Description

Your list might include some of the following:

• Conversion rate
• Average order value
• Promoting new products
• Clearance 
• Sale items
• First price sales
• Coordination with marketing campaigns
• New customer acquisition
• Brand experience
• Look and feel
• Supplier stipulations, brand adjacency etc.

That’s by no means an exhaustive list and already you’re probably starting to see potential conflicts here. If you push discounted products hard it may drag your AOV down. Pushing best sellers down the page to promote unproven new products might impact your conversion rate. Coordination with a marketing campaign geared towards your loyal customers may not work so well for acquiring new ones.
It should be obvious then, that trying to satisfy all these drivers at once is going to end up with a dog's dinner. One merchandising strategy across the whole store won’t work.
A better approach then is to shape a merchandising strategy around each of your most important drivers and apply them to pages accordingly.

For each page, decide one key driver and apply the appropriate strategy. 

“The outcome we want for this page is to boost sales of clearance products so we put clearance products at the top”
Great, did you sell more clearance products?

Another important question to ask yourself is “how will I know it’s working?”

For each of your drivers and strategies you should choose something measurable and make sure you can measure it.
Some are easier than others, AOV, conversion rate etc are simple measurements to make. You may have to do a little work to measure at the level of a specific page but don’t be tempted to just assume success because what you’re doing seems to make sense.

One Algolia customer found that promoting new products on some category pages didn’t increase sales of new products and marginally reduced sales of perennial best sellers but having a dedicated section for new products allowed regular shoppers to quickly browse the latest items.

In an A/B test we ran with a luxury goods retailer we found that over optimisation for conversion rate on certain pages reduced the sales of high price, high margin products.Although conversion went up the shift in product mix pulled down overall revenue.

It’s not always intuitive and without good measurement you can easily leave money on the table.

Differing merchandising objectives and strategies for pages can be a little daunting if you have a large catalog, so grouping them into sets with similar objectives can streamline things and make it more manageable.

The next level up from per-page strategies is to work customer segments into the picture.

If you have a good customer data platform consider varying your approach for different customer types. The details will depend on your customer data but an example might be showing a different product mix to people who have never purchased before to drive customer acquisition. You could make the same page favor higher priced products to previous customers who are already comfortable with your business.

Customer segmentation can help you manage conflicting drivers; highlighting product featured in a marketing campaign only to customers who came to your store from that particular campaign for example. 

Segmentation can add a lot of value but individualized tweaks to a page for every customer are possible without an awful lot more effort. Building a customer profile based on browsing and shopping behaviour allows you to weight the product mix and positions towards products the customer may like based on attributes like color, brand etc.