In a cost of living crisis how do retailers ensure that if you choose to spend your money, you spend it with them?
Retailer analysts have long said that they need to have a personalization solution so the next question is buy vs build. They can’t sit back and do nothing.
There are 2 options:
-
1) Use your existing data scientists to build personalized recommendations in-house
-
2) Leverage a third party solution such as Dressipi to deliver personalized recommendations
The obvious solution would be to simply build something from scratch with in-house teams asking each other ‘How hard could it be?’. Building personalized recommendations in-house will give you full control over the finished product and it might seem to offer cost savings. The reality is not so simple.
What are the important factors?
There are many important factors that you need to consider when making this decision. Every case is unique but there are some simple questions you can ask yourself:
Cost: How much will it cost to build my own vs buying a third party solution?
Time: Customers are demanding personalized recommendations. What will allow me to deploy personalization faster?
Customizability: The end solution needs to be one that meets all of my needs. Can I get this with a third party solution?
Maintenance: If I build in-house, how will I maintain a best in class product that can keep up with the competition?
Workflow: Is it easier to manage a solution provider or a team that is building personalized recommendations?
Scalability & Performance: Will building technology in-house provide better long term results?
We’ve dug deeper into the pros and cons of these factors to help you make the essential decision.
Building personalized recommendations in-house:
Getting things started are simple enough for most development teams with basic item to item or user to user algorithms. But it doesn’t end there, advanced algorithms take years of research and development. Dressipi has around 15 different algorithms. These have been built specifically for the apparel industry and take into account the complexities around how people shop for fashion. Think seasons, trends, individual tastes and body shapes.
Building a scalable personalization solution requires vast engineering resources and years of expertise to successfully execute. In a recent article after acquiring Thread, Marks & Spencer said that they think it would take 3 years to build the capability in-house.
Even if in-house teams get to this point, the work is still not done. You’ll need to be constantly updating, maintaining security, managing infrastructure, checking compliance requirements and adding new capabilities to keep up with the competition. It’s a significant commitment!
Additionally, creating new features is exciting for development teams but this maintenance of existing back-end technology in short - isn’t. Expert engineers won’t want to be fixing bugs over building a new AR or VR experience. In addition, while doing all this you also need to ask yourself, what are my engineers not working on? Could they be using their time better to add value to other parts of the business?
A common misconception is that external companies don’t understand a retailer’s customers as well as they do. So how could they build a personalization solution that performs as well? Personalization companies stand out because they are customizable to the needs of each specific retailer. Control is not lost and rules can be written so the solution is completely in line with brand DNA.
Buying a third party solution to deliver personalized recommendations:
By leaning on a solution that is already available, many obstacles are eliminated with a condensed timeline being a huge benefit. With Dressipi all you need is a product feed and some tracking on the site. You can be ready to go in as little as 3 weeks. The longer it takes to build your solution in-house, the more expensive it’ll be in salaries and the cost of not having personalized recommendations.
This takes us onto costs. For small and medium retailers, buying personalization equates to less than hiring one engineer. One engineer would definitely not be able to take on a daunting task of building personalization in-house!
With enterprise deals being larger, the temptation to build grows. That is until you break down your in-house costs. With a minimum of one senior engineer and five junior engineers with salaries of $200K and $100K each over 3 years, the total labor cost alone has set you back $2.1m for a basic solution. For this cost, you could purchase a solution with years of engineering power behind it and invest long term cost savings into other parts of your business.
Final thoughts
Retailers need to remember that if they buy a personalization solution, they won’t give up complete control. They are just outsourcing the infrastructure and algorithms to make a speedy, cost effective way to service their customers.
This is where Dressipi comes in. Personalization is what we do, and we do it well. We’re driving a step-change in personalization for apparel ecommerce. Given that organizations that excel at personalization generate 40% more revenue from those activities than average players (McKinsey), speed of deployment is essential. Everyday that the technology is not ready, money is lost.
Even if buying is just for the short-term. Dressipi can be a building block in the experience that retailers want to build and can save years of time. We have data scientists, engineers and stylists working on our apparel personalization solution to make it the best it can be. Today, Dressipi’s personalized recommendations deliver incremental improvements to profit (+21%), revenue (+12%), returns (-15%) and full-priced sell-through rate (+10%).
Build or buy? What’s the best solution for your business? Get in touch to find out.
