Wrong Way (for ROI). Source: Pixabay.com

Time: Why Your ROI Case Is Wrong – Part One


In theory, calculating the Return on Investment (ROI) for improvement projects is simple: take the amount of costs you’ve saved, add it to the amount of new revenue you’ve generated, and divide the whole thing by how much you invested to get those results. When the project is complete, our job is a bit easier. We have actual cost outlays and real-world results, assuming we’re keeping track of all of the right information and doing the right calculations. Of course, we may still calculate ROI elements differently than our peers. And that’s where everything starts to break down. One of the big culprits affecting the accuracy of ROI calculations and projections is time. Specifically, I mean clearly defining the time period for which costs will be tallied and savings will be measured. That’s what we’ll cover in today’s post.

Accuracy: It’s About Time
Imagine two competing software investments, one offered on-premise with maintenance and the other hosted in a SaaS model. Here’s a straight-forward example with some basic rules: (1) both solutions produce the same amount of savings and revenue generation; (2) the on-premise solution has a higher up-front cost with an annual 10% maintenance fee; and (3) the SaaS solution has a consistent price over time. The incremental savings and revenue generation shrink over time, in the time honored 80/20 tradition: year one is the full amount, years two through five are 50%, and years six through ten are at 25%.1

[NOTE: This example pits an hypothetical on-premise solution against an equally hypothetical SaaS alternative for illustration purposes only. It should NOT be read as a comparison of on-premise vs. SaaS solutions generally!]

Year One ROI: A Clear Winner!

Investment Option Amount Saved New Revenue Generated Cost of Solution ROI: Year One
On-Premise w/ Maintenance $100,000 $50,000 $55,000 273%
SaaS $100,000 $50,000 $12,000 1250%

At the end of the first year, it seems obvious which choice was better. Both on-premise and SaaS had respectable performances, but only one really shined. Why is that? It’s because we’ve failed to account for the different cost structures of the competing options. One incurs the majority of its costs up front, with smaller pieces in future years. The other is higher annually, but does not entail a large up-front investment.2  The error of basing our ROI calculations on just this time period is that it will give us a false picture of the project’s true costs and benefits. This can happen not only with different pricing structures, but also with different trends in savings and revenue generation. In any scenario where the early periods (weeks, months, years) differ from the later ones, only looking at the short-term doesn’t tell you the whole story.

Year Five ROI: What’s Going On Here?

Investment Option Amount Saved New Revenue Generated Cost of Solution ROI: Years One through Five
On-Premise w/ Maintenance $300,000 $150,000 $75,000 600%
SaaS $300,000 $150,000 $60,000 750%

Hmm. It’s the same two alternatives, achieving equal amounts of savings and revenue generation, but the results are much closer. I know that, to some of you, this may come across as too basic. If you’re already dismissing this with I know, Scott, you’re just explaining Total Cost of Ownership, then… you’re exactly right. I can’t pull anything over on you. For the sake of your business and the market in general, I hope you had the same reaction the last time you read an ROI case study. Now, if TCO is a new concept for you, I think you’re probably seeing where this is headed. Now that we’re a few years farther on, the cost differences begin to even out. What looked like a lopsided contest at first is much more competitive. If nothing else, Year Five shows us that we were right not to take the early returns and extrapolate from there. That would have led us to the wrong conclusion. So let’s jump ahead a bit and finish the journey.

Year Ten ROI: TCO In Action

Investment Option Amount Saved New Revenue Generated Cost of Solution ROI: Years One through Ten
On-Premise w/ Maintenance $425,000 $212,500 $100,000 638%
SaaS $425,000 $212,500 $120,000 531%

Now that’s interesting. We’re a bit older and grayer now, and assuming we’re still keeping an eye on things we see that the tides have turned. Ten years of annual SaaS costs have finally added up to more than the initial on-premise cost plus annual maintenance. Without any performance differences to separate the solutions, the only difference in our equation is the denominator — and on-premise pulls ahead. That’s the power of TCO. As basic as the concept is, it’s also incredibly important, both when assessing your project’s performance or projecting your expected ROI for future investments.

A Quick Wrap-Up
A lot more than just software/service cost can vary over time. Can you realistically expect year-over-year savings to stay the same? To increase, perhaps? How about new revenue opportunities? I have two take-aways that I’d like to leave you with, depending on your perspective:

      1. If you’ve already completed your project and are taking stock of the ROI, please take care — especially if you will be sharing the numbers with your peers. Be clear about what time periods you’re discussing. Make sure that they’re fair choices, accurately matching costs and benefits across the expected life of the solution. If you don’t have the benefit of years of hindsight, then be mindful of trends and how the elements of the ROI calculation can change over time. It will give you a more accurate picture of your own success, and it will go a long way towards giving others an accurate basis for comparison.
      2. If you’re evaluating technology or service investments, don’t take those published ROI numbers at face value. Dig into how they were calculated, what assumptions they made, and where they may have gone astray. If you can’t review the methodology, then build your ROI case from the bottom up. That is, instead of expecting to achieve the ROI results others claim, map your process flows and cost components and be detailed about what aspects you believe will change post-investment. Will you reduce transaction costs? Will you trim labor count? If so, by how much — all in the context of your own organization.

I think we all have the same goal when it comes to accurately assessing the benefits of software and service investments. Having accurate projections helps businesses make smarter investments. Inaccurate projections can lead to missing project goals, which hurts both the corporate practitioners who championed them and the solution providers who delivered them. Neither job advancement nor customer satisfaction are positively impacted when projects fail to meet expectations. Sometimes it’s unavoidable, but basing an investment on questionable numbers is something we should all endeavor to avoid. Deal?

Until next time,


1. This “diminishing returns” perspective on technology investments is realistic, but is not always reflected in the ROI case studies you see in the market. Be careful when reading papers that attempt to project Year One returns out into the future. When the low-hanging fruit is picked, it takes greater time and effort to reach those higher branches!

2. We’re mainly looking at license costs here. It is absolutely true that SaaS solutions can entail up-front consulting, development, and customization costs too, just like on-premise solutions.

Leave a Comment