Build vs Buy of SaaS tools: what SaaS teams need to know

The biggest debate in SaaS is: Should you buy or build your own tools? Proponents of the build side are saying that you have enormous cost savings by building the tools. Opponents are saying that it's risky and you will not be able to maintain it. At Revenue Wizards, we are primarily on the buy side.

Since I started my year in tech, I was often just surprised by the margins of some SaaS companies. Some have margins between 60% and 90% or more. Of course, many companies invest heavily into growth, so those margins are shrinking, but on its own, every new user is just a click of a button on average. It's pretty much free in terms of technology. If you are a rational human being, you're wondering: "Okay, this does not really make sense and it's quite against any economic theory." I'm wondering why companies are paying for that?

They pay for value, perceived value. A CRM is making the sales organisation way much more efficient compared to using an Excel spreadsheet. A marketing automation tool helps them to automate things. A CPQ helps with quoting. So many tools help processes to improve them, so they often argue that they can spend more time on high-value activities to make more money. That is why companies pay the premium.

In the Buy and Build debate, this paradigm gets challenged because many people are empowered to build an equivalent tool or a tool that fits their use case by themselves at a fraction of the cost. That means that the value argument of the SaaS provider is eroding. There is not much value in paying the SaaS premium if you can build it yourself.

To me, this is a no-brainer, and more companies will be building. In a recent survey of 123 warehouse professionals, we asked, "Are you building custom tools or not?" 61 people said that they are building. That's a lot, and my assumption is this will only continue over time as OpenAI and Anthropic are making it easier to build.

What should you build?

There are dozens of go-to-market tools that you can build, maybe a partnership platform, a commission calculator, or even a CPQ tool. See our blog on it.

If you build dozens of go-to-market tools, you can save anywhere from $140,000 to $340,000, $400 on SaaS licences (that would be for a company with 20 resellers and users). We built a calculator that shows you how much you can save on licensing fees. 

You can find it here.

What do you need to build?

You will need to have the skills in-house to do this build. Tools like Lovable, Claude Code, or Codex from OpenAI help you build those tools. Even business users are nowadays empowered, but it's recommended to have experienced operators who can build it.

I would go even further and argue that you should even invest in proper engineering resources that are becoming part of the Revenue Operations team. I don't mean just a regular go-to-market engineer that does not have coding skills, but really a proper developer. Those people are going to save you a lot and make a very reliable tooling infrastructure for your commercial team that will give you a lot of growth opportunities. 

Probably, for the higher the team size, you won't really invest into an engineer. Smaller team sizes, let's say under 20 employees, probably can do it. At 20 sellers, you can do it just with RevOps resources. Should you need help building here, feel also free to reach out to us.

Common objections

There are objections from the opponents, and let's go through them and provide our view on it.

First, security.That argument is fading fast. Low-code, no-code, and cloud platforms now handle the heavy lifting on proper security setup — and plenty of GTM tools don't even need that level of lockdown in the first place.

• You can make it SOC 2 compliant.

• You can self-host on your own infrastructure.

• You can wire in SSO across all the relevant features.

  
  

Second, Departures. People are worried if your talent is leaving, who's going to manage the platform. That is a valid concern, but it's also getting less. The newer models of the LLMs are extremely good at setting up proper documentation so that if anyone new joins, they can take it over simply. In addition, this documentation helps the LLM to pretty much self-manage it, and this is where the future is going: it can be continuously improved by itself with the right prompts.

Finally, as mentioned earlier, you should be investing in a proper team that can maintain those tools. This is really your competitive edge over the long run.

Third, complexity. The other objection is that some tools are too complex to self-build. There is some truth in it, but everything can be built yourself. Companies like Coca-Cola and Booking.com have built their own CRMs before there were even AI technologies. Of course, those companies have massive engineering teams, but even else, there was a whole market of custom tool developers that developed tools for their companies and maintained those. This is nothing new. Everything can be custom developed, including your ERP, CRM, or Marketing Automation Platform. Yet you should probably not build everything, which we will discuss next

What not to build (yet)?

There are three tools that I would suggest not building for now, which are an ERP, CRM, or the Marketing Automation Platform.

Those three tools are systems of record, which are highly integrated with other platforms, and it can get quite complex with custom processes. Software providers like Salesforce, HubSpot, Xero, NetSuite, and the like invested millions, if not billions, in perfecting those tools. Having a custom-built one can maybe work for smaller companies, but if you're any mid-sized SaaS or bigger enterprise, building this yourself is a heavy lift. It's not impossible, but it's a very heavy lift, so rather start by focusing on the tools that are simpler and have better value. Simple doesn't always mean cheap, right? The CPQ is expensive, but it can be rebuilt, so builders, let's go out and build.