retail detail omnichannel congress 2015 - data science for e-commerce

35
Veldkant 33A, Kontich [email protected] www.infofarm.be Data Science Company Data Science for e-commerce RETAIL DETAIL OmniChannel Congress 05/02/2015

Upload: infofarm

Post on 19-Jul-2015

491 views

Category:

Data & Analytics


2 download

TRANSCRIPT

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Data Science Company

Data Science for e-commerce

RETAIL DETAIL – OmniChannel Congress05/02/2015

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Agenda

• About InfoFarm & Elision

• What is Data Science?e-commerce vs Data Science vs BigData

• Example Data Science applications in e-commerce

some inspiration to spot your opportunities…

• Applying Data Science

how to get started with all this?

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.beData Science Company

About InfoFarm

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

InfoFarm – Mixed Skills Team

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

InfoFarm + Elision – e-commerce!

Expert Hybris partnerin omnichannel

solutions in B2C & B2B

SAP/Hybris Nominee Service Delivery Partner of the year 2015 EMEA

Highly focused on Data Science and

Big Data

Technical Knowledge

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.beData Science Company

Introduction: what is Data Science?

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

What is data science?

• Data Scientist: “A person who is better at statistics than

any software engineer and better at software

engineering than any statistician”

- Josh Wills

• “Getting meaning from data”

Finding patterns (data mining)

• Complementing business

knowledge with figures,

enabled by IT practices

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Data Science & Big Data

• Relevance for e-commerce - use data insights to:

– Increase conversion rate

– Increase operational efficiency

– Understand your customers’ needs

– Make better offers

– Make better recommendations

– …

• Many successful online businesses thank their position

to smart data usage:

– Google was the first search engine that didn’t rank by keyword

– Amazon is the e-commerce leader thanks to BigData practices

– NetFlix is a world leader in personalized recommendations

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Data Science & Big Data

• Classical presentation : the V’s in BigData

• Can be very hard to make the leap from this technical

capability towards the added value for your business!

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Data Science & Big Data

• Most of us don’t run a business like the ones referred to in stereotypical Big Data cases

• Big Data does not necessarily mean or require much data

• Data Science is very affordable for companies of all sizes

• Typical Data Science projects are 10’s of man-days of work

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Data Science & Big Data

• Non-structured data: weblogs, social media content, …

• Secondary use of data sources is the key

– Example: Weblogs

• Are there to log webserver activity

• But can also tell you how people find, compare and choose products!

– Example: ERP / Cash register software

• Prints bills

• But can also tell you what products are typically bought together in a shop

• Much data is present, valuable information is hidden in it!(aka “dark data”)

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

OmniChannel considerations

• Customer intimacy leads to customer loyalty, referrals, …

Combine data from online and offline channels for a holistic view

• There is a lot of unused information on customers !

– Does your store personnel have any idea if a customer has bought things via

your online touchpoints? Or has been looking for specific items recently?

– Do you take contact moments (in-store, callcenter, …) into account for the

personalisation of the online touchpoints?

• Key enablers:

– Ability to ID customers in-store (purchase? visit?)

(personal contact, CRM, ibeacons + mobile app, loyalty card, …)

– Ability to ID customers online (quite obvious if logged in)

– Ability to combine / integrate both

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Data Science – some key techniques

Data Mining - Machine Learning algorithms applied to business data

• Clustering data – finding similarly behaving/interested customers…eg: can be used to determine marketing segments

– Do we segment our customers on (assumed) business knowledge?

– Or do we cluster them in segments based on their behaviour?

• Predictive modeling – finding factors and their weightings that predict customer behaviour with a certain probability

eg: can we predict the probability that a sale will be made, an order will be cancelled, etc… based on historical information.

• Classification algorithms – auto-classify data in predefined labels.

eg: Can be used to auto-classify messages in question, complaints, etc… or products in budget-class, high-end, etc …

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.beData Science Company

Example Data Science applications:

#1: Recommendations

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Recommendations – Why? How?

– Why?• Attempt to cross-sell or up-sell

• Provide customers with alternatives that might please them even more

– Traditional approach• Making no recommendations at all

(would you be happy with such a salesperson? why then online?)

• Products in the same category

• Manually managed cross-selling opportunities per product

– Why are these approaches fundamentally flawed?• They all start from the seller perspective, not the customer!

• “We know what you should be buying”

• Manual recommendations are too costly and time-consuming to

maintain – even impossible with large catalogs

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Recommendations

– Product based recommendations

• Main focus on online, but can be elaborated to in-store?

• Who knows best what products to recommend?

• Learn from your data, don’t take decisions based on a feeling

– Time based recommendations

• Recommend or cross sell different products depending on

– season?

– holiday?

– weather?

– Customer based recommendations

• Learn from your customers and their past.

• example: Android vs iOS smartphones.

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Recommendations – Omnichannel

– It’s the same customer (more or less) online and in the store

– It’s about the same products!

– How to align:

• In-store recommendations and personal, professional advice

• Online automated recommendations and advice

– Some ideas:

• Analyze cross-selling realized in-store and online

• Are there any differences? How come?

• Can sales person use the information from the online touchpoints to their

advantage? (browsing history, …)

• Can cross-selling information realized in-store by used to optimize the online

touchpoints?

