Bus 421 Marketing Analytics

areas of AI

○ Machine Learning ○ Deep Learning ○Artificial Intelligence

Measuring ROI and ROMO

○ Marketers want to know if they are spending on ways to get the most impact with media overall ○ Don't want to overspend on an invaluable media (Waste of money) ○ Don't want to stick to miss out on a better media/ opportunity (Waste of impact) The Foundation of Media Planning

Proprietary programming tools:

○ Programming tools that are developed by a firm and distributed for sale to the public.

Open-source programming tools:

○ Programming tools that are made freely available, often developed by and for the community.

●Native Social Listening Tools

○ Some social media outlets provide detailed analytics about who is interacting with your content (example: Facebook) ▪Analytic tools provided direct by the social networks ▪Insights about your Facebook Page •Likes/ New Likes / Net Likes •Where Likes came from •Reach •Demographics •........

statistical programming tool

○ a programming tool focused on statistical analyses of data.

programming tool

○ a software package that allows for the execution of a programming language.

Non-programming software for data

○ specialized computer programs for analysis using a graphical interface

Programming software for data

○ specialized computer programs for analysis using programming code

MAX

○Generates the largest value that fits a specified criteria.

Document Store Databases

○A database that uses complex data structures known as documents for storage and queries; similar to key-value but the key is paired with a document instead of a value. ○Benefits - schema free, documents can have different structures, fast write performance and fast queries

Graph Databases

○A database that uses structured relational graphs of the interconnected key-value pairings instead of relational tables ○Consists of nodes, edges (relationships), and properties

BETWEEN

○Displays Data between a specific range of values. Example: One can select the customers between the ages of 18-25

AND

○Displays data that is a combination of two conditions. ○Example: One can display the customers between the ages of 18-25 and live in California.

AVG

○Generates the average value of a numeric column. Example: One can determine the average age of all customers

MIN

○Generates the smallest value that fits a specified criteria. ●Example: One can determine the customer who made the smallest or biggest purchase last week.

SUM

○Generates the sum of a numeric column. ●Example: One can retrieve the summation of sales made in a certain year.

Frank

○Pay-Per-Click tool ■Uses Machine learning to find best paid advertising channels depending on specific audience

data integration

Combines data from multiple sources: 1.Relational databases 2.Data management platforms 3.Statistical programming languages 4.Commercial data visualization

STRATEGIC METRICS

Companies need an overview of the potential revenue available within the market ●For market entry and exit: ○Market size and marketing growth should be measured ●Return on Marketing Investment (ROI): ○ROI = (Marketing Revenue * Contribution Margin) / Marketing Spending

Inbound Marketing Tools

Creation and management of cross media content and campaigns ●Aid in search engine optimization (SEO) ○Content strategy, keywords and landing pages ●Blogging ○Relevant and conversion optimized content, formatting, link to SEO ●Social media ○Track engagements, schedule posts, monitor conversations ●Produces marketing automation ○Use lead behavior to personalize messages and delivery ●Email ○Templates, personalized content, A/B testing ●Website ○Templates and SEO ●Ads ○Track ROI

Multi-Model Databases

○A database that can support multiple data models against a single, integrated backend ○Examples: ■Amazon DynamoDB - Document, Key-Value ■MS Azure Cosmos DB - Document, Key-Value, Graph, Wide-Column ■Arango DB - Document, Key-Value, Graph ■Orient DB - Document, Key-Value, Graph

UNU's SWARM Insight (Tool)

○ Create algorithms that generate optimized group answers & decisions ○ Each individual tendency, preference, bias, and objection is identified by AI ○ Shows what decisions were made & by which people

Person Level Analysis

○ Create profiles, layer in target audiences for accuracy ○ Includes impact on your objectives ○ Starts with individuals then grows outward + other influential facts The Foundation of Media Planning

●Single-Attribute Social Listening Tools

○ Focuses on a single aspect of a post such as the text or an image (example: TweetReach) ▪Focuses on one social media attribute like text or images facebook, instagram, twitter

Wide-Column Stores

○A database similar to a document database that uses a column-oriented data structure ○They store data tables in columns instead of rows ○Each row can contain a different number of columns than the other rows

preparation

organizational level Per Visualization

BRAND EQUITY INDEX

(Effective Market Share) x (Relative Price) x Durability

Undifferentiated Marketing

- offer the same marketing effort to everyone in the population

tab delimited file

- the columns of data are stored as a text file with a TAB character between values. TAB delimitation has an advantage over CSV: commas inside text values (e.g., "Smith, Rodrigo") can confuse the software about where values actually begin and end.

predictive anlaytics

- the practice of interpreting data to predict the likelihood of future marketing outcomes.

marketing optimization

- the process of refining the marketing efforts of a firm to maximize marketing outcomes.

marketing mix

-Value Creation - product -Value Capture - price -Value Delivery - place -Value Communication - promotion

Market Segmentation

1. Identify the main segments in your data 2. Profile the segments

Which Clustering Variables to Include:

1. Typically people use ALL segmentation variables initially (big as an elephant) 2. People do not typically use outcome measures (like sales, overall satisfaction, etc...) to determine one's cluster. 3. After fitting the cluster solution, you can see which variables are different between the clusters (profile) 4. Then see how each cluster affects the marketing outcomes to decide which to target first. 5. Use marketing mix variables to determine HOW to target each segment.

the scientific method

1. ask question 2. hypothesis 3. research 4. analyze data 5. conclusion.

How to Set Up Instagram Insights

1. select the settings button in upper left hand corner: ▸Under account settings, select "Switch to Business Profile" ▸You need to be public Select Category and Subcategory Enter Contact Information Select Insights tab in upper left hand corner

First Party Data

Data collected by your own organization Usually involves personally identifiable information (PII)

Main Inbound Marketing Companies

Hubspot marketo salesforce pardot eloqua oracle

programming code

○ statements written in a particular programming language

Three Types of Visualization Tools

1.Commercial data visualization tools a.Tableau b.Power BI c.Excel 2.Commercial geographic information system tools 3.Open source tools

4 Attributes of Data Visualization

1.Comprehend information quickly Communicate large amounts of data clearly and cohesively, so that it's easier to draw conclusions. 2.Identify relationships and patterns Recognize patterns that are highly correlated which helps organizations focus on areas most likely to influence their most important goals. 3.Pinpoint emerging trends Get an edge over competition. It's easy to spot outliers that affect the quality of the product or customer churn to address issues before they become bigger problems. 4.Communicate stories. Once a business has uncovered new insights from visual analytics, the next step is to communicate those insights to others.

steps for using data

1.Data Collection 2.Data Storage/Management 3.Data Cleaning 4.Data Integration 5.Data Analysis 6.Data Visualization

How to Set up Analytics on Instagram

1.Edit Profile 2.Try Instagram Business Tools 3.Choose any category 4.Opt out where you can 5.Home 6.Hamburger (upper right) 7.Insights 8.Switch back to Personal Account (hamburger -> settings) Do you see anything that surprises you?

