GCP for Retailers: Use Consumer Data Analytics for Predictive Purchase Insights

SADA Says | Cloud Computing Blog

By SADA Says | Cloud Computing Blog

In an era of cloud and digital commerce, retailers are no longer responsible for merely responding to consumer demand. New data sources, both real-time and batch, internal and external, allow retailers to both predict and influence demand. Actionable insights are not possible without an organized, thorough approach to your company’s consumer data sources for understanding your target market.

Google Cloud Platform makes organizing data pipelines easy. It’s time for you to assess which data sources your ecommerce company needs to take advantage of, and what consumer insights they might deliver. Your analysts are probably already using G Suite for collaboration and user surveys, so throwing in BigQuery and DataStudio for consumer insights will be a natural addition to a marketing toolkit.

Bringing Internally Generated Data Into Google Cloud Dataflow for Predictive Purchase Insights

Many online retailers run on ecommerce platforms such as Shopify, Magento or Drupal. Using these platforms, retailers are able to track internally generated data all the way from transaction logs, referral sources, conversion rates, bounce rates, abandoned shopping cart values, to click-stream data tracking detailed user behavior on site.

While highly visited sites generate truly massive amounts of real-time consumer data, this data can be organized and sliced in such a way to provide precise analytics. Real-time data can be routed from third party sources into Google Cloud Dataflow using Google Cloud Pub/Sub. From there, data can be transformed and modified such that it is more readable to analysts downstream in the pipeline in business intelligence tools.

This internal data can be pulled into Google BigQuery and used to inform inventory, marketing and store design decisions, for example:

  • What time of day are most of our purchases occurring versus customers simply adding to cart?
  • On which devices do customers perform certain operations (product research, browsing, purchasing?)
  • Which product collections do users spend the most time on?
  • Which day of the week is product XYZ typically purchased on?

Insight into purchase and visitor behavior allows you to customize timing of email campaigns and the release of certain offers to optimize for conversion. If your storefront has a physical presence, it might be worth exploring physical internal data you can collect from IoT connected devices. With a plethora of new data sources and visualization tools comes the need for training employees. Cloud adaptation is a culture shift, and SADA Systems offers change management and employee training services to help foster a data first workforce from top to bottom.

Complementing Internal Data With External, Third Party Data

Internal data alone might not be sufficient to create a clear picture of your customer segments. By pairing internal transaction data with social, financial and user preferences data, you can better understand the behavior of your returning, high value customers. It’s important to note that while third-party data can be a win for both you and your customers. Shoppers benefit from being displayed relevant information but it’s necessary to use third party data responsibly. When selecting third party data vendors, reference guides and standards for the collection of such data.

Social data, for example, might be made available by social networks and used responsibly to map the interests of your customers. For example, what other brands do our customers interact with? What activities are our customers engaged in? This information can be used to shape content and targeting of digital marketing campaigns. On Facebook and Google ads, for example, advertisers can run campaigns targeted towards custom audiences of visitors and fans of other brand pages. A helmet company, for instance, might run ads targeted towards customers of certain types of high end road bikes.

Financial and demographics data can also narrow down the targeting of both digital and physical advertising. This data can be purchased from credit rating agencies, for example, and be used to answer: Are there any certain geographies where our customers are concentrated? What are the demographics of customers within these geographies? These basic insights can then lead you to new markets with matching demographics and similar geographies (e.g. urban or rural, climate type).

As a last example, user preference might be collected from user behavior on other sites and market research surveys, such as the quick one question surveys you see as pre-roll to Youtube videos. This preference data can act as a marketwide focus group that describes in detail the complex preferences, beliefs and attitudes of your target market.

All of this external data can be brought into your data warehouse and matched with internal data to create an actionable profile of your primary and secondary target markets. When coupled with Google’s machine learning capabilities, retailers can create robust product recommendation systems tailored to consumers. If your current data infrastructure is sprawling, then it might be difficult to envision what a data warehouse would like in the cloud. SADA Systems provides cloud migration consulting and planning services to help you best map your current business intelligence in your new cloud environment.

Retailers Not Taking Advantage of Available Data Are at a Disadvantage

If you’re reading this and you run marketing for an ecommerce company, you might be thinking that there is a lot of available data you are not making use of. Perhaps your company currently engages in traditional market research and is looking to complement it with big data insights. Or, maybe you’re focused on innovative new ad tech such as blockchain to predict demand in a lightweight fashion. If either are the case, it’s time to consider a data-first approach to customer insights in Google Cloud.

It’s important to define what customer insights you would hope to derive in a fine-tuned data warehouse. From there, SADA Systems can help you outline the architecture of your warehouse and your ingestion pipelines from your ecommerce and third party vendors to get your analysts answers. Contact SADA Systems today for more information.

Seth Moffitt

Seth Moffitt
Solution Architect, Google Cloud Platform


Our expert teams of consultants, architects, and solutions engineers are ready to help with your bold ambitions, provide you with more information on our services, and answer your technical questions. Contact us today to get started.

Scroll to Top