How to improve conversion rate of online Lifestyle shopping platform?

No single, best answer to this question! Online Lifestyle product shopping and customer experience have matured over more than a decade now and the preliminery deterants like trust and quality have been addressed by most of the leading platforms and the so called rules/ standards of doing this business are also laid out. Millions of dollars are spent on awareness and improving the online shopping experience so that more and more users shop without any fear or concerns. Still the online conversion rate for most of the leading platforms is an area of concern. It reflects a lot about the brand and overall perception of the platform.

Generally, most of the online platforms are perceived as "Discount" platforms by customers and usually people shop for cheaper price and they compare online prices vs store before shopping. In a price competitive market, "price" becomes the most influencing factor and lot of money needs to be burnt to provide price discounts to acquire a customer. In this scenario, an online shopping platform can never be profitable until other 'X' factors influening conversion are identified, measured and optimised to the extent that price war can be managed. This article is all about rest of those 'X' factors.

The article is purely based on my experience and standard best practices of conversion rate optimization or growth hacks are deliberately not covered here (as there are tons of articles available on that). Inspite of that you might still find most of the article with obvious or known measures of Conversion factors but for sure are the ones which matter.

How's e-Commerce platform conversion computed?

Conversion = Orders/ Visits; where as, Order is an online shopping transaction and Visit is an online session

There are many factors which impact Conversion but we will look at those which impact the vital stages of the Conversion funnel -

Stage-1: Awareness

Its just not about telling everyone about your brand but also marketing Category and Brands of products available on your platform. Its not just about telling this to everyone but only to the target audiences and the ones who have very high intent to purchase a certain brand/ category of product. To achieve higher returns on Ad spend the above two guidance are to be followed as thumb rule.

X Factor: Traffic Quality - They say, "garbage-in, garbage-out", which means if the quality of online traffic (and quality means "intent") is not good, your chances of order conversion is equally bad.

To measure the quality of traffic you should look at the following KPI/ Metrics by traffic channel and put your money and effort on the best performing channel.

KPIs - Visits, Product Views/Visits, PageViews, Exits, Bounce, Session length, App Installs, Installs with Logins, Daily Active Users, Monthly Active Users

Organic channels like "Natural Search" or "Direct traffic" always convert better than in-organic, paid, channels as inherently "intent" is high in them. The "Direct traffic" volume is directly proportional to the amount of Brand awareness campaigns done. The "Natural Search" volume is directly proportional to the amount of Search Engine Optimization (SEO) done.

KPIs to be measured for SEO performance - Keyword Ranking, Organic ClickThroughRate, Organic Conversion, Core Vitals (FCP, LCP, FID from GoogleSearchConsole), Indexed Pages #, Referring Domains #

Tip - try not to send paid campaigns to already installed customer base and notify them using 'Push' notifications. An optimal use of paid channels will increase your "Return on Ad Spend (ROAS)".

Esentially, you need to get volumes of good intent traffic!

Stage-2: Consideration

This is when a Product View happens. In an online session, more "Product Views" would mean more engagement and there's a very strong, positive, correlation to the number of orders placed. In other words, if there are fewer products to be viewed then the chances of finding the desired product are fewer.

In the offline parlance, "Product Views" would mean "Eye Balls" and more products which are kept in the "Eye-level" will get more "Eye Balls". In online parlance, this would mean, the Site Merchandising needs to provide visibility to the "High-demand" brands/ categories by placing them in the "high-visibility" slots, thereby, increasing the chances of more "Product Views".

How do we measure Site Merchandising efforts to increase "Product Views"?

I suggest two KPIs here -

a) % of "Promoter" brands in Merchandised list of brands; a "promoter" brand is a brand which is in demand (gets more Product Views) in non-Sale period as well; to identify your "promotor" brands compare "Product Views" of few good days (with high Product Views which has to be a non-Sale day and consider only organic traffic channel) Vs few bad days (with less Product Views on a non-sale day and organic channel) and filter out those brands which show more Product Views between good and bad days. I took a sample of 2 good and 2 bad days to compare, you may derive your sample size.

b) Avg # of clicks to reach Product Page; from your Digital Analytics tool, you can get the "Depth levels" to your Product Page for every visit and do a weighted average of that. This has a very strong inverse correlation to ProductViews. With slightest of increase in this KPI the number of ProductViews will decrease.

Another KPI which impacts the "Product Views" is "Relevance". There are thousands of products in the catalog but only a few would be relevant to the customer. Your Relevance engine should identify the best mix of product listings basis chances of conversion, demand, profitability, etc. The better is the Relevance so will be the number of "Product Views". Statistically, they are highly correlated.