Target Marketing

3. Develop measures of segment attractiveness (which are most profitable? Reachable?) 4. Select the segment(s) you will target

product positioning

5. Develop product positioning for each target segment 6. Develop marketing mix for each target segment

A/B Testing

A randomized group of experiments used to collect data and compare performance among two options studied (A and B). A/B testing is often used in refining the design of technology products, and A/B tests are particularly easy to run over the Internet on a firm's Web site. Amazon, Google, and Facebook are among the firms that aggressively leverage hundreds of A/B tests a year in order to improve their product offerings. A/B Testing = A/B/N Testing = Split Testing

Experimentation

An experiment is a procedure in which one or more causal variables - for example, a marketing message about a product - are systematically manipulated and data on the effect variable - for example, choice to buy the product or not - are gathered while controlling for other variables that may influence the effect variable (e.g., Christmas season, weather, etc.)

Low Cost Metrics

Measure the ability to deliver goods and services at a low cost

problems with big data

Big data is hard to: 1.Gather Must produce and capture thousands or millions of data points 2.Store Big data takes up a whole lot of disk space (petabytes) 3.Analyze Performing computations on millions of data points takes a long time

Tabular Spreadsheet Data

Data come in all shapes and sizes. XLS and XLSX are both native files that can be opened by Excel. A lot of data are already in tabular form. Tabular form means it has rows and columns. For example, in a Comma Separated Value file, each row of data is stored in a text file with a comma separating each column's values from one another.

third party data

Data owned and generated by another organization Can also be generated anonymously how is it gathered? ●Each browser on a desktop or laptop generates unique cookie IDs ●Each mobile phone and tablet generates a unique device ID DMPs can store data down to the cookie/device ID level Personally identifiable information is not stored (multiple people can use the same machine), but individuals can be identified through hundreds of millions of cookies or device IDs

second party data

Data owned by another organization but shared with you

what does Deep Learning use

Deep Learning uses capsule networks which can recognize relationships between various parts of an object so that they can more easily and quickly recognize that object in another photo. I.e. Deep learning replicates the human brain and neurons, where the ai will analyze many patterns instead of just one, and will make predictions based on that.

Quality Metrics

Ensure that companies produce high quality products and services ●Includes certifications, training, employee involvement, etc. ○Hospitals often conduct ongoing training for surgical staff to ensure that they are constantly delivering top-quality healthcare.

data analysis

Excel, R, and Python are some of the many tools to transform complex data into insights: ●Marketing mix models ●Cluster analysis ●Moderation ●AB testing Experimental design

Number of Factors

Factors can be any number of things; these are typical: •Product •Promotion •Place •Price •Segment I recommend 2 or 3 factor designs in which: •1-2 factors are one of the 4 Ps (e.g., price, $10 versus $20) •1 factor is a segmentation variable (e.g., high versus low income)

leaders in deep learning

Google Google's self-driving cars will utilize deep learning in order to help them react quickly to unexpected things on the road DeepMind Revered as the foremost experts in A.I. deep learning, DeepMind has created self-learning A.I. capable of beating globally ranked players at chess, Go, and Shogi Facebook Through the Facebook A.I. Research initiative, (FAIR) Facebook has open sourced many of their A.I. breakthroughs, helping smaller companies begin their own work as well Baidu The largest Chinese search engine, Baidu's A.I. department has seen amazing developments when it's neural network achieved a 97% accuracy rate in voice recognition and 99.7% accuracy in facial recognition

Comprehensive Marketing Measurement and Optimization Solutions

Has features of an inbound marketing platform + Has features of a data management software + Solutions to marketer problems A software that allows marketers to collect and analyze data, create and manage campaigns both digital and non digital Focus: Optimize Media Mix Return on Investment (ROI) Predictive Marketing Analytics

Current A.I.

In the current state of machine learning, algorithms need to process millions of photos of a subject before being able to recognize that the subject of that photo is in other photos as well. I.e. Machine learning looks at one general pattern and tries to make predictions based off of that.

Purchase Funnel & Journey

Inbound Marketing tools aim to create lead generation and increase the retention of customers

Importance of SQl

Leverage Valuable Customer Information Understand Company Metrics Make Data Driven Business Decisions ●SQL is important because it is the language you use to interact with a database ●Two biggest language categories of SQL are table functions and query functions

Profiling People

Looks at behavior and sales record but not current perceptions

Marketing Fundamentals

Marketers are responsible for managing and adjusting the... ●Marketing Mix: the combination of mediums companies use to reach consumers and meet their needs successfully ●Four P's of Marketing: product, price, place, promotion

Current Campaign & In Flight Optimization

Marketing Evolution can track current campaigns and compare between medias, with sales, against the optimized plan, and external factors Benefit: In Flight Optimization allows the user to open business drivers in Media Planning and adjust spend on media rollups and levers

inbound marketing

Marketing activities that draw the attention of customers through blogs, Twitter, LinkedIn, and other online sources, rather than using more traditional activities that require having to go out to get customers' attention, such as making a sales call. ●Collect data about campaigns ●Generate ROI ●Create metrics ○Ex) amount of visitors, leads, click-through rates, average time on site, conversion rates Provide campaign recommendations

Benefits of Using SQL and NoSQL

Marketing analytics professionals can: ●Understand their company's metrics ●Compile demographics of their customers ●Run marketing campaigns based of their data ●Better contribute to making profitable data-driven decisions

Innovation Metrics

Measure a company's ability to innovate and develop a mix of new offerings ●Measure levels of breakthrough products, next generation products, major enhancements, minor enhancements and corrections

Responsiveness Metrics

Measure whether companies are attentive to customer needs ●Include measures of development speed, attentiveness to needs, and market feedback ●Many companies release new updates in the App Store to better the users experience.

