Modern retail is a function of establishing trust and imbibing intrinsic simplicity in all selection and purchase funnels. Consumer shopping experience over digital and offline channels are increasingly being seen as one unified channel. Omnichannel is no longer just a collective term for offline and online but an aggregated operational framework for delivering a more simple and cohesive shopping experience. Modern approaches for better retail practices focus more on consumer habits and create an ecosystem of product-driven follow-up which furthers a circle of value delivery.
Consumer’s data security is at the heart of retail
Zivame ensures to create a tangible value for its consumers which can be realised across its digital product experience, post-buy experience and differentiated micro-interactions. Needless to say, these facets are individually and collectively powered by our in-house products and business-first technology initiatives.
At the heart of Zivame technology, lies consumer security, data protection and platform reliability. Our tech platforms cater from baseline to tower peaks with equal ease and infrastructure is provisioned on demand in seconds – these secure and scalable components are built in partnership with AWS (Amazon Web Services) as the choice of our IAAS (Infrastructure-as-a-Service) provider. Some of the core stack components and integrations to enable infrastructure at scale are developed as per AWS scalability principles and commodity technology.
Core technology engine is fuelled by our data and insights platform which is enabled at multiple levels namely communications layer, data ingestion layer, data processing layer, data storage layer, data querying and visualisation layer.
Foundation-wise, it is created bottoms-up with an N-dimensional understanding of our consumers and products. This underlying knowledge asset serves multiple data science decision support systems and consumer use cases at Zivame. Our AI stack is closely integrated with Sagemaker as an enabling ecosystem and all the second-order derivatives essentially are powered through Apache Spark-based distributed processing.
Enriching ‘digital experience’ via data insights
B2C companies hugely rely on the market space they create over a period of time in the form of RTUs (Repeat Transacting Users) and referred NTUs (New Transacting Users), towards the same goal. Zivame believes in creating as well as facilitating a digital experience via our onboarding experience, product relevance engine and occasional serendipity.
Our core tech stack enables this via a sequence of behavioural events’ understanding, extraction of short-term and long-term consumer intent, identification of ongoing trends and our fulfilment of the same. For computing these consumer and product insights, from an infrastructure and scale perspective, we depend on customer ingestion pipelines hosted over custom deployed components like functional language-based API servers, asynchronous distributed ingestion pipelines and multiple spouts ingesting the same based on individual use cases.
The processing comes to PySpark clusters in an on-demand and pre-reserved setting, sequenced by a use case-specific data warehouse. These warehouses serve as business intelligence serving layers in both real-time and on-demand.
Digital learning is an ongoing process in a ‘Phygital’ environment
Zivame’s business culture, in accordance with our technological advancements, follows a strong mandate of learning from experimentation across. There are outward (consumer) facing experiments we do over digital channels for both online and offline initiatives spanning AB / Multivariates / Multi Arm Bandits / Limited Exposure Explore Exploit. There is a second order of derived learning as well, which creates a sustainable advantage for all business lines.
This is inward-facing and more towards an attribute-level insights mechanism at different pivots, calculated independently for respective business channels. These practices going forward will be introduced in our non-digital format as well, which brings us to the next logical state – Omnichannel!
Phygital retail technology strategy is a combination of multiple sensitive elements and validation of GTM works in a very different way for both channels. Simply put – digital consumers are more open to discovery and are more responsive to offers but their intent of purchase generally is much weaker as compared to offline and they are more susceptible to purchase funnel drops. On the contrary, offline consumers are much more likely to convert. Hence accuracy of how well the limited number of styles cater to their demand determines the overall success.
Technology plays a very vital role in both channels – digital channels are driven by wide discovery, huge amounts of exploration divided into smaller consumer cohorts whereas offline is much more focused. Zivame engages its consumers in omnichannel settings via a deeply integrated CRM solution for its offline POS products as well as end-consumer applications like mobile apps. This enables Zivame to create a long-term relationship and value delivery for our consumers.
All Zivame stores are enabled with auto-replenishment systems which predict demand at an SKU level given the ROS (Return on Sales) trend and the historic seasonality for the same. This makes sure that movement of styles is done seamlessly and in a predictable way. The same inventory and more can be explored via the store version of our digital web products as well.
The next layer of initiatives are highly experimental given the current stage of their maturity, but are viable explorations into technology-led common phygital inventory models. One such initiative would focus on creating and exploring a product-embedded first virtual showroom which would enable users to maintain digital inventory of their intimate wardrobe.
More details on the same will be shared in upcoming conversations. Stay tuned and excited!
