Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Upcoming SlideShare
Embrace the Known: Predictions for the Future of Personalization
Next

Share

Improving Revenue through Dynamic Product Recommendations

We power tens of millions of product recommendations per day on behalf of our clients. Needless to say, we've learned a lot about what works, how and where to implement recommendations in the purchase funnel. Global retailers who have implemented Dynamic Yield’s recommendations have seen a substantial uplift in revenue within the first 90 days. Your site can do it too. Contact us to see a live demo.

Related Books

Free with a 30 day trial from Scribd

See all
  • Be the first to like this

Improving Revenue through Dynamic Product Recommendations

  1. 1. All Users Recommendation Clickers Visitors who engage with Product Recommendations generate 280% higher revenue per visit compared to all users RevenuePerVisit $20 $15 $10 $5 $0 +280% Users with 2+ PV’s Key Insight: Product Recommendations are an integral part of the shopping experience. Maximize exposure to recommendations by optimizing widgets with the most effective number of items, layouts and strategies.
  2. 2. Conversions 2.3X AOV 1.2XRPV 2.8X Revenue = Conversion Rate x Average Order Value Compared to all site visitors, 
 users that engage with product recommendations yield:
  3. 3. RevenueAfterClick ClickRate Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8 Day 9 Day 10 Day 11 Day 12 Day 13 Day 14 Day 15 Day 16 Day 17 Day 18 Day 19 Day 20 Day 21 Day 22 Day 23 Day 24 Day 25 Day 26 Day 27 Day 28 Day 29 Day 30 $4 $6 $8 $10 $12 $16 $18 $14 9.6% 9.0% 8.4% 7.8% 7.2% 6.6% 6.0% Revenue After Click Click Rate Key Insight: Instead of optimizing recommendation widgets for CTR, optimize for revenue metrics like Revenue per Visit, Average Order Value, Lifetime Value, etc. Higher CTR ≠ Higher Revenue (In Product Recommendations, higher CTR does not entail higher revenue)
  4. 4. Diverse Recommendation Strategies Only Similarity-Based Strategy Only Popularity- Based Strategy +14.7% +11.2% ConversionRate Conversion Rate by Recommendation Strategy 7.0% 6.6% 6.3% 5.9% 5.6% Key Insight: Use multiple strategies in a single recommendation widget in order to maximize revenue. 
 Combine different recommendation strategies to identify the highest yield specific to each product category. Similarity-based recommendations (e.g. bought together) Popularity-based recommendations 
 (e.g. hottest, most viewed etc.) Example of diverse recommendations 
 (using multiple strategies within the same recommendation widget) Multiple Recommendation Strategies in a single widget produces better results
  5. 5. Diverse / Contextual 
 Strategies Personalized 
 Strategies Diverse / Contextual 
 Strategies Personalized 
 Strategies Conversion Rate for the Average User Conversion Rate for Engaged User Segments 
 i.e. with rich purchase history -7.3% +7.7% Personalized recommendations refers to a class of algorithms that takes the user’s past site behavior, purchase behavior and/or CRM data into account 6% 5.5% 5% 4.5% 4% 6% 7% 8% 9% 10% For engaged users, personalized strategies outperform all other strategies Key Insight: Onboard all available transactional and behavioral data - online, offline, CRM - to develop a rich data set and tailor your recommendation strategies.
  6. 6. No Price Consideration Price has no impact on recommendation algorithm Revenue Per User +13.77% -3.34% Moderate Price Consideration Price has moderate impact on recommendation algorithm. Medium range products are pushed more in this approach. Aggressive Price Consideration Price has heavy impact on recommendation algorithm. 
 More expensive products are included in this approach. $80 $70 $60 $50 $40 Take price into consideration when recommending products, but don’t overdo it Key Insight: Test price consideration (medium to high) to identify the sweet spot that maximizes revenue.
  7. 7. Algorithmic Fusion Combine multiple recommendation strategies within a single widget to maximize performance.1 2 3 4 5 Dynamic Yield’s Personalized Product Recommendations Contextual Recommendation Layouts Change the layout of recommendation widgets based on user’s visit context, traffic source, purchase history, etc. (e.g. for 1st time users with no history, show widgets with more products to infer interest) Omni-Channel Recommendations Place product recommendations on any page or any channel. (e.g. on Homepage, category pages, product pages, cart pages, pop-ups, navigation menus, mobile apps, email & more) Data-Driven Merchandizing Create recommendation rules that automatically update widgets according to product availability, price changes, real-time behavioral interactions, local weather forecast, etc. Flexible Merchandizing Rules Give power to your merchandizers by giving them a flexible and easy-to-deploy merchandizing rule builder that allows you to pin, suppress and exclude specific items in the automated results.
  8. 8. Our recommendation engine assesses the level of valuable data about an individual visitor and deploys the most appropriate strategy. Gathered information per user Revenue Popularity Predictive Personal Automated Segment-Driven Strategy Selection New Users Occasional Users Frequent Users Savvy Users
  9. 9. Global retailers who have implemented Dynamic Yield’s recommendations have seen a substantial uplift in revenue within the first 90 days.
 Your site can do it too.  SEE A LIVE DEMO

We power tens of millions of product recommendations per day on behalf of our clients. Needless to say, we've learned a lot about what works, how and where to implement recommendations in the purchase funnel. Global retailers who have implemented Dynamic Yield’s recommendations have seen a substantial uplift in revenue within the first 90 days. Your site can do it too. Contact us to see a live demo.

Views

Total views

3,357

On Slideshare

0

From embeds

0

Number of embeds

1,035

Actions

Downloads

0

Shares

0

Comments

0

Likes

0

×