Personalized Recommendation Magento Case Study

Personalized  recommendationsHow  to  help  customers  do  the  shopping  in  

your  Magento +  www.quiosque.plCase  Study

• Why  e-­shop  needs  recommendations• Magento  recommendation  out  of  the  box• 10  best  recommendation‘s  strategies  that  is  NOT  Included  in  Magento

• Cases  &  Examples

E-­commerce  manager  pain

How  to  help  customers  buy  any  product?

• How  to  increase  conversion  rate?• How  to  maximize  cart  value?• How  to  speed  up  my  IT?• How  can  I  have  more  time?  

Personalized  recommendations

WHEN  YOU  HAVE  1M  VISITORS  ONLINE,  YOU  SHOULD  HAVE  1M  VERSION  OF  YOUR  E-­

SHOP

How  we  define  recommendations

Showing  additional  product  to  e-­shop  buyers  when  they  are  shopping

UpSelling,  Related  products,  Cross-­selling,  Similar,  Alternative,  trending,  bestsellers,  

personalized,  sold-­out,  long  tail

Similar  products  recommendations

Related  products  recommendations

Up-­SellingProducts  you  wish  to  your  customer  to  buy  instead  of  the  

product  that  they  are  viewing

They  might  be-­ More  expensive  -­ Higher  margin-­ Regular  price

Cross-­selling

Products  to  be  purchased   in  addition  to  the  product  being  view  or  added  to  the  cart

Magento  out  of  the  box  recommendations

Product  Page-­ UpSelling-­ Related  (cross-­selling)

Cart  Page-­ Cross-­selling  (impulsive)

Manual  recommendation  settings  

Recommendations  can  be  automated  

Machine  Learning  algorithms

Collaborative   filteringContent  based  filteringAssociation  rulesNeural  Networks

Black  Box  engine  algorithms

BIG  DATA

Real  time  personalization  algorithms

Visitor  browse  e-­shop  webpage,  each  click  does  matter  and  is  collected  by  the  reco engine.

AI  smart  mechanism  learn  immediately  the  visitor  needs  based  on  his  behavioral  data.

Recommendation  engine  predict  and  decide  automatically  what  products  the  customer  might  like  and  display  personalized  offer  to  him  in  the  real  time.

AI  based  Recommendations

What  with  Furby?

What  with  balaclava?

Strategies  that  work1. Personalized  landing  page2. Dynamic  trends3. Smart  Cross-­selling  4. Smart  Up-­selling  5. Similar  products6. Zero  search  recommendations7. 404  – page  recommendations8. Add  to  cart  cross-­selling  pop-­up9. Impulse  Cross-­selling10. Personalized  promotions

Case  study

How  www.quiosque.pl is  enriched  with  personalized  recommendation  engine  -­ QuarticON

Step  1  -­ Integration

<script type="text/javascript">(function(){ upsParams = window.upsParams || [];upsParams.push(['customer','xxxxxxxxxxxxx']);var _qS = document.createElement('script');_qS.type = 'text/javascript';_qS.async = true;_qS.src = '//rec.quartic.pl/qon.js';(document.getElementsByTagName('head')[0]||document.getElementsByTagName('body')[0]).appendChild(_qS); })();

</script>

One  script  before  </HEAD>

• Tracking  user  interaction  data• Product  feed  crawler• Widgets  displaying

Step  2  – strategies

Step  3  -­ The  results

12%  total  sales  generated  with  recommendations

Cross  selling  Pop  Up  – Must  haveCTR  15%,  Conversion  5%

Thank YouPaweł  Wyborski,  Founder at [email protected]