Development Metrics

Measures a company's ability to leverage competitive advantage and product or service development. divide into 1. low cost 2. customization 3. quality-oriented development 4. responsiveness 5. product innovation

Frequency to Lift Response Function

Measures the impact of each incremental frequency point on any objective for the target market level or between medias ●Found through experimentation with media exposed versus controlled groups ●Lift : the difference between the exposed and controlled groups in profitable or impactful behavior of consumers ●Have a diminishing returns relationship: each additional time the customer sees the message, the less the consumer is influenced

Spend-to-Impact Response Function

Measures the impact of each incremental marketing spend on the objective. ●Used to compare the impact and ROI of different medias leading to recommendations ●The combination of factors taken into account FLRF (initial impact) + other factors Ex) In this example, the marketer should first invest in email then direct mail

capsule networks

Neural networks that recognize relationships between various parts of an object so that they can more easily and quickly recognize that object in different images

How deep learning works

Neural networks, developed in the 1950s not long after the dawn of AI research, looked promising because they attempted to simulate the way the brain worked, though in greatly simplified form. A program maps out a set of virtual neurons and then assigns random numerical values, or "weights," to connections between them. These weights determine how each simulated neuron responds—with a mathematical output between 0 and 1—to a digitized feature such as an edge or a shade of blue in an image, or a particular energy level at one frequency in a phoneme, the individual unit of sound in spoken syllables.

Commercial Data Visualization Tools

Online or desktop software packages that offer a wide range of options for data visualization for a price

Different Relational Databases

Oracle Microsoft SQL Server MySQL Amazon Redshift

bounce rate

Percentage of people who leave the webpage after viewing only one page

R and Python

Python are open-source programming languages for data analytics.

Infographics

Tells a story by combining pictures, words charts, stats, and percentages in one effective graphic Can be made using Adobe Suite or free tools like Visage, Piktochart, and Canva Three major parts: 1.Content 2.Design 3.Knowledge

Marketing Campaign Life Cycle

The Process of creating and running a campaign through several stages: 1.Design •Starts with exploratory research •Social media analytics (likes, retweets, shares, etc.) 2.Implementation •Time to do everything to ensure success of campaign •AB Tests are useful doing this stage 3.Evaluation •Compare results to original goals

SQL Query Functions

Query functions are commands that manage and change tables within a database ●SELECT ●FROM ●WHERE ●JOIN ●GROUP BY ●SELECT ○Gathers specific data from a table. ●FROM ○Establishes which table the data is gathered from. SELECT CustomerID FROM CustomerDemo ●WHERE ○Defines a specific condition desired in the outcome. SELECT CustomerID, CustomerName FROM CustomerDemo WHERE Age = 35 ●GROUP BY ○This function tells SQL how to segment the data that is selected. SELECT OrderNumber FROM OrderInfo GROUP BY State ●JOIN ○Combines two tables of data. ○The command "ON" labels which two tables to be joined. ●4 Variations ●Inner Join ○Only displays data that has matching records in both tables. ●Left Join ○Displays all data from the left table, and matching records from the right table. ●Right Join ○Displays all data from the right table, and matching records from the left table. ●Full Join ○Displays all data from both tables, together.

click-through rate (CTR)

Ratio of people who click on a link compared to the total number of people who saw the link.

Make a Radar Chart in Excel - Basketball Skills

STEP 1: Download the file, playerattribute.csv STEP 2: Select the five players STEP 3: Highlight data STEP 4: Insert=>other charts=>radar chart (looks like a spider web)

social listening

Social listening tools are platforms that connect to various social media networks in order to extract consumer data. Examples: Hootsuite Facebook Audience Insights Tweet Reach Benefits: You can get consumer data directly from the consumer themselves. This will help marketing professionals make better decisions when it comes to the 4 P's.

Why Machine Learning Matters?

Speed to support faster compute calculations Power to process and analyze large volumes of data Efficiency to generate more models Intelligence through the ability to learn autonomously and uncover latent insights.

Summary Query Commands

Summary queries are SQL commands that can answer simple statistical questions of the data ●COUNT() ●AVG() ●SUM() ●MAX() MIN()

how machine learning works

Supervised Learning Develop predictive model based on both input and output data Trains a model to generate reasonable predictions for the response to new data. Supervised learning uses classification and regression methods Unsupervised Learning Group & interpret data based only on input data Finds hidden patterns or intrinsic structures in data. Applies cluster analysis, association rules and dimension reduction ●Does not need inputs or need to be trained ●Has led to "Deep Learning"

Machine Learning

Systems that automatically learn and improve without being programmed to do so using large amounts of data and algorithms 2 Categories: ○Supervised → Human guidance to train the algorithm ○Unsupervised → Does NOT need algorithm training Important Elements: ●Data Preparation Capabilities ●Algorithms ●Automations ●Iterative Processes ●Scalability ●Ensemble Modeling

GIS (Geographic Information Systems)

Take massive amounts of data and present it in a map.

Independent variable

The experimental factor that is manipulated; the variable whose effect is being studied.

Dependent variable

The outcome factor; the variable that may change in response to manipulations of the independent variable.

new sessions

The total number of new site versus recurring visitors.