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Recommendations – example

• Ex: Buying paint material: paint, roller, brush, primer, tape, …

• Omni-channel: try to combine advantages of both for all customer profiles (Online viewers, in-store buyers and vice-versa)

• Customer perspective– I hate painting - I don’t want to spend much time on this

– What did others buy? I can’t be the first to do such a paint job?

– I don’t want to forget an item I’ll need

• Customer wants information on all needed products that fit together– In-store: professional advice

– Online: <here you can still make a real difference>

• Fact: most paint-selling sites don’t even do recommendations!The ones that do could go a lot further if it wasn’t so labor-intensive to maintain.

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Recommendations – many questions/opportunitiesWhich

similar productsto show?

Color alternatives?Glossy/matte alternatives?

Cheaper/better?

Which complementary

products to show?Link to category

without match with specific product?

Which brush would be appropriate for this

paint?Which primer?

Related Products

Similar / alternative ProductsCurrent Product

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Recommendations – what does Amazon do?

Cross-selling as realized with other (similar?) customers

Starts from customer point of view!

Recommendations based on perceived customer journeys

Re-use the product comparisons that

previous customers made!

DATA DRIVEN!

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Recommendations – other ideas

• Data Science ideas

– “x % of the people who viewed this item eventually bought product X or Y”

– Get cross-selling information from ERP in the physical shops and let this feed the

online recommendations!

– Similar product in different price ranges

(“best-buy alternative”, “deluxe alternative”)

– ...

• This is very achievable for a webshop of any size

– Just generate ideas, and test to see what actually increases sales!

• Secondary use of various kinds of non-structured data = BigData !

– Weblogs of e-commerce site (use to deduct customer journeys)

– ERP info with bills and/or invoices (use to deduct cross-selling in physical shops)

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Recommendations – other ideas

• Auto-combination special offers based on cross-selling

infoWhy not give a small discount if bought

together?

Testing will show if and for which products and

customers this increases revenue!

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.beData Science Company

Example Data Science applications:

#2: Personalized offerings

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Personalized offering

– Browsing habits and patterns.

– Spending patterns.

– Personalized discounts and/or content?

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Personalized offerings

• Customer should be central in the approach– Provide a truly personalized online shopping experience

– Like high-end stores with personal approach to VIP customers

• Gather data about your customer (on- and offline)– Surfing history – what products were viewed? For how long? …

– What products were bought? When?

– Brand preference?

– Product-segment preference? (budget, high-end, best-buy?)

– Abandoned shopping carts

• Take action based on information mined from this data– Triggered e-mails, personal recommendations, …

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Personalized offerings – some ideas

• Anticipate customer behaviour:

– Use all customer contact moments

eg: if customer calls customer service, they should know what

products the customer was looking at during his last visit to the

webshop

– Build prediction models

(surfing behaviour vs % deal-making)

eg:

Low chance? Go to checkout immediately.

High chance? Offer extra cross-selling opportunities

Possible with well-known machine learning techniques!

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.beData Science Company

Example Data Science applications:

#3: Anticipatory shipping

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Anticipatory shipping

– Patent pending by Amazon.

– Ships an order before it is placed.

– Based on order history, search, wish list and click behavior!

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Anticipatory shipping

• High-tech? Actually not that complex at all …

• Steps:

– Gather many info on past orders

(customer info, country, product info, price, product group,

product combinations, time of day, season, …)

– Build a prediction model predicting “cancelled or not” based on

all this information

– Assess the quality of the model by training it with 90% of your

historical orders and testing it with 10% of your historical orders

– Pass each potential orders’ info and predict the likelihood of it

getting cancelled (0 .. 100%) and act accordingly

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.beData Science Company

Example Data Science applications:

#4: Customer Service optimizations

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Customer service

– Losing sales/conversion/money by poor customer service.

– Optimize information for all communication channels.

– What issues are your customers concerned with?

– Allocate resources in a better way

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Customer Service – Some Ideas

• Text mining– Mood analysis: detect negative messages on social media, forum, reviews, …

Put TODO on action list of customer care to contact with certain priority

– Auto-classification of e-mails, letters, messages: Is this e-mail a question or a complaint?Is it about the quality of the product or financial (wrong invoice, …)?Automatic routing of messages to the right person! (operational optimization)

• Social media– Social media status of customer (scoring based on profile)

What’s would be the impact of this customer being unhappy about our service?

• Omnichannel insights– What did this customer buy or view?

– How did he rate the last products (online) he bought (in-store)?

– Which contacts (brochure, mail, phone, …) did we have and what seems to be the most effective deal trigger?

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.beData Science Company

Applying Data Science

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.be

Applying Data Science

• Data Science does not replace business knowledge– Need to find balance between the two– Confirm or deny assumed business knowledge

– Detect changing trends early (customer behaviour, …)

• Not a development cycle, rather exploratory process:– Formulate hypotheses

– Data mining and modeling

– A/B testing (test new idea on x % of your customers/products/…)

– Conclusions: did the test group show better conversion?

– Rollout or cancel and start over!

• Potential issues– Privacy law and other legal restrictions

– Feedback loops, information leakage, wrong assumptionseg: trying to gather customer preferences when an order could as well have been a gift to someone else (perfume, …)

Veldkant 33A, Kontich ● [email protected] ● www.infofarm.beData Science Company

Questions?