People often underestimate the importance of SaaS marketing optimization.
Certainly in most cases, once a plan is worked out, some initial level of optimization is conducted, or else every marketing campaign in the history of SaaS would fail horrendously. Unfortunately, the cursory tweaking is usually all that is done. Sure, the marketing plans usually work out if the plans themselves are solid on a basic level, but they could work better.
Considering you must have a good positive ROI on your SaaS marketing strategy, optimization is a great way to work toward this goal. Unfortunately, how to optimize marketing in this industry seems like an elusive bit of knowledge to many, so with that in mind, maybe we ought to talk about some of the more important and basic tenets thereof.
First and foremost, we need to talk a little bit about demographics. I don't need to talk about how important demographics are to any marketing campaign, be it SaaS or otherwise. But, while you surely regard it as important, and pay a lot of attention to it, there's probably some optimization here you might not think of trying. Heck, I didn't think of this until a far wiser marketing guru pointed it out to me either!
There's something to be said for allowing some slack in relevance when targeting demographics. In SaaS, this is mostly in the purpose or industries your software may target, but this is a generality in marketing as a whole. See, if you find the relevant points that make this your demographic of choice, then you can expand on demographics this target may associate with, and relate to them on all or some of the same points that form the triad. This creates a larger outreach and potential user base, while still keeping focused on a general demographic set. It sounds odd, but it does work!
Now, with this retargeting of your demographics, your campaign should shape up a bit and not seem as oddly specific or out of sorts as it probably does, because a bit more generality of demographic will also bring general relation in the campaign as well.
Now, let's talk about testing a little bit. When testing, there are of course a couple phases you have to work through. Now, initial in-testing before doing anything external, there's probably nothing you really do wrong. This is pure scientific method and mostly, it's the programmers who are doing this testing, and they see the world in a unique and alien way, compared to that of marketing.
It's the external testing, or the Alpha/Beta phase, that we need to address here. This is where civilians come into the equation, and you can get real hard data on how potential users perceive the product. This is where big immediate flaws in functionality, design or UX overall should be spotted. Many software firms mess this up, SaaS and otherwise.
Here's the scoop. Everyone does this testing far too passively. The best way to see customer reaction to errors, or how well a scheme works when being used complexly is to put the test subjects in complex situations. Basic encounters only test the surface of the software. Have scenarios of significant complexity you wish them to test. This may mean being a bit more selective with your test subjects, but that is a small sacrifice to make. Also, with good tutorial software that can onboard with your SaaS, you can guide them through processes if need be.
Now, we come to one last place where everyone really messes SaaS marketing up. Metrics, statistics and logistics are easy to be overzealous with. They say you can't measure something too often, but "they" are wrong. You need to know the right time, pace and level of detail to take these metrics, otherwise you'll have too much, and numbers in statistics will be too diluted with fluff.
Consider which metrics are the most important, and consider how often to measure different ones. If you're more selective and synchronous with your measurements, you'll be able to see much more clearly, and avoid mistakes or working hard rather than smart in many scenarios.
So, there's a lot to be done with optimizing SaaS marketing, but none of it's terribly complex. Of course, it can go deeper than this, but you haven't time to read a lot of literature on this subject, and really who does? In that case, these are the big points to take away from the subject, in my very modest opinion.