This is the 1st of 2 parts. Looking for Part 2? Tech Player – Steve Jacobs, CIO Gilt Part 2.
Recently we had the pleasure of sitting down with Gilt CIO Steve Jacobs for an extended conversation on his approach to IT management. We touched on a range of topics including IT’s role in the business, how Gilt thinks about the cloud, working with the line of business leaders, and more.
Let us know your thoughts on Twitter @CloudGathering.
GC: What characterizes your approach to IT for Gilt and what you’re looking to achieve?
SJ: There are technology organizations that are very in-service of the business. I’ve worked in financial services, an online marketing startup, and here at Gilt, and the main difference with Gilt is that we’re trying to be a peer business partner, whether it’s in marketing or in merchandising. You need to do this to enable technology to have the biggest impact on the business.
Engineers by trade think differently than non-engineers, and when we as engineers relegate ourselves to a bit more doing and not as much thinking about what we’re doing, we do ourselves and the company a disservice. For instance, engineers are obsessed with automating things. I had an engineer who called himself lazy, and it sounds like a derogatory term, but he was actually complimenting himself. He would never do anything manually because he would automate everything that he had to do. Technology companies, like Google or Facebook, or companies that are very engineering-led, are making different decisions than if it was line of business-led. One of the things I’m trying to do is to bring some of that to a company that doesn’t necessarily look from the outside like a technology company. A lot of the opportunities that exist for Gilt’s further growth are absolutely coming from technology.
GC: What does cloud means in terms of your approach manifesting that sort of technology ethos in your organization.
SJ: Cloud for Gilt is really important because of the nature of our business. We get spikes in traffic at noon EST when the sales start, so being able to anticipate usage is an important way of keeping costs down. One of the key things for the technology department culture is empowerment: we define a particular business goal, want to improve our search engine optimization for example. We then come up with a Key performance indicator (KPI) for that goal, for example the percentage of revenue that Gilt gets from search engine optimization. Then we put a cross functional team against that—which includes engineers, someone from marketing, a product person, a user experience person, and whatever other skills that are needed on a cross functional team. We put them on the problem and we don’t micromanage them or ask them to put together a six month roadmap. A lot of these customer-based initiatives are somewhat experimental to determine what’s going to work, and there are a lot of hypotheses. We’re extremely test-driven, and we use results, and a lot of A/B tests to see what actually does work. One of the things that would be frustrating for an empowered team is to have to wait three weeks to get some hardware or infrastructure. From the perspective of getting quality results from the empowered teams we need to determine what things can be done to ensure that those teams are actually empowered and move quickly. One of those things is easy access to the infrastructure and not having to wait three weeks to test out a new idea.
GC: What are the other challenges that you’re solving through and discovering with cloud services?
SJ: Performance has been a challenge that we’ve had in the past with cloud. The world is a lot easier when you own everything and you know what’s happening on your switch. When you’re going down to the level of detail where you need to understand the ins and outs of what’s happening on the network to understand cloud performance, cloud becomes a less ideal environment because of shared tenancy. I think we’ve encountered some of those challenges which have made us focus on the easier things for using the cloud.
For example, we migrated to Gmail from Exchange, and though there were certainly some hiccups, the performance difference has been such a breath of fresh air.
But there are people for whom it’s a cultural challenge to adopt cloud and who are not thinking about it from a benefits perspective.
GC: What’s the hesitance there for them?
SJ: When you have people who in their job spend a lot of their time in email, and then you tell them that the tool, let’s say Outlook, that they’ve known and loved for 10 or 15 years is going to change, they naturally have some resistance to that. But it’s really just change management and training, and the people who have been on both Outlook and Gmail can see the benefits of going the Gmail route.
GC: What about change management more broadly in terms of integrating new platforms, new technologies into how you are approaching your IT? Do you have a same or similar sort of, sort of uphill climb to get that level of buy in?
SJ: Anything that has to do with people can be hard, and so change management in general can be difficult. With Gilt, we have the tools to build a brand new store every day. Whether it’s new internal systems, new inventory, or new creative, there is a lot of internal tooling that goes on behind the scenes to create that new and exciting website. Change management is an important discipline for us to get buy in with business users and get people excited about the change. The good news is that culturally, the people who work at Gilt are obsessed about making things better. That’s one of the best things about being at Gilt, and Gilt culture, and so therefore there is more than your average amount of embracing change. So we have a leg up in that regard, but it’s still not easy.
GC: In terms of your interaction with the line of business units within your organization, how do you go about getting buy-in or getting the process for getting buy-in going with them? What is their level of understanding in terms of what you’re bringing to the table now compared to perhaps five or ten years ago?
SJ: We have this team of 130 people in technology who impact how we invest, and they are in touch with understanding the opportunities and challenges facing the business at any given point in time. We pick all the possible things we can work on, and we can condense them down to a fairly compact list, maybe 15 or 20 initiatives that we could focus on. Then we actually model out the impact to the business, and whether it’s the customer-centric or internal systems or business-level efficiencies, it comes down to revenue or cost savings.
We look at that as a first pass, purely by the numbers, and then we look at those initiatives compared to our strategic initiatives. By wanting to improve our internal systems, we make our valuable asset, our people, even more productive. That’s one lens, and wanting to do the right thing for the customer, or other growth opportunities are some others. We look at that combined list as a team, first purely numerically and subsequently with a strategic view in mind, sort of bottom up and tops down. We use that to figure out where we want to invest. We all hold hands, look at where we want to go, and understanding the effort involved.
GC: Do you feel that the line of business is coming to the table with more knowledge, which makes your job easier or does it present new complications?
SJ: The goal is to be in touch enough with the business so we all know what we should invest in. The business partners are all in line, and maybe with a couple of tweaks or through discussion it can be done, but the majority of that work should be done up front. The hard thing about making those decisions is not what to invest in; it’s what’s not to invest in. A lot of technology organizations have lists of say 150 really good features, and everyone has a favorite feature, so it’s very subjective. We’re talking about rolling those up to more initiatives, whether it’s a marketing initiative, or a merchandising initiative, or performance improvements, for example. Whatever those macro initiatives are is where we want to invest. When people talk about innovation they are often talking about the solution to some problem. But the real source of innovation is figuring out what the right problem is to solve. Take Gilt for example: Gilt took the invite only friends & family sample sale invite that existed for coveted brands in cities like New York and moved that model online. This allowed consumers all over the US to have access to designer apparel and accessories for a short period of time, similar to a two day offline sample sale, and it allowed the brands to have another outlet to sell end of season merchandise. Gilt was perfectly positioned to meet both the brand’s needs and the customer’s needs. The identification of the problem is the hardest part. The solution, once you’ve identified the problem, tends to be pretty easy. If you give any set of engineers a problem, they’ll figure out a solution with confidence. But really, picking which problems to go after in a world of limited resources is where the magic of innovation happens. I think often times people focus on a solution or cool technology that’s really innovative, but it’s only innovative if it solves a substantial problem.
Check back tomorrow for the conclusion of this interview.
By Jake Gardner