Scaling Success: How We Elevated a D2C Apparel Brand's Digital Storefront and Operations

About The Customer

The company, a subsidiary of a large Indian business conglomerate, is one of the few D2C apparel retail companies with a track record of profitable operations. The company has a growing popularity amongst gen-z and produces a diverse range of apparel.

The company has been an established reseller on India’s top 4 e-commerce platforms and earlier in the year 2023, the company set its strategic objective to strengthen its D2C footprint at scale.

D2C is always a whole new problem even for established players. The problems of setting things from scratch and finding a Product Market Fit (PMF) in an uber-competitive space often makes established businesses get a feel of what it means to be a startup.

And if going D2C is a challenge, going D2C at scale is an even bigger challenge. Here setting up digital storefronts is only a small part of the problem needing optimisation. The bigger challenge lay in ensuring the operations, supply chain, etc and geared for the additional scale. While this often is a blind spot for many players, the company demonstrated early pragmatism by looking at the larger picture from the get-go.

The Problem

The problem then was to establish a digital storefront, find its PMF and to keep a keen eye on key operational parameters.

Owing a digital storefront means ensuring the end consumer gets the right user experience. Growing scale means using different engineering approaches to ensure the tech stack continues to perform optimally.

On the operational front, tracking order leakage posed a significant challenge. Order leakage is a mismatch between orders placed and orders delivered or returned. Leakages are often an outcome of scale where on-ground operations teams are not able to stay on top of the influx of orders. Often many third party operators for associated tasks like logistics are used. Dealing with many such third parties often opens up room of error at scale. Needless to say, unaddressed leakages cause significant margin losses along with disgruntled customers.

Returns pose another significant threat to any D2C venture. The apparel D2C sector is known to suffer returns as high as 20% - a number that often exacerbates at scale if not tracked properly.

Input supply chain is where the challenge often peaks. Here having only a view of the supply chain is not enough. If a company has to grow more and grow faster, they need to start predicting various aspects of their supply chain.

Yet supply chain, leakages and returns form only the tip of the iceberg in the D2C world. Depending on the scale of operations, hundreds of aspects need tracking in real time. Only a well designed visualization can help teams make sense of this mountain of data.

Bringing these 100s of data points in place meant integrating data from a melange of third party operators, each having different sources, formats etc. Many of them did not have any real time APIs that could be used to pull the data on demand.

Solution

It was clear that the previous digital storefront was not going to handle the scale. A new engineering design was needed. Hence a top-down approach was adopted to solve the design problem. Clear Key Process Indicators (KPIs) were established to measure the performance of the tech stack of the scale.

The entire tech stack, starting right from the architecture of the underlying cloud system, was redesigned for scale. The database was remodeled such that it could prioritize the digital storefront yet service the internal requirements of real time prediction, analytics, dashboarding, alerts and so on. The entire solution was finally configured to scale with the incoming traffic.

With the storefront taken care of, the internal tech stack was designed to ingest data from a wide variety of third party solutions. Since more third party sources could integrate in future, design focussed on being able to do it faster whenever needed. Modular ingestion methods were designed that could be invoked as per need.

Reconciliation is a recurring need with diverse data systems and parts of it were automated from the get-go so as to keep things efficient for operations. Real time alerts were configured so teams could have the right level of responsiveness to incidents when needed.

To do all this, a lot of data had to processed very fast. Conventional software engineering approaches cannot deliver such fast throughputs hence specialized engineering approaches, seamlessly combining hardware with custom software, were used to solve the speed problem.

With the operational visibility problem solved, we could now solve the final piece of the puzzle - predict the supply chain. As of today, custom models are being used to bring reliability into the supply chain.

Outcome

The near real time data ingestion solutions have managed to bring visibility into key problem areas, like returns. The company is able to address these key areas along with being able to shift focus to the changing needs of operations in a timely manner.

The carefully designed cloud and digital storefront delivered top notch user experience even as the traffic increased 3x in 4 months.