Cobrainer is a machine-learning data analytics company focused on the analysis and visualization of expertises in organizations.
What started out as a project for academia in Munich, Germany soon found traction in project management, enterprise product collaboration and innovation management.
Now, their clients range from Deutsche Bank to Boehringer Ingelheim. Although Cobrainer co-founder Amelie Spath had worked for large enterprises before, she had to learn how to target enterprises from the outside to push the company forward.
I caught up with Amelie as her team is re-focusing their enterprise product and preparing for scale.
I got a lot out of our discussion. In it, you’ll learn:
- How Cobrainer transitioned from being an academic project to targeting large enterprises
- How they got their first customers and how they motivated early adopters to work with them
- How the business got its start bootstrapping and the challenges they faced
- How they conducted problem interviews to identify the best customer segments
- How they raised funding and pivoted the product to be able to build a scalable business
- How they focused on understanding the enterprise reality and the impact it had on their growth
You can listen to the full interview below:
Etienne Garbugli (EG): How did your team go about selecting a product/market opportunity initially?
Amelie Spath (AS): Three of my co-founders are researchers. We always found that for any research project at Universities you need an interdisciplinary team. How do I find the expertises I need to make my project a success? There was really no mechanism to learn from past projects, which expertises they combined to make it work and which people have those expertises.
My co-founder, Martin, who is a machine-learning scientist said, well that’s a problem we could actually analyze with technology. And this is how we started out analyzing all the research projects at our University here in Munich and creating an expertise map. And then we pitched the whole thing at a conference where there were a lot of enterprise managers from large companies in the audience. And one of them told us: “Hey, what you’re doing for academia is actually really interesting for the enterprise… but they’d pay you money. So, why don’t you pivot and go in that direction?”. That was pretty much the start of the company.
EG: How did you go about validating the opportunity?
Amelie Spath (AS): This manager became our first customer and then we found a second customer, and with these two, we started developing the product. There was a lot of collecting requirements, understanding the enterprise context, getting feedback, iterating with the users and this is how we got started. We built the first version of the product with paying customers from the start. They took a big leap of faith.
EG: How did you motivate early adopters to work with Cobrainer initially?
Amelie Spath (AS): So, they really had this problem of not knowing which expertises were in the organization, how to best leverage all of this implicit tacit knowledge that is there. But employees cannot access this (data) and there’s no real strategic way or no direct way to collaborate with people you don’t know in your organization. They realized that with digitization, new business models and shifting dynamics in the industry, they really had to do something. They could also position themselves as thought-leaders trying new technologies like machine-learning. So, they really felt like okay, we take a leap of faith here but it’s extremely relevant for us.
EG: What were the core challenges faced by Cobrainer in the early days of the business?
Amelie Spath (AS): We started out bootstrapping and basically being revenue-financed from the start. On the one hand, that was great for us because it forced us to deliver a product that is useful to our customers. But on the other hand, we were really prone to accepting customization requests. It destroyed our focus because, when you have to make revenue, you have to make revenue. And if you have to stretch or go a bit outside of what you actually want to do to make those revenue, you do it. The real challenge was feature creep. Machine learning and AI, it’s a powerful technology. Almost everyone we talked to had new ideas of where to apply it or new use cases that we could do with expertise analytics.
Initially, we were thinking: “Hey, this is awesome. There’s so much opportunity, so much potential”. And then at some point we were like, “No. Hey stop! This is not going to work.”. And then our mantra became okay, what is the smallest product we can build.
EG: How did you know that Cobrainer had found product-market fit?
Amelie Spath (AS): We realized that the very core thing that we do is the data analysis and the expertise intelligence. We understood that this should be the core product. But then there came all of this enterprise customization on top. It was a full-stack product and then we decided to modularize it, build an API only for the core analytics engine and everything else was now coming from third parties. And now we’re considering that we have product-market fit. Before that, there was too much customization to really say this.
EG: What value did you get from reading Lean B2B?
Amelie Spath (AS): I think one of the key things is really how to properly do user testing in B2B. How to avoid lying to yourself when people say “interesting” and really drilling down on understanding the processes of the organization. The buying mechanisms, the buying centers, all of these things in the enterprise and how to engage customers even when you don’t have a product.
EG: How have you been using what you’ve learned from the book?
Amelie Spath (AS): I think one of the things we used the most extensively is the user journey. We conducted around 50 problem interviews. We had ideation sessions and design thinking workshops with the team to map out user flows, processes and come up with solution ideas around our technology. We went back to these 50 people to collect their feedback. It was now a solution to address their problems. And so we really full-on went through this process from the book.
EG: After having been through it yourself, what validation process would you recommend to entrepreneurs starting in B2B?
Amelie Spath (AS): I recommend for other machine-learning / AI companies wanting to go in B2B to really invest time and brains in understanding the enterprise processes and legacy systems used by your target users because that’s the reality your users have to work with. And even though companies are fed up with it and want to change, they cannot change it overnight. It takes probably two years to replace a legacy system. So, in the meantime, you either don’t work with that company or you accept that reality and make the enterprise product work in their context.
EG: What advice would you give to new entrepreneurs starting in B2B today?
Amelie Spath (AS): I definitely encourage new entrepreneurs to go talk to people in the enterprise. Even if you don’t know them, go to conferences, talk to relevant people, cold call if that’s the only way. Don’t be afraid to show a product even if it embarrasses you. Don’t make too many assumptions about how the people in the enterprise are, what they want or how their day is. Really understand the reality of enterprises; what’s a matter of life or a matter of fact in the enterprise.
EG: Thank you for your time.
AS: Thank you.
More on Enterprise Product
- Is Bottom Up SaaS the New Way to Get Products in the Enterprise?
- Why B2B Startups Should Target SMBs First, Not Big Enterprises
Working on a B2B Startup?
Learn B2B customer development with our free email course: