No marketing decision should be made without proper data analysis, as simple as that.
It’s too important a topic to be ignored, while at the same time it is a complex environment to jump into, not knowing where to begin, how to advance or how to combine data and convert it into meaningful information.
Today, data complexity has become an advertiser’s biggest challenge. Someone has rightly said, “Data is what you need to do analytics; insights are what you need to do business”. Here at Brand New Galaxy, we have unjumbled the complexity through a simplistic e-commerce data-management approach.
It might be useful to begin by understanding the data-source landscape and its benefits. The number of data sources in the e-commerce world are varied, but let’s cover some major ones.
Some e-retailers are more advanced than others in terms of collecting data and more open to sharing data. At this point, without doubt, Amazon is leading the way with brand analytics for deeper knowledge around on-platform shopper behaviour including search terms, repeat purchase or basket data. In addition, third-party tools like SKAI (previously Kenshoo) provide a share of voice analytics and advanced campaign reports.
Amazon also keeps updating its data arsenal and, going by recent reports, may introduce brand metrics from spring 2022, which will allow benchmarking against peers in the category or measure shopper engagement rate, amongst other things.
The amount of data might be overwhelming though, and you may need a doctorate in data science to make sense of it, but luckily data is also available through APIs (application programming interfaces). In other words, more digestible dashboards could be built for specific business objectives that simply filter out the noise. With Amazon, now available across three countries in our region, data learnings have been used by advertisers across other platforms that struggle with collecting and sharing e-commerce data.
On the other side, we have new-age e-retailers and aggregators like Instashop and El Grocer in the UAE, and Hunger Station and Nana in KSA.
These e-retailers are also interesting in the way they populate and share their data. For example, the importance of shopper purchase habits or media attribution.
Then, we have others that are picking up gradually and finding ways to shape more advanced analytics products on data, and sooner or later will have to open their walled gardens, besides sharing basic post-campaign reports.
Digital shelf-management tools
These include Profitero, Edge, and Brand New Galaxy’s proprietary tool Synthrone. The story is pretty straightforward: regular data that one gets will focus on the fundamentals of any e-commerce tracking across multiple e-retailers and markets. Is my product listed? If yes, is it in or out of stock? Is my content properly implemented? What is my share of search in a given category? How are my ratings and reviews? How do my prices compare with competitors?
There is plenty of data out there, but we need to know when and what type of data we will need. This is influenced largely by the advertiser’s data maturity curve and how equipped are they to use it. You also need to be sure that your organisation is equipped with talent that is going to digest data and create strategic action plans and naturally optimise operational excellence in a way that they can react faster to market dynamics.
Regardless of the mix of analytical data matrix and where an organisation starts the journey, the real power of analytics is in actionable conclusions that comes in few forms.
The main benefits of real-time data tracking are speed of knowledge and opportunity to react fast in a given circumstance (e.g. optimising a campaign in case you are out of stock, or the competition is running promotion that can kill your product sales). Post-action analysis is great for lessons learned and overall future planning of actions. And specific targeting is specific tactical analysis that should target one segment of the e-commerce ecosystem (such as a basket suggester, or cross-sell and up-sell analysis).
At BNG we would advise to follow best practices based on actual learnings from some of the big spenders in our region on e-commerce.
1. DISCOVER. Create product segments (based on product performance, market trends, and competitors’ activities) and prepare further recommendations adjusted to a specific group of products.
2. IMPLEMENT. Execute recommendations.
3. MONITOR. Track the performance of your products over time with the support of custom-built dashboards. Define your main competitors and KPIs (listing position, search visibility, prices, sales, conversion, traffic, content completeness, share of voice, and others). Create automated notifications when urgent actions are required (e.g. the launch of a new product by competitors requires an aggressive advertising strategy targeted on specific keywords);
4. MEASURE. Measure the isolated effect of the implemented changes. Check whether your products have achieved the benchmarks for predetermined KPIs. Rearrange the strategy to the changing market and competitors’ strategies when needed.
Despite the multiplicity and complexity of data sources, a structured framework around data collection, data analysis and insight generation can go a long way in generating meaningful business impact. At the end of the day, it’s all about connecting the dots.