By Yvonne Hyland, Head of Portfolio Business Development, B Capital Group

Artificial intelligence, machine learning, and other advanced technologies have incredible potential for business. With the right application, a large enterprise can use these tools to accelerate productivity and time to market while trimming costs, helping the company gain competitive advantage.

But just adopting a new piece of software is not enough to take full advantage of its benefits. You need a strong partnership with a trusted vendor. So how do you find the right vendor and build that collaborative relationship?

B Growth Network recently attended the HMG 2022 CIO Summit of America in New York City, where a select group of Fortune 2000 CIOs and CISOs gathered with investors and startups to explore how large enterprises can take more innovative approaches to designing their futures. 

As part of the summit, I participated in a panel on successful technology partnerships. Other panelists included executives from cloud integration platform SnapLogic, AI services management platform Aisera, and New York’s cutting-edge Hospital for Special Surgery. We discussed strategies for evaluating technology vendors, and how to drive results once you’ve found your match.

The truth is, evaluating a new service provider can be daunting, which is why you need support through that process. One place to start is by building a relationship with a venture capital firm that invests thematically by sectors, that has its finger on the pulse of shifts to come, and—crucially—that has done extensive due diligence before investing. 

Start with trust, and build from there

Vendors want to do the best job possible. Make your expectations clear so they can meet or exceed them. Set a reasonable scope, provide feedback as it comes up, and just generally keep lines of communication open. Find out what resources your vendors need to succeed, so they can help you succeed.  

“Vendors don’t want to fail, they want to do the best job they can. They want to help you get promoted. They want to help you to progress your career and have great stories to tell your CFO when he or she is asking about your business case and say I exceeded it by 25 percent a couple months early,” said Aisera CRO Anthony Palladino.

Trust requires ongoing work, which is why you should treat your vendors as partners. Work with them. Communicate regularly with them, find out what resources they need to be successful, and offer feedback along the way. 

“There’s got to be trust,” said panelist Craig Stewart, the chief data officer for SnapLogic. An early step toward building trust, he said, is to start with references. Ask your community what vendors they use, or ask them about the vendor you’re considering. 

“Find people that are solving similar problems to you, where you can see what other people are doing, find people that they trust, and build out those relationships,” he said. 

That can mean asking other CTOs at companies similar to yours; it can also mean striking up a relationship with a VC firm that does months of due diligence before investing millions into a company. 

Due diligence comes first

The thing with rapidly evolving technologies like AI, which have a constant stream of new and emerging applications, is that the best vendors are often startups. That’s just the nature of software. The thing with startups is that they come with unique risks. 

A lot of startups fail, which can be a problem when you’ve grown dependent on their tools. A startup with weak leadership could see high turnover, which could lead to poor staff performance and ultimately problems with the product. Startups that are bringing new technologies or use cases to highly-regulated industries could be working in regulatory gray areas. Finally, new technology comes with new problems. 

Panelist Bashir Agboola, CTO at the Hospital for Special Surgery, pointed to the problem of bias in algorithms. A few years ago, an algorithm used by a major healthcare provider reportedly misinterpreted the significance of certain data, leading to inaccurate assumptions about which patients required care. The result was that the software recommended treatment for healthier white patients over Black patients who were actually more ill.  

“We would want to know before we start using your technology that the models you use, the people training those models, that they’re conscious of whatever biases they might need to watch out for and that might creep into the data,” he said. 

He said the consequences of failing to do this due diligence in a health care setting could be deadly: “In healthcare, we’re dealing with human lives and wellbeing. If it fails, it’s someone’s health that’s going to get compromised.”

Algorithm bias could be a problem within your own company’s tools that a startup can actually solve; or it could be a problem that a startup brings into your company. In either case, you want to make sure you are contracting with a capable vendor. That means performing due diligence. The problem with due diligence is that it can take a lot of time and resources. The way to get around that, without cutting corners, is to work with an intermediary that has already done the research. 

If you think about a VC firm—any VC firm—the amount of due diligence that’s done before investing millions or hundreds of millions of dollars in a company is quite substantial. We look at how well the product fits its purpose, the product roadmap, the strategy, and of course the management team. 

A startup with funding from a highly-regarded VC firm is substantially de-risked before your company even considers contracting that startup as a vendor. That means that when you work with a VC firm, you’re getting the best of all worlds: You’re accessing startups that have already been vetted, that offer the latest tools, and that are ready to integrate rapidly with your systems.