On this week's episode of Inside Outside Innovation, we talk about why 70% of startup acquisitions fail, why UX didn't die, and how everyone is still building their startups backwards. Let's get started.

Inside Outside Innovation is the podcast to help innovation leaders navigate what's next. Each week we'll give you a front row seat into what it takes to grow and thrive in a world of hyper uncertainty and accelerating change. Join me, Brian Ardinger and Miles Zero's, Robyn Bolton. As we discuss the latest tools, tactics, and trends for creating innovations with impact, let's get started.

Podcast Transcript with Brian Ardinger and Robyn Bolton

Why Startup Acquisitions Fail: Learning Problems vs. Execution Problems

[00:00:30] Brian Ardinger: Welcome to another episode of Inside Outside Innovation. I'm your host, Brian Ardinger, and I have with me Robyn Bolton as always from Mile Zero. Welcome, Robyn.

[00:00:48] Robyn Bolton: Thank you. Great as always to be here.

[00:00:51] Brian Ardinger: We are excited to have you. Excited to get into the news of the day and some of the amazing things that we're hearing in the world of innovation.

We are going to start with the first article. First article comes from our friend Elliot Parker. Elliot is with Allied Partners. He's actually coming out to the summit, so not only are we going to talk about his article today, but you can come and see him live and in person April 13th. Let's now talk about his article, Why 70% of Startup Acquisitions Fail: the learning versus execution problem.

And Elliot talks about, first of all, he cites some statistics that large companies acquire startups at a 70 to 90% failure rate. Yet the same research shows that bolt on acquisitions, when you buy a company in the same industry that's doing similar work, the success rate climbs to 80 to 85% of the time.

And he poses the question, what's the key difference? The key difference really is the fact that you're really working in two different worlds. You're working either in a learning problem world, such as a startup, trying to understand who their customers are and what they're building, et cetera, or an execution problem world where you figured a lot of that out, and your job then is to efficiently scale and predict and move that business model forward.

And I think based premise is that large organizations oftentimes don't know exactly which startup they're buying. Are they buying a startup that has figured it out or have they bought a startup that's still learning. And then that integration is where the, it all falls down.

[00:02:12] Robyn Bolton: Yeah. I will continue the shameless plug. I am a huge Elliott fan. We've worked together, we've co-authored articles way back when together, and he is just a really smart, really great guy. So highly recommend everybody come and see him. Mob him at the IO 2026 conference, and again, he hits the nail on the head of learning problem and an execution problem.

It's different worlds innovation and operations are different. Pilots and scaling something are opposite problems. And the fact is big companies are designed for execution. I mean, I still remember my days at P & G when we were test marketing Swiffer Wet Jet, and our test markets were Canada and Belgium.

Those are countries, not test markets. But that's just how big companies are wired, and he makes a great argument backed up by facts around what the problem is and honestly, what companies need to do about it is kind of recognize that these are opposite things and I had to structure and approach the problems accordingly.

AI, UX Design, and Why User Experience Is No Longer Just About Screens

[00:03:25] Brian Ardinger: It'll be interesting to see how this plays out in the day when you can spin up a startup in five minutes and, and all the new things that are happening out there. How many large corporations might fall into that trap of looking for the shiny new thing and not realizing that it's not fully baked, and then it won't necessarily fit into the existing structures that they have and kill it from that perspective.

Or we'll get it to a place where you can build a startup and get to execution much faster, such that those acquisitions can dovetail right into an existing business. So it'll be interesting to see how that changes over the time period as well.

[00:03:59] Robyn Bolton: Yeah, and you know, will organizations, the failure mode I see most often is they think, oh, you know, there's market traction, there's revenue. The startup may even be profitable, and they think great. It's no longer a learning problem, it's an execution problem. So realizing that just because there's revenue, just because maybe it's even cashflow positive, doesn't mean it's ready for scale.

[00:04:20] Brian Ardinger: Absolutely. Alright. The second article is UX Didn't Die, it just stopped being about screens. This is from Nurkhon, if I'm reading that right. N-U-R-K-H-O-N. He has a medium article talking about this particular thing and he's basically saying that the skills that matter are different than what they've been in the past. So he's goes through an example where he asked Cursor to, you know, re redesign a checkout flow for a thing he was building.

It generated the perfect uI in 30 seconds, all the correct ratios, proper button states, et cetera. Then he showed it to three customers and they all abandoned the project at the exact same stop. The UI was perfect, but the problem was something else. And this gap between what it looks right, the functionality and what actually works is where UX value really lives in 2026. And that's an interesting thing that we're seeing more and more. What's your take on that?

[00:05:13] Robyn Bolton: I think it's another great example of what noted innovation philosopher Mike Tyson said that everyone has a plan until they get punched in the face. And this is another example of AI designing something that is perfect, but then what exposed to reality and that reality being the unusual, illogical, wonderful nature of human beings. It just gets punched in the face. It doesn't work.

So I was actually really glad to see the seven skills that he listed as mattering: systems, thinking, feedback, translation, there's judgment again, I feel like that's becoming a theme, pattern recognition, trust building. All of these skills are fundamentally human skills, and I think it's just another great illustration of how AI can't replace us yet.

Customer Discovery Still Matters: Why Startups Keep Building Backwards

[00:06:04] Brian Ardinger: It'll also be interesting to see from the perspective of, again, if these systems are, they're basically coming to commodity decisions. They look at everything out there. Yep. They find the best route to it and say, here's what the average says about it. Most people shouldn't be building average. You have unique customers with unique problems with unique environments around that. And so at the end of the day, that's still your job as a UX UI designer or a business o...

Learning vs Execution with Brian Ardinger and Robyn Bolton