“Homegrown martech is like homemade pizza.”
That’s the kind of quip I might have likely shot back a couple of years ago if someone asked me, “Should we build our own martech software or buy a packaged app from a martech vendor?”
It’s not that you can’t make your own pizza at home. Anyone can. I have. But making great pizza at home? Eh. You need a combination of skills, ingredients, the right pizza dough, perfectly tossed and stretched, and proper equipment — in particular, a pizza-sized wood-burning or coal-fired oven that you can get hot enough and a long wooden paddle you can shuffle the pizza with to evenly cook it.
And then a heck of a lot of practice.
Hey, some people do it. If you’re one of those home pizza chefs who regularly dazzles your family and friends with your magnificent, original pies, hats off to you. I acknowledge that you can craft something unique, tailored to your exact tastes.
But for the vast majority of us, it isn’t worth it. Pick up the phone, call your local pizzeria, where they’ve built a whole business perfecting the art of pizza-making at speed and scale. In under 30 minutes, you’ve got a delicious pepperoni and mushroom pizza on your kitchen counter, in a festive white-and-red box, with hypnotic, mouth-watering aromas wafting into the air. Just picture pulling that first perfectly-shaped, thin-crust slice to your plate, with a few strands of golden, melted cheese being drawn out from their heavenly mothership… ahhhh…
I digress.
There’s still truth to that metaphor today. But it’s more complicated.
You probably shouldn’t be building your own CRM platform from scratch. And certainly not your own web server or email delivery engine. There’s so much standardized capabilities that you can affordably buy off-the-shelf, documented, debugged, optimized, maintained — don’t underestimate the importance of who’s maintaining a large piece of software.
But “don’t build your own web server” is an apt example to dig into. Because while you shouldn’t waste time and money hammering out your own HTTPS protocols or cached file systems, you absolutely must build your own website on top of those commoditized foundations.
The information architecture, content, functional web app logic, etc., of your website are all unique to your business. You have to build that — and continually evolve it to keep pace with the ongoing progression of your business, customer demands, and competitive market dynamics. It’s what makes you, you.
Obvious, right?
But it’s more than your website now. As companies go through digital transformation — which, to sharpen the meaning of that beyond a hackneyed buzzword, I’ll define as having your entire business fluidly orchestrated with software — it’s not just your website that must be custom-tailored.
The whole way your back-stage business operations and front-stage customer experiences are designed and optimized — and inexorably entangled together — becomes unique to your business too, across every channel and touchpoint. (This is the context in which Big Ops is becoming a big deal.)
As with web servers for websites, you should fully leverage off-the-shelf platforms and apps that give you the immediate, foundational capabilities to implement those business-specific processes and experiences. Don’t reinvent the wheel.
But that “implementation” almost certainly now includes unique software logic.
Granted, a lot of that logic may be implemented with no-code features. The superpowers that are available to marketers now for making workflows, apps, bots, dashboards, etc., without “coding” anything are pretty remarkable.
However, for all the wonders of no-code, many digital products and services still need to be built with code, by expert software engineers. Or, increasingly, have key components built with code, which are then leveraged within no-code tools that let you flexibly rearrange business logic around them. (Definition reminder: with digital transformation, business logic = software logic.)
So set aside the code vs. no-code distinction for now.
The point is that, either way, you are building custom “software” to build your business.
But you want to be strategic about it. You don’t want to build things you can buy off-the-shelf. Ideally, you really want to buy open platforms that give you both a ton of standardized functionality out of the box, but also expose data and services through APIs that let you accelerate the development and integration of the components unique to your business that you must build.
This shift from “build vs. buy” to “custom apps & ops on a common core” is part of the Second Golden Age of Martech that I described a while back. (Here’s a follow-up post on how platform dynamics enable this.)
“Okay, Scott. Nice concept. But do you have data that this is actually happening?”
Why, yes. A recent report from MarTech Alliance revisited the age-old question of suite vs. best-of-breed. (Note that almost all “best-of-breed” stacks now revolve around platforms.)
Not surprisingly, “best-of-breed” is 3X more popular than “suite.”
But the most interesting finding was that the most popular option was “hybrid” — a blend of in-house/custom developed and vendor solution(s).
To my knowledge, this is one of the first times this question was asked with a hybrid option. Kudos to the research team for asking it, as it reveals a major trend that has clearly been building.
To close out with an overextended metaphor. Martech isn’t like pizza. It’s more like a trip to the grocery superstore, where you buy a mix of prepared foods and raw ingredients. You combine them however makes sense to enjoy the best possible meals, in a way that balances time, talent, comparative advantage, and the experiences and outcomes that make you happy.
P.S. Flatbread != Pizza.
We’ve actually seen companies build their own email engines because at the scale they send emails (billions) off the shelf solutions become extremely expensive. One of the key factors in the buy/build decision is long-term costs and scale.