Foundational Marketing Analytics Tools

These include: ●spreadsheet tools ●programming tools

Marketing Data Platforms Summary

Three Major Marketing Data Platforms ●Inbound Marketing Data Platforms ○Digital content management ●Data Management Platforms (DMP) ○Data storage, analysis, and segmentation ●Comprehensive Marketing Measurement and Optimization Solution Leaders ○Digital and nondigital analysis maximizing ROI across a media mix

unique visitors

Total unique visitors to a webpage (monitored by IP address)

total visits

Total visits to a webpage (non-unique, including unlimited visits by the same person)

why integrate data

Two or more database sources (e.g., two data sources or tables within a single data source) may offer more insights together than separate

Targeting and Positioning

Use means analysis of outcome variables (e.g., past patronage) to know which segments to cluster. Use means analysis of positioning variables (e.g., importance and perception variables) to know how to position the product via the marketing mix.

secondary data

Were collected for some purpose other than solving the present problem (e.g., census data, syndicated data, primary data collected previously)

data management platform

When dealing with Big data, data management platforms make it easy to: collect data create audiences analytics/insights push to marketing campaigns

Make a Word Cloud

Wordclouds.com is a free online word cloud generator. STEP 1: Download the Chapter 11 word document STEP 2: Choose wizard 🡪 upload Word doc STEP 3: Change Shape, Theme, Font, etc. STEP 4: Answer the clicker question

Positioning

a marketing strategy that emphasizes serving a specific market segment by achieving a certain position in buyers' minds

which is most valuable and most difficult?

a. prescriptive analytics

which is least likely to be an objective for analytics? a. qualitative reasoning b. market effectiveness c. segmentation d. demand estimation

a. qualitative reasoning

referrals

number of people who clicked a link from another website to get to the website.

Social traffic

number of people who found the website through social media.

Advanced Attribute Modeling

acknowledges a synergy between medias and the influences they have on the consumer and conversion event

Deep Learning

algorithms identifying relations in data inspired by process of the human brain ○ Utilizes artificial neural networks, which are computer systems modeled after the human brain & nervous system ○ Allows ability to adapt to changes to maximize results

SQL (Structured Query Language)

an international standard language for processing a database Data Management for RDM's

support metrics

brand metrics customer metrics digital marketing metrics development metrics ●The ability to retrieve a brand from memory. ○When you think of soda, you think of Coke, Sprite, etc... When you think of fast-food, you think of McDonalds, Taco Bell, etc... ●Brand recall guides our decisions every day.

Content Analysis Tools

commercial R

Concentrated Marketing

commit all marketing resources to serve a single market segment

which is most false about unstructured data? a. generated by humans b. created by machines c. may have an internal structure d. fits a relational data model

d. fits a relational data model

which is least likely to be an example of unstructured data? a. chat b. tweets c. email d. spreadsheet

d. spreadsheet

Data Visualization

describes technologies that allow users to see or visualize data to transform information into a business perspective ●When data is presented in interactive maps and charts, people will understand the information quickly and easily ○Tableau and GIS

Attribute Modeling

finds the exposure point at which the customer performs the conversion event, whether buying or just reaching the landing page. Often notes last-click-attribution

Types of Data

first party data second party data third party data

demographic

gender, age, income, race/ethnicity

Behavioral

how often do you buy gum? Where do you usually buy it? Is there a day of the week or time of day you are more likely to buy it?

analyze the metrics

identity metrics analyze the metrics

cleaning data includes dealing with data that are

incomplete, outliers, or inconsistent

IBM top priorities for marketing

increase digital acumen

artificial intelligence (AI)

is an area of computer science that "emphasizes the creation of intelligent machines that work and react like humans"

spreadsheet tool

is an interactive software application for structuring, transforming, analyzing and storing data in rows and columns. Examples: Microsoft Excel and Google Sheets

Internal validity

is the extent to which the observed results are due to the experimental manipulation

data management

is the practice of organizing and maintaining data processes to meet ongoing information needs.

Attribute Modeling problems

last click attribution should not get all the credit

implementing STP

market segmentation target marketing product positioning

Total Conversions

number of people who have taken a desired marketing outcome, as defined by the marketing team. Examples include completing a checkout on an ecommerce site; filling out a lead form; subscribing for a service; or signing up for a newsle

organic traffic

number of people who reached the website by perform a search from sites like Google or Bing.

direct traffic

number of people who typed in the URL to get to the website.

customer retention rate (CRR)

percentage of customers who return to buy again.

Power Map

presents data through 3-D GIS

types of data

primary and secondary

psychographic

psychological characteristics, values, and lifestyles

Differentiated Marketing

target each of the potential segments it discovers with a different marketing mix

open rate

the number of people who open an email that has been sent to them dividing by the total number of people to whom the email was sent.

total reach

the number of people who received impressions. Or the total number of followers, retweets or friends.

Mobile Traffic

the number of people who see the marketing content through a mobile device.

impressions

the number of time marketing content has been displayed, including the same person multiple times and regardless whether there is any response to the marketing content.

conversion funnel rates

the portion of customers who make it through to the next level of each step in the journal from impression to conversion (e.g., from impression to click-through; then from click-through to placing an item in the shopping cart; then from the shopping cart to checking out).

data cleaning

the process of checking data for errors after the data have been entered in a computer file Tools like Google refine will help clean messy data

marketing attribution

the science of assigning credit to unique events that lead to key marketing conversions like the sale of a product to a customer.

Analytics

the use of mathematics and statistics to analyze data •Tools helpful for analytics: ●Excel is a spreadsheet tool ●Excel, R, and Python are good marketing analytics tools ●SQL or structured query language, a database language

Advanced Marketing Mix Modeling

uses some historical data but focuses also on current data and other medias for predictive analytics

A/B Tests

•A/B Tests are field experiments in the digital marketing context making use of big data •A/B Tests are often run as part of a LIVE, REAL marketing effort. •A/B are often automated to be adaptive. ●A/B Tests are rooted in experimental design ●However, often: ○A/B Tests often test two or more competing messages without necessarily seeking to understand why. ○A/B Tests are often about ENDS rather than MEANS ○Experiments test hypotheses, seek to understand outcomes and reasons I recommend approaching A/B Tests more like traditional experiments, to understand WHY. •In order to understand why, be sure that A vs. B is EXACTLY the same content EXCEPT for the "why" you wish to test.

Designing a Test

•Anyone at the company should contribute •Ideas can come from user testing (your regressions, for example) •Use best practices •Be creative, try crazy ideas •Use common sense: don't piss off your users for the sake of metrics

primary data

•Are collected specially to address a specific research objective (e.g. survey data, customer feedback, experiments)

Why structure data?

•Because it comes to us in ways that would be hard to analyze. •Think about a database of newspaper articles from the last 50 years. •How would you "analyze" those millions of "unstructured" articles?

