Photo: Getty Images
Is Dick’s making the right move by bringing its software development in-house?
Paul Gaffney, chief technology officer at Dick’s Sporting Goods, is leading an effort to bring the chain’s software development in-house rather than rely on third-party tech vendors, according to a Wall Street Journal report. The goal of the program is to give Dick’s the ability to create and control solutions that meet its specific business needs and, in turn, place it in a position of competitive advantage over rival retailers.
Mr. Gaffney, who led a similar effort at Home Depot before joining Dick’s, began by having his team tackle the retailer’s inventory management system. The goal of the program, which is about halfway to completion, is to create a system that would earn at least 10-times in annual revenue the cost of the eight-person team that developed it.
Before the end of the year, Mr. Gaffney and team are planning a complete overhaul of Dick’s e-commerce platforms, including how products are displayed online and how search, checkout and shipping estimates work.
The changes to Dick’s systems is reportedly enabling the retailer to be more responsive to its customers and internal stakeholders, as well. In speaking with the Journal, Mr. Gaffney pointed to Dick’s ability to list new merchandise online within 30-minutes of breaking news such as a team winning a championship. Previously, the same task took up to five days to complete.
Dick’s, similar to other retailers, has placed greater emphasis on e-commerce operations as more consumers go online to research products and make purchases. On the company’s fourth quarter earnings call earlier this month, CEO Ed Stack said that online now represents 23 percent of Dick’s total sales, up from 19 percent a year ago.
Lauren Hobart, president of Dick’s, said the company is building two new e-commerce fulfillment centers to open in the third quarter. The new facilities located in California and New York will enable Dick’s to deliver the majority of online orders within two days.
“Our technology investments extend beyond just improving our website. We will continue to increase our investments in technology, talent and capabilities to make the shopping experience easier and more convenient for all athletes (customers) regardless of when, where, or how they shop with us,” said Ms. Hobart (via Seeking Alpha).
Dick’s is also heavily focused on providing technology to its teammates (associates) to improve productivity, as well.
“We recently delivered an application, which we’ve named merch search, to all Dick’s stores nationwide,” said Ms. Hobart. “This app provides real-time product information including detailed product descriptions, inventory availability, and alternative product recommendations so that our teammates can guide athletes through their shopping experience.”
Discussion Questions
DISCUSSION QUESTIONS: Do you approve of the decision by Dick’s to bring its software development in-house rather than rely on solutions from third-party vendors? What do you see as the pros and cons to be considered by retailers considering a similar move?
Poll
BrainTrust
Lee Peterson
EVP Thought Leadership, Marketing, WD Partners
Tom Erskine
CEO, One Door
Dick’s Sporting Goods is a retailer. They should focus their efforts on being a great retailer and taking care of their customers the best possible way they can. They should focus their efforts on retaining customers. Becoming a software developer when you have no core competency in it will distract from being a retailer. There is no competitive advantage to be gained from this initiative. Too many great development teams are already building systems far beyond what Dick’s can think of.
If you want something custom, partner with a development group that can do it better than you and stick to being a retailer.
I always have concerns when a company insists on doing it all in-house. First, it’s hard to find the best talent because those with the highest skills prefer not to work for one company but to remain free demanding top dollars for their skills. Second, when you’re an outside vendor, it is imperative that you stay competitive and stay on top of the latest and greatest new trends in your field. In-house people deal with different pressures and often become inundated with too many demands making it impossible to meet deadlines or stay on top of new techniques offered in their field. The company often feels they have more control doing it in-house which may be true but without top talent and the latest technologies they may find they’re not receiving what they could be from a capable third party. This concern is not just with technology but other essential departments such as advertising and training. Doing it all in-house tends to be a bit ego driven and often does not save the company money. So, I see this is as not the best move for Dick’s.
Internal development is a double edged sword. With internal development you gain greater control, but you’re limited by your own capabilities. With external development, you should have access to a broader range of expertise and theoretically save cost, but you lose some control. The pendulum on build vs. buy decisions swings in both directions. It seems that Dick’s believes that these systems are mission critical and that the greater advantage is to bring development in-house. While I respect Dick’s decision to move development in-house, I think more retailers are looking to outsource than build/expand an internal development capability.
After reviewing about a dozen research papers on this exact topic, I found no research to support that doing software development in-house for a retailer is a good idea. McKinsey & Oxford showed that 66 percent of these projects take on cost over-runs. 33 percent experienced schedule over-runs. 58 percent of large companies found their own project had content shortfall. Standish’s CHAOS research (2015) found that 61 percent of these projects failed, or were seriously challenged. All told – not a wise choice.
E-commerce is growing. Omnichannel is starting to take hold too. It is enticing for retailers to get their teams to just hack together solutions that are exactly customized to their needs. A better strategy is to use these resources to configure and customize existing best of breed technologies that are already developed. Perhaps listing new merchandise (a previously five day task) needed a new process to get it down to 30 minutes. Ultimately, do retailers want to become programmers? Or is it best to outsource the peripheral development, and stick to one’s core capabilities and strengths?
