One of the most familiar quotes we hear from retail and ecommerce teams using WooCommerce is that
“Generally, different teams have unique definitions of metrics or KPIs. They’ve been built up quite naturally from the different sources and goals our teams have. Ops and LP are watching revenue in proportion to shrinkage. Ecom monitor revenue vs acquisition spend. We don’t have a single source of truth to build everyone’s targets off, in a unified way. Our executive team spend too much time de-tangling the overlap between reports to get on with building tests or making decisions”
What would seem like basic questions, quickly become too hard to bother with, so everyone reverts to doing it the same way as last month. But there’s analytical gold to be mined asking questions like:
- Do customers from campaign A have a higher or lower forecast LTV?
- Which channel sends customers with high propensity to spend, and how can my CMO focus budget on acquiring more like them?
- Do choices made in marketing and merchandising move the needle on overall customer spend?
- What is the most efficient use of my marketing & advertising budget to maximise growth without harming loyalty?
WooCommerce is an ecommerce platform with more than 650,992 active users according to Builtwith.com, the ecommerce technology monitoring site. As the go-to wordpress ecom offering, it benefits from a large development community, and an easy-to-use interface. But it’s not an analytics or BI platform, and pretty soon an ecommerce business that wants to drive sales will hit the limit of its capability.
Reporting does support a high level of detail at the individual transaction level. But all the intermediate metrics that actually mean something for decision making are in the main, absent.
Measure profitability of marketing channels? You’ll need to have Indiana Jones-level skills to walk the rope-bridge between GA/Site Catalyst & your WooCommerce installed reporting.
Calculate average time between purchases? Not supported. And these metrics are vital for guiding your customer acquisition and retention strategies. With SEO and PPC channels wrung for every last drop of value, the new approach is to own these intermediate metrics, join the silos they sit in, and build your new growth strategy from them. Otherwise, you’re in the dark about the metrics that your competitors are profiting from.
Imagine if you knew precisely which products acted as introducers to a longer relationship with your brand? What the new acceptable CPA (cost per acquisition) should be for customers buying those products, since many of them will become loyal in the long-term?
These actionable take-aways come as standard when you’ve matured your approach to data science, marketing, and measurement using Vuzo’s tools.
Bundles and refunds are two big areas where big growth opportunities sit. Most of the time however it’s too laborious to unpick the impact of merchandising choices for bundles, so ecommerce managers just press ahead and do ‘what worked last time’. This is not sustainable. With data science you can spot the 10 most common items purchased together and run individual tests to work out which is the most persuasive offer that lifts AOV highest. And remember, not all customer segments buy all bundles – do you have a dashboard for that yet?
Do you know which products are most commonly in your abandoned carts this month? Why is that? Are they just slightly under the free shipping threshold or are they simply cheaper from a competitor? Again, there’s money on the table here for the retailer who is on top of their data.
Then, it’s time to bake in some margin. Without including a consideration for net profit per SKU or per category, big ideas and campaigns that drive volume can leave head office feeling less than overjoyed. These metrics need to become standardised so that marketing, ops, and merchandising all pull together at the common goal.
And speaking of which – do your teams even know what the common goal is? It’s hard to align and bonus everyone to increase market share at all costs, if you don’t have reliable metrics for market share. Getting all teams to pull on the Quarterly Gross Profit rope together is much easier when there’s one figure they can all see. Accountability, not to mention morale, is much easier to engender when there’s a clear link between what a person does, and how it helps team and company achieve the goal.
And finally there’s the anomaly detection aspect which is not supported at a deep level in WooCommerce out of the box.
If like me you hate the idea of reporting for the sake of it, then adding a ‘priority’ layer to reports brings a lot of value. If each WooCommerce data report contains 5 recommendations or items that need to be investigated this week, then you go a long way to incrementally improve ecommerce performance, and give a fulfilling set of tasks to the front-line operative, reducing management overhead needed for constant spoon-feeding.
Another part of maximising revenue from WooCommerce is by using AI to build per category forecasts, then use automated monitoring of metrics for detection of opportunities/issues on individual products and categories, to maximise revenue and on-site conversion rate. E.g. gloves sales were predicted to trending up in October but actually they are static and instead the bounce rate is trending up… This could trigger a content review.
What are your most re-ordered products, and who buys them? Is it time to re-orient your content marketing strategy around these products and buyers? What is the expected ROI from increasing sales by 10% on those lines? You won’t know until you dive into the data – and you certainly can’t bonus the team on achieving the increase, if there’s no credible baseline they can see and agree on.
Products such as ZigZag are revolutionising the returns space, and rightly so. But if there’s no returns line in your monthly dashboard, then it may be too late to detect a faulty product set or a drop in standards at the courier – the affected customers have already walked into your competitors open arms.
So, make sure you use data science in these 3 key ways
- Define some KPIs that are profit oriented, and extract them regularly from your data
- Share this insight to the relevant teams so they are all aligned on the same goal as the C-suite
- Find out your net profit from each marketing channel, and for bonus points, work out the average LTV of customers coming in from each
- Tie marketing spend dashboards to revenue and profitability by item, and then build smarter marketing tests that are built on profit not volume
- Simplify the output so that one or two key messages per month are shared up to the board, and down to the front line. Drowning everyone in a million new metrics will mean low adoption of data driven decisioning while you’re around.