How do you measure "Relevance"?

I suggest, the following three metrics -

1. PLP CTR (Product Listing Page ClickThroughRate); more Relevance will mean more PLP CTR as there's a very high positive correlation between them
2. Search CTR (Internal Search Listing ClickThroughRate); more Relevance will mean more Search CTR as there's a very high positive correlation between them
3. % of Products with ProductView; better Relevance would mean more CTR which eventually translates to more Orders, which means the Relevance engine, if working perfectly, should get your most relevant products stocked out fast and which will make space (in Ranking) for the next relevant product. This should follow a FIFO queue for more and more products to get ranked higher and get more Product Views.

X factor : Size and Variety of your Product Catalog - influences "Product Views". The product assortment matters - freshness, styles, product range, brands and categories!

How do you measure the "Health and Strength" of your Catalog?

I suggest the following metrics -

1. Inventory distribution of Catalog by Freshness
2. Catalog density at L1 level = total listed Styles/ total # of listed brands
3. Catalog width at L2 level = Avg, min, max and median brands at category level
4. Catagory depth at L2 level = Avg, min, max and median range of style#

All these metrics are positively correlated with Product Views.

Stage-3: Preference

This is when a Product is added to Cart. Primarily, customer will prefer to add a product to cart if the customer likes the style, finds the right size in stock, deliverable at the location within expected number of days and finds it in expected price range.

How to measure these factors?

1. Affinity of Style (has very strong positive correlation to Cart Additions) - After a Product View, if a product is not added to Cart inspite of meeting all the factors from Availability to Delivery TAT, can be safely assumed to be due to affinity of Style.
2. Availability of right size and style (has very strong positive correlation to Cart Additions) - You should tag an event for "unavailability of size for a certain style" in your DA platform and compute a metric as "% of Product Views with Out-of-stock error"
3. Price expectation match (has positive correlation to Cart Additions) - Avg Price of viewed Product without Order (by category and brand) vs Avg Order Value of Category and Brand
4. Pincode Serviceability Issues (has very strong positive correlation to Cart Additions) - You should tag an event for "undeliverability of product at a certain pincode" in your DA platform and compute a metric as "% of Product Views with Pincode Serviceability Error" and in turn derive the missed Conversion Opportunity (due to Pincode error)
5. Promised Delivery TAT (in my experience this doesnt have much correlation though with Cart Additions but highly recommended to find the correlation for your platform; theoritically speaking this should have a strong negative correlation) - You should tag an event while showing the "Promised TAT" and also capture the "Promised Number of Days" in your DA platform; compute the weighted average of shown "Promised TAT" for Sessions with Cart Additions vs w/o Cart Additions.

Stage-4: Intent

This is when a Product is Checked Out. The Checkout action almost guarantees that customer has 100% intent to buy the product but what are the factors which are in consideration before making the check-out. The two most important X factors which impact the Conversion the most are in the lower most part of the funnel -

1. Promotion - as I called out at the beginning of this article, this market is driven by promotions and deep pockets to burn more and more money; customers rarely shop without promotions no matter how insignificant is that. In order to measure the impact of promotion on Checkouts, i recommend, computing - Promotion_Burn / Gross Merchandise Value ratio. As the ratio goes up, the Checkouts go up and vice versa.
2. Delivery charges - no matter what price segment your customer falls, paying "Delivery charges" hurts! Many a times (and there's a strong correlation) customer wont check out if they find the "Delivery charge" unreasonable. To measure what % of missed orders would be due to that, compute - % of product view with Delivery Charges and if that number grows it negatively impacts the Checkouts.

With that I have some high-level guidance to provide before you take on your Conversion improvement journey -

1. ensure your Digital Analytics tagging is razor sharp and accurate; any error in capturing the raw events will lead to blunders as all your analysis will be basis incorrect data
2. always remember, something which can't be measured can't be improved; don't live in utopia and set and define measurable goals and targets
3. Conversion improvement of a e-commerce platform is not a onetime activity and there are more number of moving parts than you have found; not everything is what your DA or BI platform shows you - there could still things which matter but couldnt be tracked as they couldnt be measured
4. Not necessarily everything can be a zero sum game; at times you might derive a directionally close metric which is not tighly bound to goal
5. Process of Conversion improvement should be - Define Measure -> Track -> Optimise and Measure again -> repeat until you find an improvement

All the best!