Since day one of my work at Bazaarvoice, I’ve been fascinated by the innovation coming out of our Labs team. So I sat down with Jon Loyens, Bazaarvoice Labs Director, to see what makes his team tick.
Can you give us a brief overview of what the Labs team does, and what it’s responsible for?
Bazaarvoice Labs is the new product research and development group at Bazaarvoice. I’d like to put some emphasis on the ‘new’ and ‘research’ parts of that statement. Creating a new product that tackles a new market or brings our existing products into new and existing channels comes with certain assumptions. Labs is a team of engineers that works with Product Managers to test assumptions about user behavior and community engagement by building prototypes and running pilots with our customers on our most exciting new product ideas. By doing this sort of live testing up front, we can stay on the forefront of social commerce by testing things like Facebook, Twitter and mobile integrations while making sure we deliver our customers solutions with proven business value.
Why is a Labs team important at a software company?
As companies get bigger and become more successful and get more customers, a lot of times they can get stuck between a rock and a hard place. Their expanding customer base means they have great opportunities to grow relationships but their customers, rightfully, want them to continue to improve their existing products and add more and more features. This often leads to bloated products with features that few customers take advantage of. On the flip side, that same company will try to introduce an entirely new product but often it will have ill conceived features, no value proposition and then see limited adoption by the customer base. A labs team helps a company by ensuring that there are engineers who can operate in a truly agile environment unfettered by the processes that tend to start to slow down successful software companies. When we create a prototype or run a pilot in Labs we follow agile methodologies and release minimally viable products with a small group of customers that we work closely with. This allows us to let real data guide our product decisions in a discovery driven design process. This means that when we introduce a new product at Bazaarvoice it’s been tested in the real world and has a solid business case associated with it. Ultimately Labs allows us (and any other software company that would choose to adopt a process like this) to experiment with new ideas and products without fear.
Are there Labs teams at other companies that you look to for inspiration or best practices?
I think the most obvious example of another successful Labs process is Google’s. How companies run their Labs can vary widely from company to company. At some companies, Labs is more of a marketing moniker for passion projects executed by their engineers, at other companies, like ours, it’s a dedicated team with their own mandate. Intuit has a Labs group more like ours. I think from a best practices standpoint, I don’t really look to other Labs groups but rather companies who do discovery-driven design and development and really let the data guide their product decisions. Many, many consumer oriented web companies do this: Facebook, Google, LinkedIn, Amazon, etc. You see this less in B2B companies which generally collect requirements directly from their paying customers. I think what we’re doing at Bazaarvoice is unique in that we rely on our customers to deploy our software on their sites but they rely on us to follow the data and give their users the best possible social experiences.
Do you have any advice for companies that want to start their own Labs team?
Make sure that the team has clear goals and don’t be afraid of failure. It’s a bigger failure to come up with something that only has mediocre results and then letting that limp along. If something sort of works, keep changing it until it fails or until it’s great. We started our team when Bazaarvoice was still relatively small so it was very important to have great results. We needed to work quickly to make sure that we met the goals that were laid out for us. Failed projects were fine but we definitely needed to iterate and get those failures successful.
How do you learn from the experiments that fail?
I don’t think that any particular experiment is a failure as long as we’re learning and our technical execution wasn’t flawed. We take the data from every experiment we run back to product management and they evaluate whether or not we have a business case for a new product. If user engagement isn’t there, if we don’t add that feature or product to the core offering, then it’s a win for both Bazaarvoice and our customers.
Which project was the most fun to execute, and why?
They’ve all been really fun in unique ways but the most exciting to me are the ones where the customer we’re piloting with is really engaged and pushing us hard. Like I mentioned earlier, we’re completely reliant on our customers to be able to pilot new features and customers. We don’t have our own ecommerce site like Amazon does to be able to try new things out on and many of the brands we work with are pretty conservative. That said, a lot of the brands we work with really want to fly in the face of convention and push the envelope. Facebook Ratings and Reviews and Ask and Answer have been awesome because the social team at Benefit Cosmetics has been so wonderful to work with. Intuit pushed us really hard when we did our first on-site Facebook integration with them for their Friends Like You campaign and it was awesome too. Travelocity is the first customer to launch with our new Social Connect Discovery project and they shocked me with how accommodating they’ve been letting us getting our new ideas out there.
What can organizations do to support the success of their Labs teams?
Honestly, it’s not about the success of the Labs team. It’s about supporting the right processes organization wide to foster innovation and let people learn by doing. Making sure that the creative folks in your business have the free time to follow their ideas will go a long way. Make sure that people can take risks and appropriately postmortem anything that doesn’t go well. I’d always say that if an idea fails, that shouldn’t reflect on the person. In fact, as long as the right lessons were learned, I’d say that person should be celebrated. One thing I don’t tolerate though are failures of execution. An idea can fail because the UX isn’t right or the market isn’t ready for it or any other number of reasons, but because we are launching features on our client’s sites, we need to make sure they scale and are bug free. I don’t want technical issues confounding our business results.