Multi-Seat Pricing, Software-as-a-Service and the Decline of the "Hosting" Industry


There’s been a lot of discussion (especially amongst “old timers”) about our business model change at the end of 2014.

If you’re looking for just “the facts” you should read this other post first as it’s a much more focused piece that breaks the discussion down into a Q&A format. You can read that here: Your Multi-Seat Pricing Questions Answered.

However if you’re interested in more of a stream of consciousness history of our business model(s) and a bit more color on the whole story, then this post is for you.

Back in October (of 2014), we announced that simultaneous to the release of Miva Merchant Version 9, we’d be changing the way we charge for Miva Merchant licensing. To sum it up, we shifted from a model where we previously charged for “hosting” and gave Miva Merchant away free with our own hosting plans, to a model much closer to the industry standard for Software-as-a-Service and charge based on Simultaneous Administrative Users (Backend Users). We now include a base level of hosting free with the use of Miva Merchant and you can add more hosting resources if you need them.

At the time we examined the other models that are both prevalent in the SaaS and/or ecommerce world and decided the per admin seat model was both the best and most equitable for our client base (other models we examined but decided against were percentage of sales model, a page-view model or radically increasing the base price, a la Magento Enterprise which costs $18k annually without hosting).

While this might not seem like an important change, or worse might seem like a bunch of semantic hooey to justify a price increase, it’s actually a very important foundation and precisely explains why we needed a price increase and more importantly a pricing model change.

When Miva Merchant first came out in 1997, a customer would buy it outright, for roughly $500. The business model back then was Miva would release a new version every year or two and hopefully the customer would upgrade and pay another $250 on top of the $500 they paid for the first version.

There were a number of problems with this model both for Miva specifically (upgrading Miva back then was difficult, just as it still is for most other distributed ecommerce platforms today) and for the software world in general as this model of selling software was on the verge of being disrupted by Software-as-a-Service.

So in 1999 Miva changed course a bit and also started selling Miva Merchant bundled with hosting (via third party hosting providers). The model was identical in theory to how Windows was being sold at the time. Microsoft would sell Windows in bulk to a company like Dell or HP and part of the agreement was they had to include a copy of Windows with every computer they shipped.

Miva tried to emulate that model with hosting providers whereby hosting companies like or Earthlink would buy large bulk purchases of licenses and then had to bundle them with every ecommerce account sold.

This model sort of worked at the time; Miva got a lot of market share out of the arrangement, and the hosts got a high value product to use to differentiate their hosting. It also helped keep the price of their hosting from being undercut by competitors by offering premium software as a differentiator.

Fast forward 15 years later and this is still in its core form, the same way we’d been selling Miva the whole time, except that we’ve transitioned to being the primary hosting provider as opposed to selling it through third party hosts.

Essentially, you purchase a $50/month plan; get Miva included with the plan for “free” and a little bit of disk and bandwidth. If you need more disk and bandwidth it would cost you more money, but Miva was the same price regardless of how you use it, what features you use, how big your store is, etc.

Not rethinking our pricing sooner, caused some odd long term unintended consequences. We found our sales people explaining why we have so much less disk space than GoDaddy, BlueHost or HostGator (all commodity “low end” hosting providers). This is what I’ll describe as pricing problem number 1.

Those routine conversations got so ridiculous, where people were stuck on the price of disk or bandwidth compared to other commodity hosting providers (and for the record all disk and bandwidth aren’t created equal and there is a big difference when it comes to performance but that’s for another blog post) and not at all considering the actual value of what they were looking at purchasing, namely the underlying platform that is going to power their online business, that I personally realized we had to rethink our pricing strategy from the ground up.

The other major unintended consequence was from the success we’ve had with our long-term existing customers. We had grown our company while they grew their companies.

Our average customer uses Miva for close to 9 years and the industry average for customer churn is roughly 2 years. This difference between the norms in the industry and what Miva customers accomplish is a true distinguishing factor when it comes to Miva and the competition, but it ended up making our pricing problem more painful.

Since 2007, we’ve been heavily focused on keeping our platforms capabilities ahead of our most successful clients growth to ensure they would never wake up after succeeding and have to re-platform because they’d accidentally outgrown us.

Not only do we not want to lose customers out of the top, but from a customers point of view, a re-platform is scary at best and a downright risk your business proposition at worst.

To add some perspective to this evolution, in 1997 if someone had tried to process 10,000 orders a DAY via a Miva store, they would’ve failed catastrophically.

For that matter today, if you try and process 10,000 orders a day on most (if not all) competing solutions that start with a street price of $50 a month (or less), the odds of catastrophic failure are incredibly high.

Yet with Miva Merchant in 2015, processing 10,000 orders a day is not a problem.