This is interesting given the very small relative size of Dick’s versus Home Depot. Diverting a lot of resources in this direction as opposed to sourcing best in class tech may be risky. Absent in the news is anything about Dick’s investments in analytics to better inform how it runs its business through technology. If anything, retailers need to invest more here and have a strategy.
So Dick’s wants to get into the software development business? With plenty of third-party software vendors, which most would be happy to customize to their needs, I’m surprised Dick’s would want to spend the money and time it takes to develop their own solution. Retail is what they do best, and most might say, “Stick to what you do best.” The exception to this may be that they can’t find what they want (highly doubtful, but possible). If that’s the case, it might make sense to create their own software. And to recoup some of the effort and cost, they may want to license their solution to other retailers.
There are two schools of thought. After decades in this business, I belong to the school that says: unless you have devised the better mouse trap, and no commercially-available software comes close to supporting your need, buy it! There are many reasons for this. The next best thing to re-inventing the wheel at huge costs, which a single retailer cannot amortize over a customer base of 50, 100, or 200 other retailers, is to license the best available solutions, and modify them to suit your business, but I don’t recommend this either.
Having been a retailer, I REALLY like this idea. A little like private label development though, it’s not anywhere near as easy as it sounds. But to have the ability, let’s say, as a merchant, to talk to your developers and tell them what kind of reporting you thought you really needed and then to work with them on-site and then actually get it? That is SUCH a home run (opening day, sorry) for all involved.
Also, think of your insights team having the ability in-house to sort information and derive what they think is most relevant and apply it to the business they are intimately involved with — wow. When you consult or are a vendor to a brand, it’s like being a grandparent. But if you ARE the brand, you’re absolutely the parent. That metaphor really holds in this case.
In-house development is nothing new; Crate & Barrel did it for years, and Target had a huge development farm in India. I’ll defer to Mr. Gaffney’s good judgment.
My gut says, “Bringing it in house is difficult and resource-intensive.” BUT – they’re looking for something that is purpose-built for their business. If the tech isn’t available, they have to innovate. I agree with Cathy, Mr. Gaffney has a vision and a track record of success. I wish them luck!
“Transform our technology stack into a weapon?” Good! “Think about solutions ‘store in’ versus ‘HQ out’?” Great! “Get a 10x return on our software investment?” Check! “Write all our own software?” Wait, what? You lost me.
There is a lot of business value in the innovation happening in the retail software industry, and it is hard to understand how Dick’s could be smarter and faster than the industry at large.
I am guessing (and hoping) that the Journal’s editing of this piece and headline misinterprets the spirit of Mr. Gaffney’s intention. But if not, then the bright side is at least his inbox will be less clogged with vendor meeting requests.
Many online retailers see themselves as tech companies first and hence do much of their development in-house until they realize there are significant disadvantages to doing so. The argument that they get “just what the business wants” also means they are limited to the knowledge within that business. The fact that it is slower and unless he builds a huge empire will create a big backlog of projects and the projects delivered get out of date very quickly are just a few reasons why this is a flawed strategy.
It is a brave or foolish man that believes he knows better than everyone else. I thought most retailers had grown out of this some years ago. Technology moves too fast, especially in the dynamic world of retail.
While software dev is not my area of expertise, I did make a textbook omnichannel purchase from the them last week and some areas of this tech revamp were working, while others were not.
Purchasing new soccer cleats. Started on Google (on desktop) to check options, reviews, and decide top option. Went to Dicks.com to look at sizing availability at my local store – they had my size in a different color so I went in to the brick-and-mortar store to try on. Verified fit in-store and was told I could order online with an associate at a kiosk (not sure this was the same thing as the “merch search” mentioned in article). This was a frustrating experience as the kiosk employee did not let me type in my own info which took 5x longer to relay to him and correct his misspellings, etc. only to have an issue where the page would not advance so checkout could not be completed (associate seemed frustrated as if this was not the first time this had happened). He said I would have to order online on my own.
At this point I would have normally opened up my search to the wider internet again to get the best price, but they were lucky because I had a Dick’s gift card so I stayed in their channel. Ordered on my mobile device while in the food court at mall, regular shipping. My order arrived in 26 hours which I was ecstatic about, so the delivery within 48 hours is something they knocked out of the park in my case. Seems like this will make their “express shipping” option obsolete, but for a consumer this is a win and something that Dick’s and every store who finds themselves with customers inside their walls ready to make a purchase but not able to because of in-store inventory availability need to be able to deliver on.
Um, how do I say this nicely? I think it’s a very strange decision. I thought we were over the “do it yourself” era of software development.
I think this can make sense for Dick’s, but only if they can achieve something referenced in the first paragraph: a position of competitive advantage over rival retailers. If they believe they understand their business better than the tech vendors and have identified genuinely unique solutions that they want to protect, then I think this can work.