Big Data

•Data sets of such size, complexity and volatility that their business value cannot be fully realised with existing data capture, storage, processing, analysis and management capabilities

Experiments Procedure

•Decide on your treatments ○ What are you trying to test? ○ How many factors? ○ What will you do if you have the information? •Design the experiment design ●Potential experimental designs •After only •Multi-factor •Before and after •Between subject vs. within Subjects ●Create an outline on paper ●Implement the design in Qualtrics •Recruit participants and run the experiment •Analyze the experiment

Netflix

•Lifetime Value calculations drive pricing and marketing decisions •Average subscribers •$150 to acquire and maintain each customer •Year 1: 12 months * 10.99/month = $131.88 •Solidifies budget for new movies, shows, and marketing expenses •Netflix has truly mastered Customer Lifetime Value "down to the penny" •Track each individual subscriber's tastes and preferences •When a customer is about to leave, Netflix: •Knows they watch less •Uses profile information to maximize Retention Rates by finding ideal shows and movies

A/B Testing Tools

•Mailchimp (email campaigns) •Abtasty (social media) •Google Optimize (Adwords) •Optimizely (website, mobile) •Splitly.com (Amazon)

Benefits of Machine Learning for Marketing

•Massive data input from unlimited sources •Rapid processing, analysis, and predictions •Action Systems •Learning from past behaviors

Use of Data for Decision Making

•People are poor at learning from experience •People have poor intuitive senses for the dollar value of information from marketing analytics •People are limited and biased information processors ●Overconfidence: "What I've experienced is key." ●Availability bias: "I know someone who..."

marketing analytics strategy

•Planning a data-driven project to maximize marketing effectiveness •Allows marketers to maximize efficiency and minimize marketing costs •Process to accurately report on the past, analyze the present, and predict the future ●Success of campaigns need to be measured ○How much money to spend on them ○How to improve outcomes ●Marketing analytics is a continuous function of measuring results of campaigns ● Campaign Strategy Steps:

Recruitment

•Population •Incentives •Timing (all conditions at once) •Potential pools •What is a CONVENIENCE SAMPLE? •For the real world a convenience sample wouldn't be good enough. Why not?

Randomization

•Randomization is typically the way that other variables are controlled for (e.g., flip a coin to decide what a subject receives) •Randomization of subjects over treatments is the strategy for eliminating biases in measuring treatment effects due to selection between the experimental units.

Marketing Channels for A/B Testing

•Website promotion (copy, images, video) •Email campaign (messaging, subject line) •Social media ads (messaging, images, video) •Digital retailing (messaging, pricing) •Text coupons (messaging, price promotion) •Mobile ad (messaging, images)

External validity

•is the degree to which the experimental results are likely to hold beyond the experimental setting •Usually there is a tradeoff between the two •Without internal externality, external validity means nothing ==> Fix the level of internal validity and strive for the highest possible external validity.

Moderator variable

•it changes the way the independent variable affects the dependent variable

Customer Satisfaction

•renewal rates for services •likeliness to recommend to friends •effectiveness of online customer service •customer ratings on sites such as Yelp or Amazon

Marketing Analytics

•the use of data to maximize marketing outcomes. •Marketing Analytics specifies the information needed for decision making and links the information to actionable decisions.

Design

▹Often starts with exploratory research ￭Coke→ custom built listening software ⬝Find out what is going on; likes, shares, comments ▹Form hypotheses about which strategies work well and which do not

●Multi-Platform Social Listening Tools

○ Allows connection to multiple platforms (example: Hootsuite)

categories of AI

○ Artificial Applied Intelligence more common ex: Pandora ●Run aPerform single task extremely well ●utomated and repetitive tasks ●Does NOT involve decision making ●Ex) Netflix, Spotify, self-driving cars ○ Artificial General Intelligence less common ex: robot ●System that can handle infinite number of tasks ●Involved in decision making + analysis ●Think and function similar to human brain ●Still in development of artificial neural network ●Ex) Virtual assistants like Alexa

COUNT

○Reveals the number of rows (respondents/products/etc.) that fit the stated criteria. ●Example: In a relational database of a company's customers, one could find the the number of female customers aged 25-34 who purchased an item last week.

Key-Value Stores

○The simplest of NoSQL databases, uses an associative array (aka hash table) as the fundamental data model where each key is associated with one and only one value in a collection ○Benefits - Scalability, Reliability, Simplicity, Speed

Albert

○Used for autonomous media buying ■Buying digital media on behalf of clients ○Analyzes, manages, and optimizes paid advertising campaigns

UNU's SWARM AI

○Utilizes interfaces & algorithms to create human swarms (singular intelligence) ○Interface → hexagon with 6 answer choices ○Virtual magnet pulls group in certain direction

Programming language:

○a formal set of instructions that can be used to produce various kinds of data output.

Relational Databases

●"A means of storing information is such a way that information can be retrieved from it" (Oracle) ●A set of tables containing data fitted into predefined categories ○Tables might include employee, customer, vendor, and product and transactional information ●Used when working with structured data ●All systems run using the same code (structured query language, or SQL)

components of marketing analytics

●3 Components: ○Data ○Analytics ○Visualization

Where Versus Having

●A WHERE clause is used is filter records from a result. The filter occurs before any groupings are made. ●A HAVING clause is used to filter values from a group.

Data Scraping

●A computer programmed extraction of information from individual computer screens, websites, or reports. ●The legal and legitimate uses of data scraping focus primarily on scraping from public websites.

radar chart

●A graphical method of displaying multivariate data in the form of a two-dimensional chart (typically a pentagon or a hexagon; depending on the number of variables) ●Three or more quantitative variables represented on axes starting from the same center point. ●Radar charts are a good alternative for 3D bar charts

Advantages of Python

●A growing community that includes computer science software engineers and programmers ●There are more opportunities to take advantage of artificial intelligence ●Flexibility; e.g., data analysis can be integrated with website and mobile apps or a production database ●Ready for programming tasks besides analyzing data

marketing evolution company

●A leader in comprehensive marketing measurement and optimization solutions ●Attempts to optimize ROMO (return on marketing objective) ●Creates custom in-depth optimizations for current and future media plans with in-dashboard adjustments ●Provides visualizations, metrics, and best practices engine ●Utilizes an immense amount of data to get a complete picture Marketing Evolution's platform for optimizing media mix performance ●Collects and analyzes data ○First, second, and third party data ○Marketing Evolution's own data ○Marketing Evolution collects new primary data ■Experimental: Measuring "lift" ■Survey Responses: Measuring perceptions ●Comprehensive data leads to recommendations Applies information to ●Create metrics, segment, identify leading indicators, and business drivers ●Measure ROI ●Manage overall business and current campaigns ●Create Media Planning ●Recommend best practices

