL by itself (because its relatively easy) might not be marketable. Thus, the ability to render a policy for GL (MMF2) might be necessary to go to market. Still cheaper than auto by itself, and a marketable product.

Minimal Business Increment (MBI) is the smallest release of a product that is consistent with strategy. An MBI is an increment of an MMP, or MMP?, that adds value and reinforces that you are on the right path. Again, given our MMP with General Liability rating and policy client tools an MBI might be to release Auto rating and auto client tools. Assuming the success future MBIs would include other lines of business, like Fire, Commercial, or Marine.

It is up to the business to figure out which approach is the right one. For a startup, MVP may get you past capitalization and moving to a positive revenue stream as early as possible may make MMF or MMP the next appropriate approach. MBI is the sustaining business strategy. (Think Apple iPhone.) For an internal application, viability may be moot. An MMP and MBI approach might be the right choice. Regardless, Value, Complexity, and Risk must be considered.

Value, Complexity, and Risk (VCR) is an assessment of a feature or product’s value in the market place, and a technological complexity and a risk mitigation assessment. Continuing our P&C example, General Liability is so trivial that it may not provide substantial value even if it is easy. It certainly doesn’t mitigate risk or address complexity. If we switch to a more complicated example, the VCR approach becomes more apparent.

Our product is a time machine. MVP, MMF, and MMP only work if the most valuable, most complex, and riskiest element is addressed first. You can’t mock time travel, but an MVP could be to send a hamster ten seconds into the future. An MMF might be to actually transport something to and fro—back to the future. Traveling through time is actually the most valuable, the most complex, and certainly the riskiest part of our product.

Degrees of VCR may not be so dramatic as time travel. You could do P&C GL first and then CL with the GL release being sort of meh. Where risk mitigation plays a role is that if you do easy GL first you still may not be able to do the much more complex commercial or auto.

From a technology stand point working backward from high VCR to low VCR requires the most patience but yields the most consistent and reliable result. Low VCR to high VCR tends to produce the most mis-leading results. Euphemistically we call this low hanging fruit. By picking low hanging fruit everything looks easier than it may be. Low VCR to high VCR redresses impatience but most often results in the least consistent and least reliable result.

Business decides what is valuable, and they may have some insight into to what represents risk. Architecture decides what constitutes the greatest complexity and how to mitigate risk. A high VCR approach always yields great results, if not the cheapest and fastest. A low VCR, might appear to be cheaper and faster, but success or failure will not be revealed until you tackle the tough stuff.

?

Paul Kimmel is a software architect who has delivered working software on most of the software platforms of the last 30-plus years. The VCR rule was first more than ten years ago. “Low hanging fruit” tends to be a common managerial anti-pattern that most often leads to failure.

?

Roman Pichler’s website at romanpichler.com seems to have many excellent blogs on these subjects and much more.