2024 has not been kind to technology, but it’s been outright rude to tech startups.
When you connect the dots, the issues that have been plaguing tech startups and even startups in general for the last 18 months or so have most definitely been rooted in the decline of the strength of the SaaS market.
3 Top Market Shifts Affecting SaaS
- Artificial intelligence
- Overused subscription models
- Rising processing costs
Now, if you’ve been following my diatribes against SaaS over this period, you might think I’ve got something against SaaS. I do not. I’m a fan of SaaS. I just see the way the market is changing.
These market changes were happening under the radar about a year ago, but, after a number of conversations over the last several months — with CEOs, with investors, with customers — I think four concepts are starting to cement in a way that anyone running a SaaS startup, a tech startup or even just a startup that uses or engages with technology platforms, is going to have to understand.
So here are the thorny problems that snuck up on SaaS. They will continue to manifest themselves, so I’ve been doing everything I can to counter them, and I’ll give you a little insight into my thinking along the way.
Technology Fatigue Is Real
And it’s accelerating.
Most people, especially technical people, get caught up in the definition of SaaS as “software that runs in the cloud.” This is a little like saying that cars have wheels and an engine. So trucks are cars. Airplanes are cars. Lawnmowers are cars.
A better definition of SaaS is that it’s “software for the people.” Stay with me on that.
A lot of folks discount the fact that the rise of SaaS was not only correlated to, but in my mind, caused by, the proliferation of a computer within everyone’s reach, and by everyone, I mean all business people and all consumers. Every. One.
Once everyone was using a computer during almost all waking hours, for business and personal reasons alike, software had to not only be instantly accessible, but it had to be slimmer, friction-free and elegant as all hell.
So there are two things I say these days that very much take that second definition of SaaS into account.
One: Customers don’t buy technology, they buy the benefits of technology.
This is very much coming to a head in the sense that software no longer should be all things to all customers. Software that tries to do too many things for too many people is being rejected as bloated, expensive and, well, yesterday. Niche is in, and niche is making a comeback.
Two: Users don’t want more screens.
Users still expect the same functionality from fewer screens, with less input required and more focus placed on the immediate task at hand. I was getting raked on this opinion as recently as six months ago, but now people are hopping on board.
AI Directly Competes With SaaS
You can say what you want about AI. As a long time player in AI and one of the progenitors of generative AI, I say some good things and many bad things about it.
Here’s a good thing. AI is better at input and output than humans. If generative AI is making headway, it’s in UX and UI, especially where input is involved. AI can contextualize input better and faster than we humans can, and it can detect patterns and insights in output data much better and much faster than we humans can.
Here’s a bad thing. People are damn sick of AI. If technology fatigue is real, AI fatigue is probably the primary reason. Customers are not just unimpressed by AI now, they’re wary of it.
And here’s how it’s hitting startups. An outsized proportion of technology investment money went into a greater fool’s chase around AI. If you’re wondering why it’s so difficult to get seed or series A money these days, well, there are a lot of reasons, but a big one is that VCs started chasing their crypto and pandemic-recovery losses by panic-investing in anything that had AI stapled to it.
And as I mentioned above, this was investment made into a lot of AI solutions that nowhere near fit the primary AI use case.
Subscription Models Were Overused and Abused
I wrote a piece several years ago pointing out that not every pricing model needs a subscription option. The model ends up becoming a hindrance to either the customer or the company, by either providing too much or too little value, by adding features and functions that weren’t necessary, at the expense of features and functions that were.
Again, trying to be all things to all people. But it’s more like bloated cable television packages than Swiss Army knives.
And then I wrote another piece about how subscription-based marketplaces were also going to suffer because of an overload in the market. In a lot of cases, subscription models were being used to justify the offering of the marketplace, but too many vendors were low quality and customers were catching on.
A better way to handle a lot of commerce that’s being done under the subscription model now could, ironically, use a hand from data science and automation and AI. Predict when the customer needs the product/service, inform them in a way that explains the need and offer concisely, then make it one-touch to confirm. Like an individualized subscription. That’s actually something I think is going to work.
One problem with it is that it breaks the rhythmic recurring revenue paradigm, which investors and boards still require.
The Cost of Processing Is Catching Up to Companies
Too many companies — and not just SaaS or even tech companies — are losing money on fulfilling SaaS processing costs on a per customer basis.
I’ve got guesses as to why this is the case.
- It could be blamed on equally abused freemium model strategy, one in which the paid tier doesn’t offer enough value to compel customers to pay for the processing they’ve already grown accustomed to.
- It could be that customers are wary about overpaying for processing requirements for their unique use case. In other words, they don’t want to pay for someone’s else’s processing time.
- It could be AI processing driving up the costs of all processing.
- Maybe it’s just inflation.
And again, in a general sense, there is some all-things-to-all-customers issue in this as well, as startups try to capture larger and larger shares of markets without a decent sense of CAC and LTV of those new customers.
The Overarching Strategy: Less Is More
As these problems deepen and trickle down, I’m seeing a need for less is more in the development of SaaS and its usage as well, and even the use of more general technology solutions.
I often talk about elegance in UX and UI — fewer screens, fewer inputs, less function, more focus. When I talk about that elegance, and how important it is these days, and how that results in the need for equally elegant software, and that not enough companies are putting an emphasis on it when building new tech, this is what I mean.