Google Ads

●A more commercial product compared to Google Trends designed for company advertisement ●Gives marketing analysts tools, like Keyword Planner, to see how often keywords are searched and changes over time ●Gives indications of volume instead of an index

GDPR (The General Data Protection Regulation)

●A new standardized data collection law effective on May 25, 2018 affecting 28 countries in the EU ●Scandal of Cambridge Analytica: company used personal data to target messages to users during the 2016 presidential campaign using Facebook data ●Gives consumers the power to say no to data tracking, relevant to DMP's data. ○focuses on ensuring users know, understand, and abide to their data being collected online

Propensity Model

●A statistical algorithm used to predict customer behavior ●Customer Behavior created by feeding large quantities of data to machine learning algorithm ●Propensity model consumer metrics: ○Likelihood of conversion ○Price Point of conversion ○Which consumers to target to become repeat customers ○Predict lead effectivity ●Only as good as the data you put into them

Web Crawlers

●A web crawler uses web scraping when firms index the web periodically to make searching quicker. ●After extraction, data is stored to calculate metrics. *It is also possible to gain even greater control over web scraping through programming in languages like R.

Data Flow

●AI detects when unexpected traffic is visiting a site ●Analytics is checked autonomously by AI platforms

Marketing Examples & Applications

●According to a recent survey, one third of advertisers are using A.I. to deliver personalized web experiences ●Applications include image recognition, video analytics, language processing and much more ●Personalization is based on factors such as geographic location, demographics, types of device being used, & on-site interaction ●This info is used to display offers and content that fits each user ●One of the main applications of this can be seen from music streaming services such as Spotify when it curates recommended playlists and suggests songs based on previous listens

Major DMP technology sellers:

●Adobe Audience Manager ○Adobe acquired DMP Demdex in 2011 ●Oracle DMP ○Oracle acquired BlueKai in 2014

data storage and management

●After data is collected, it must be stored and managed ●Data management platforms: ○Adobe Audience Manager ○Oracle BlueKai

Intelligent Personalization

●Algorithms that can personalize websites to individual site users ○Based on: ■Geographical Location ■Demographics ■Types of Device being Used ■On-Site Interaction ○Companies like WSJ, Pandora, and TopFan use AI to improve conversion rates

Adobe Audience Manager

●Allows companies to build unique audience profiles ●Uses cookies to track users data ●Identify the most valuable market segments, cluster, and then target them across any digital channel ●Can decide which version of a website to show a specific customer ●Provides access to all this information in one interface

CONTENT ANALYSIS

●Allows for sentiment and text analysis of gathered data ●Serves segmentation and targeting activities, as well as opinions about brands

Uses for Relational Databases

●Allows users to link information from different tables using keys and indexes through the use of queries ●Query- a question that allows the user to pull information from the database to answer said question ●Help marketing analytics professionals to target specific customers ○Store information about products, customers, employees, etc. ○Use segmentation to create groups of potential customers

Hadoop

●An open-source software framework that stores and processes large amounts of data ●Benefits that are attractive to business: ○Computing Power and Scalability ○Fault Tolerance ○Low cost ○Flexibility

SWARM AI

●Analysis allows marketers to make decisions that are lower in risk and do so with less data ●Potential to reduce marketing costs → especially for newer companies ●More accurately predict future movements in the market ●Could replace expensive customer research and more efficiently create personas and characterizations

Online Content Count Tools

●Calculates frequencies of word or phrases embedded within a media provider's website ●Ex: New York Time has a tool to search for any word or phrase within articles published by the NY Times from 1851-Present ●Counts can be adjusted by date range, article type, and section

Oracle BlueKai

●Cloud Based ●Takes in 1st and 3rd party data from different devices ●Can target different experiences on different devices ○Uses cookies to see when users access what data on what device ●Maps all consumer IDs with device IDs to form consumer profiles ○Utilizes and ID graph to bring multiple aspects together into a user profile ○Allows increased personalization of campaigns

Hadoop Ecosystem- Distributors

●Cloudera ○Serves an enterprise data hub, which allows for real-time data processing, stores and processes data in the same machines; provide security ●Hortonwork ○ 100% open-source; allows multiple workloads of data to be processed at the same time ●Map R ○Uses its own filing system in place of HDFS; analyzes large data set, both structured and unstructured data

Cluster Analysis

●Cluster analysis seeks to group objects such that segments are created that are as homogenous as possible given the choices by the researcher ●Cluster analysis works on the principle of maximizing the between-cluster variance while minimizing the within cluster variance ●Every object is allocated to one cluster

Step 2: Analyze the Metrics

●Companies must implement systems to track the important metrics ○Web Analytics ○Marketing Automation Dashboards ●Compare current state to benchmarks ○Historical Trends ○Industry Average Performance ●Most important step is to determine the root cause of why metrics perform the way they do

Examples of Machine Learning

●Computational finance, for credit scoring and algorithmic trading ●Image processing and computer vision, for face recognition, motion detection, and object detection ●Production, for price forecasting ●Automotive, aerospace, and manufacturing, for predictive maintenance ●Natural language processing, for voice interaction

The Development of Deep Learning

●Conceptualized in the 1950's but hasn't been a possibility until as early as 2010 ●Requires massive amounts of processing power that haven't been available until recently ●R&D has been extremely expensive until companies like Facebook began open sourcing their work, allowing smaller developers to build on their research

advantages of secondary data

●Cost (internet has made search cost low) ●Time ●At times more accurate, at times the only alternative

DMP

●Creates cookie IDs to make targeting segments for digital advertising campaigns. ●Collects demographic, psychographic, and behavioral data about the target audience ●Stores, analyzes, and segments audience data ●Unifies audience and performance data across all sources ●Uses captured data to create hyper-targeted ads ●Continually optimizes campaigns

insider program

●Customer loyalty programs are often used to increase customer profits. ○Point-based program - customers get points for each dollar spent and can redeem them for beauty supply products at checkout

Power BI

●Dashboards that allow users to access, edit, and automatically update them ●Connects to variety of external data sources ●Automatically determines relationship between data from multiple sources ●Good at importing visuals with easy to use interface (can connect to R and Python) ●Simplifies data evaluation, sharing dashboards, and interactive reports ●Power BI has the advantages of Microsoft business analytics, that includes Azure Machine learning, SQL Server based Analysis Services, and data streaming ○Stronger at data manipulation ○Easily develops predictive modelling and optimization ○Visuals are selected first then data is added ○Lower cost ○Leans toward small business. ○Places a datapoint limit.

data cleaning

●Data cleaning tasks ○Fill in missing values ○Identify outliers and smooth out noisy data ○Correct inconsistent data

Why Data Cleaning?