You can apply the same logic with SKU count as well, most platforms start running into issues when SKU counts hit into the many thousands to tens of thousands, yet compare that to Miva where we routinely test all releases against a data set of 1.5 million SKU’s and never see any performance issues related to SKU count.

All the while, if a customer could achieve 10,000 orders a day with 1.5 millions SKU’s while only using limited disk and bandwidth, we’d only charge them $50 a month.

To make matters worse, if they were hosted elsewhere, then regardless of disk and bandwidth we’d only charge them $40 a month, which is where the second problem is discovered.

The economics of our business model from the last 15 years simply don’t work in a world where Miva’s focus and goal is keeping our customers long term while helping them succeed.

One irony is, the old model could work fine if our resources were all poured into sales and marketing (like our Shopify, Volusion, BigCommerce style competitors) because in a world where your average customer only lasts 2 years and they only do a few thousand dollars in sales, you don’t need a great product, you just need a passable product.

So enough with the theory of why we needed to make a change, here’s some information about what’s actually changed internally at Miva since 2007:

  • We’ve grown our staff almost 20 fold since 2007
  • The budget to run the company since 2007 is up almost 1000%
  • We’ve gone to 24/7/365 support and none of it outsourced, from only offering outsourced Monday – Saturday normal business hour support back in 2007
  • We’ve successfully launched 5 major releases of Miva Merchant all the while never forcing our customers to rebuild their site in order to upgrade
  • We host approximately 75% of our clients (and that number keeps growing as a percentage, we expect long term to host roughly 90% of our clients on our network)
  • Our hosting architecture is built on top of the absolute best hardware, network and infrastructure money can buy and our uptime and site performance reflect those investments

In addition to everything that’s changed in how we run Miva, we haven’t taken any outside investment money and so we’ve had to power all of this growth with the income we directly generate.

When you add up the pricing issues and the growth of the company combined, something had to change in order for us to keep succeeding for our clients. 

The fundamental truth is that it’s simply not possible to invest millions of dollars a year into maintaining, securing and growing a scalable ecommerce platform when for all intents and purposes regardless of how the end user uses the product we had capped our potential income for the intellectual property at around $500 a year.

One response that has come up in regards to our business model change is “that’s fine for the new clients, but why didn’t you grandfather the existing client base with your old pricing?”

Here’s the problem with that request, we’ve invested all of our time, thinking and money over the last 8 years into the existing client base as opposed to getting new clients (which is the exact opposite of how our major competitors run their businesses).

Our focus is almost singularly on the success of our existing clients.

We spent the last 8 years of blood, sweat and tears (not to mention somewhere in the neighborhood of $50 million) building the best product in the market so that the largest Miva customers today can push $100 million in annual online sales through a Miva store. More importantly our average customer sells roughly $650k a year online, which is orders of magnitudes higher than our competitors like Shopify, Volusion or BigCommerce. Their average customers are at best in the low 10’s of thousands of annual online sales.

Therefore allowing our existing customers to be grandfathered in to an old and fundamentally broken pricing model was simply impossible as it would have completely defeated the purpose of this pricing change.

Another common response to the business model change is “I didn’t ask you to build all of these features in, so why are you charging me for them?”

The new business model fundamentally solves that dilemma. If a Miva user is truly using Miva in the “old 2007 way”, where it’s basically a catalog manager and checkout, then your price shouldn’t change one bit due to this pricing model change because in those circumstances you will not need more than 1 simultaneous back end user.

However if you are using all of the latest features and have multiple employees needed in order to maximize the potential of these features (while likely eliminating other costs such as order management systems, reporting software, fulfillment connectors or even CRM), then the value proposition and return on investment should be clear with our new model.

There are a couple of items we hadn’t fully worked through when switching to this model that we’re still analyzing and working on solutions (if needed).

The most obvious one is what we’ll call the “Developer Seat”. Developers are like the bees who pollinate the crop of the Miva community (either module developers or site builders and maintainers) and they uniquely have the need to be in lots of different stores but generally not more than one at a time and not for very long in general.

We’re working on a new Certified Developer Program that will allow our Developer Community to work around this issue without the need for all of their clients to buy them a seat.

Second is what we’ll call API users. This is products like Synchro, or ShipWorks, or any other product that logs into the back end and pretends to be a human user. The basic issue is we can’t tell the difference between a human and a computer the way the system is setup today. We’re still examining the best long term solution to this issue, but we’re aware of it and keeping an eye on it.

This new pricing model properly aligns Miva with the success of our customers, providing a proper pricing structure for Miva as we add new features designed to assist our customers and their employees, without forcing out the low end by pricing them out of the market.

Our goal is to be able to offer a platform affordable for a real business just starting out online who has the room to grow into an online powerhouse without ever having to worry about their underlying technology and knowing they have a partner in success through the whole journey.

We believe these changes are exactly what’s required to both continue fulfilling on that goal and to speed up the pace at which we can serve our customers and help them achieve success.