If this is simply to develop in-house capabilities or software that is currently available or that can be developed quickly using a third party, I am less enthusiastic about this approach.
Investing in getting products online faster and gaining insights from data will improve experience and sales. I’m not convinced taking ALL software development in-house will do this as well.
It sounds like I’m in the minority on this one (and I don’t claim any expertise on software development), but it seems like a smart move on Dick’s part — provided that they can hire the right talent to execute the strategy in a cost-effective manner. After all, who knows more about the specific complexity of Dick’s business than the company itself? As the demands for system development get more complex (driven by real-time inventory management and omnichannel initiatives), Dick’s is at least trying to control its own destiny.
I agree with many of the sentiments here. Being a retailer is hard enough, but staying at the cutting edge of emerging technologies while trying to maintain the age-old inventory and POS solutions would require one hell of an internal team.
We’ve seen a few retailers (attempt to) do this before with customer marketing and loyalty applications with very mixed results.
In one case, an in-house dev group developed an application for a pilot program and it worked reasonably well but was largely absent a rich feature set. In another case, a similar size retailer earmarked $15 million for a new platform and after spending $9 million pulled the plug.
Whether Dick’s is successful depends on its IT leadership and ability to attract talented people to their team. Starting with a hurdle rate of 10x incremental revenue for the business case is reasonable but spending 10 percent of incremental sales on new technology seems high for retail. Let’s hope they got their incremental revenue right.
Dicks’ current software vendors should take this as a wakeup call – Dick’s would rather reinvent the wheel than use yours!
As a solutions provider for retailers, one of our core tenants is to empower our clients to become better merchants. As such, most aren’t by nature software development shops (with some exceptions). But I get it. The belief and determination that controlling ones own destiny, building and protecting ones “secret sauce” and staying more agile are significant. All that said, I think the reality is more nuanced. More likely, collaboration with a trusted partner to enhance and extend an existing platform for unique business needs is probably the path for the majority of clients.
The challenge of doing it in house is that you are limited by the expertise available and you don’t get the scale of knowledge of requirements in the future that you haven’t thought of. Also third party vendors would be reluctant to share information with you.
I remember long time ago working at a retail software company and we were told not to show our software to certain retailers who are known to leverage vendor IP for in house development. You also create a proprietary platform that is difficult to maintain when the talent changes.
Building a best-in-class software development team is an expensive proposition and it is typically only feasible for very large retailers. While having an in-house software development team has some benefits – agility, control and focus – the disadvantages often outweigh these benefits. Internal development team don’t have the breadth of creative expertise as a lager IT organization that has the luxury of a team of specific focus areas with detailed expertise in a wide range of topics. Also, companies (software vendors and creative agencies) that work with multiple retailers can draw upon best practices of successful development designs and implementation approaches and apply them to your project.
Why would you forge nails when you can buy them at the hardware store.
There is no magic to writing and developing software for a business. Bringing it in-house for a company the size of Dick’s is a logical move and one that makes logical sense when empowering the company to better manage its inventory, logistics, and retail trackability. I applaud Dick’s and believe that there is no difficult move to their bringing software development in-house.
Having built commercial retail software and developed custom solutions both in-house and as a consultant for retailers, I think we are at a point in time where it makes a lot of sense for a retailer to selectively build in-house solutions.
Technology companies, startups, and now even super Tier 1 retailers are leveraging their internal technical skills to raise the bar on consumer expectations. If you can’t lead or at least keep up with their innovation in your segment, you will lose business to them.
The development tools and technologies available today make it quite viable for a retailer to retake control of their destiny with strategic internal development and third-party components. This is not like in the ’90s where if you wanted to build your own house, you had to cut down your own trees. The tools and components are available.
On the commercial side, vendors are not adapting quick enough to this new reality. There are major conflicts developing in the business model, and the granularity and flexibility of their offerings. The model of buying or upgrading highly parameterized monolithic applications every couple of years to try and just keep up with the industry, isn’t working. It is not a competitive model against businesses who have an agile architecture, control their own UX, and can quickly leverage new internal and external services to enhance their business.
A retailer who hasn’t discovered that they need internal technology skills, more agility, and tight cycles of continuous improvement to be a great retailer, will ultimately fail. That should be the impetus for their digital transformation.
Kudos to Dick’s for stepping up. Vendors who want their business will also have to step up and bring them nuggets of gold that create value in their world and operate under their control. The vendors who want to be successful going forward will have to decide if they can transform as well.
I’m personally observing something of a trend more back to bespoke development, driven by pace of innovation, and reduced cost to deliver. It’s certainly not a defacto approach for everything, but in consumer facing areas the package approach can’t keep pace, resulting in increasing customisation and integration pressures that make in-house development more viable. I wrote some thoughts about this very subject on LinkedIn recently, here.