●Data in the real world are dirty ○incomplete: lacking attribute values, lacking certain attributes of interest, or containing only aggregate data ○noisy: containing errors or outliers ○inconsistent: containing discrepancies in codes or names ●No quality data, no quality results! ○Quality decisions must be based on quality data ○Data warehouse needs consistent integration of quality data

Search Volume Data Tools

●Search volume tools allow the user to see the popularity of specific topics ●Google dominates this space with Trends and Ads ○Trends is used to track the ups and downs of search engine results ○Ads also shows search popularity, but is designed to allow marketing professionals opportunities for advertising.

Non-Relational Databases

●Databases that are not stored in the format of linked tables, and generally do not use SQL for data manipulation ●For semi-structured or non-structured data ●Also referred to as "NoSQL" Databases ●Uses Include: ○Large-scale data processing ○Embedded IR ○Exploratory and predictive analytics ○Large volume data storage ●Google, Amazon, Facebook facing problems: ○Constantly changing data ○Decrease in development cycles ○Massive # of users ●Not stored in tables, but documents ●Grouped into 4 categories: key-value stores, document databases, wide column stores, and graph databases

3 Types of Segmentation

●Demographic ●Psychographic ●Behavioral

Open Source Tools

●Designed to deliver easy-to-make charts, plots and maps ●Trulia uses R to form map overlays for crime, traffic, amenities, etc

A1 marketing downsides

●Difficult to use depending on company or industry being marketed ●Too predictable and expected = Too boring ●Stifling of the creative process in marketing

Dynamic Landscape

●Digital media gives marketers the ability to reach thousands, even millions ●Consumers are in information overload making it harder to stand out ●Marketers need to successfully use data in order to create and manage online campaigns and offline campaigns

features of unstructured data

●Do not reside in traditional databases and data warehouses ●May have an internal structure, but does not fit a relational data model ●Generated by both humans and machines ○Textual and multimedia content ○Machine-to-machine communication

Conveyance of Concepts

●Easier to absorb graphs than a spreadsheet ●Eliminates barriers ●Draws attention, clarifies, and predicts ●Helps to identify relationships and patterns ●Need to engage audience

Octoparse

●Easily-configured visual scraping tool ●Can run extractions on the cloud and on your own local machine ●Exports the scraped data in TXT, CSV, HTML or Excel formats

Implementation

●Ensure the success of the campaign ○Coke→ using their data, generated over 300 customizable responses ●Carefully and regularly monitor results and making adjustments when necessary

Evaluation

●Evaluate whether a campaign has met its goals or not ○Coke→ social reach, over 22-million, increase happiness exposure by 200% ●Determine how many resources should be spent in the future, if any

excel

●Excel is one of the most widely used ●Charts are simple and tend to not be aesthetically pleasing ●Creating Dashboards is time intensive ●Lags significantly behind most data visualization tools ●Add-ins can provide enhanced functionality ●Power View allows user to create interactive charts ●Power Map allows users to map data

Hadoop advantages

●Fault Tolerance ●Low Cost ●Flexibility ●Computing Power and Scalability

Benefits of a Non-Relational Database

●Flexibility- many different types of database systems allow marketing analysts to manage data how they wish ●Relatively inexpensive, allowing companies to store high volume data like logs, call-data records, meter readings and ticker snapshots ●Affordability- they utilize open source software or use cloud storage ●Accessibility- marketing analysts can analyze semi or non-structured data such as email archives, XML files, and documents ●Scope- they can crunch large data sets and help machine-to-machine data retrieval

SOCIAL LISTENING

●Gathers content from forums, comment sections, and social media ●Converts from HTML to text readable by humans, but not interpreted - i.e., raw data

Visualization

●Graphs and other visuals help communicate findings from analytics and data ●Visualization facilitates good storytelling ●Visualization Tools: Tableau, Excel, Power BI and geographic information systems

Hadoop Components

●Hadoop Distributed File System ●Common ●YARN ●MapReduce

Tableau

●Has visuals head and shoulders above competition ●Robust drill-down and visualization tools ●Visuals can be transformed into dashboards ●Dashboards can be shared on Tableau Online and accessed by browser or mobile app ●Easily connects to any data ●Can integrate data from multiple sources. ●Integrates with most data types and platforms, including Hadoop, Python and R. ○Stronger at data visualization ○Easily develops sophisticated visualizations. ○Data is selected first then switch between visualizations later. ○More costly ○Leans toward large business ○Any number of datapoints. to download: STEP 1: Open Tableau STEP 2: Upload the Diamonds excel file into Tableau STEP 3: Slideshow describing the data STEP 4: Make a visualization with an interesting insight STEP 5: Email 1 image to [email protected] See the accompanying "Tableau Exercise.pptx" and diamonds.csv, both on PolyLearn

Step 1: Identity Metrics

●Have a goal in mind ●Come up with a question you need the data to answer ●Metrics help bring meaning to the data ○Metrics are quantifiable measures used to track the status of a marketing process ●Metrics help determine if goals are being achieved

Cluster Analysis: Basic Idea

●In order to group objects, some kind of similarity measure is needed. ●Similar objects are grouped together and those further apart are put in separate clusters. ●Euclidean distance

Defining The Fixed Unit Effect

●In the presence of others, people consume fixed number of food units based on what they perceive to be an appropriate quantity of units of a given food. ●The result ○Unit size is held constant ○Total caloric intake varies with unit size (i.e., smaller or larger units), and consumers will consume more calories with larger unit sizes.

Search-Volume Data Tools

●Indexes the search volume of terms people use on search engines ●Data can be downloaded into raw files ●Public Access

Google Trends

●Simple tool that works just like a Google search ●Indexes search interest on a 0-100 scale ○Can be misleading ●Recently expanded service to include data from Google News, Google Images, Google Shopping, and YouTube

Customer Lifetime Value

●Informs companies about how much a customer is worth to them. ○Especially important for companies like Netflix, where they want customers to continue to subscribe to its services. ●These metrics focus on the LONG TERM value a single customer brings to the company. customer lifetime value= (margin) * (retention rate)/ (discount rate)-(retention rate) Margin = customer revenue - cost it takes to service the customer Retention rate = percentage of customers who remain loyal over time Discount rate = cost of capital for the organization Retention rate = percentage of customers who remain loyal over time

Hubspot

●Intuitive all-in-one platform ●The Leader in Inbound Marketing Data Platforms ●Free add-ons: Sales and CRM ●Integrations with Slack, Salesforce, LiveChat, and MS Dynamics ●Popular wit small to midsize companies

K-means Clustering

●K-means clustering is the most commonly used clustering technique ●It is an iterative technique that seeks to allocate each observation to the cluster that is located closest to it ●The number of clusters is chosen by you

Non-Relational Database Types

●Key-Value Stores ●Document Databases ●Wide-Column Stores ●Graph Databases ●Multi-Model Databases

Per Visualization

●Know the audience ●Assess the data ○High cardinality: data variable has mostly unique values(bank account numbers) ○Low cardinality: repeated values (gender) ●Decide between visualization options

Disadvantages of Python

●Less efficient for statistical computations (it was original built for non-statistical purposes) ●Has less appealing data visualization built in ●Fewer packages

Customers as a whole

●Look at behaviors and perceptions based on survey data

advantages of R

●Made for data-oriented projects in general ●Handles big data (very large datasets) ●Large number of ready-made packages ●Built-in ways to professionally visualize data ●Developed by data scientists, important for marketing analytics ●Large community that provides support through mailing lists, documentation and blogs ●Supported by a well-established programming tool (a.k.a. integrated development environment) called RStudio for which there are no close competitors in R and for which Python has no comparative leader

Customization Metrics

●Measure the ability to tailor products and services to customers ○Can include modular design, configuration systems, flexible manufacturing, and JIT inventor ○Dell allows customers to configure and customize their PC to their liking, using online configuration tools.

Step 3: Take Improvement Actions

●Most difficult step in the process ●Changes aren't always obvious so marketers use analytical and creative skills to develop solutions ○A/B testing allows marketers to make isolated changes until the best performing marketing effort can be achieved ●Invest their resources in areas that need the most improvement

Marketing Mix Modeling problems

●Not fast or detailed enough, foresight analysis (need years of records), only looks at existing media

Data Collection

●Obtain the right data to achieve the right information, detailed information about user interactions within their own website, social media, and other digital platforms ○Time spent on website ○Clicks ○Conversion rates ●Internal sources like revenue and cost information

Profiling

●Once you determine the market segments, to make things actionable you want to compute who is in what segment (profiling) ●Here is where you use demographics, psychographics and behaviors for each segment. ●Useful for reach and targeting purposes. ●Name the segment (ex: OFES = Older Frequent-restaurant Eater Spender). ●This is how the term Yuppies came to be. ●Another from segmentation: SINBAD (single income, no boyfriend, absolutely desperate)

A1 Marketing: Data Collection

●Organisms amplifying group intelligence by forming flocks, schools, shoals, colonies, and swarms (eg. bees, birds, fish, ants) ●Research shows → groups outperform most individuals in decision making & prediction regardless of expertise ●Most surveys, focus groups, and polls fail to recognize this social aspect of decision making

Conclusion to Marketing Evolution Case

●Predictive Analysis, looking forward ●Widened focus attributing to other components other platforms fail to account for ○Measures digital and nondigital media ○Increased accuracy ○People-centric ●All-in-one content management, data management, and planning functionalities ○Fast capabilities Easy visualizations

disadvantages of secondary data

●Problems of fit ○Inappropriate level of aggregation (time, company, region) ○Wrong unit of analysis (e.g. you are interested in the price ○elasticity for the small box but the data are about the large one) ●Problems of accuracy ○Uncertainty about supplier/collection methods ○Availability of multiple sources

New Way: Inbound Marketing

●Pull consumers in during their purchasing journey ●Align with consumer interests ●Create relevant, differentiated, and quality content

Old Way: Outbound Marketing

●Push products onto consumers ●One-size-fits-all ●Assume those who need will buy

The Use of Color

●Red-Blue color blindness affects: - 8% Men and .5% Women ●Blue and Orange schemes are color blind friendly Strong contrasts in color can also help your audience differe

Causal Inference

●Requirements for Causal Statement X → Y ○(1) X must occur before Y ○(2) There must be evidence of association between X and Y ○(3) Control of other causal factors ○(1) is typically manipulated. (2) can be assessed by the data. (3) is typically done by randomizing, matching, or blocking ●Example: ○product advertisement on TV → Choice of product purchased

Best Practices for Stable K-means Cluster

●Run multiple times to see if you get the same answer ●Split the data RANDOMLY in half and try each half separately.

SQL vs. NoSQL

●SQL manages relational databases while NoSQL manages non-relational databases. ●NoSQL can handle large volumes of rapidly changing structured, semi-structured, and unstructured data.

Marketo

●Skilled in in-depth, custom solutions ●More complex and requires a professional ●Popular with mid-sized to larger companies ●Leading Salesforce integration, Netsuite, MS Dynamics, SAP, and Oracle. ●Five different components for managing digital marketing: lead management, email marketing, consumer marketing, consumer base marketing, and mobile marketing

Spinn3r

●Spinn3r scrapes entire data from blogs, news sites, social media and RSS feeds. ●Firehose API manages 95% of crawling and indexing work. ●Scraped data can be filtered using keywords.

Disadvantages of R

●Steeper learning curve than many languages ●Less efficient for general computations, sometimes due to inefficiently written packages

Benefits of Relational Databases

●Store large amounts of data and create easy access to information ●Information can be manipulated in real time ●Can be used to target and segment specific customers ●Example: Targeted emails

R

●Text analysis through tokenization and sentiment analysis through dictionaries ●TM Quanteda

Customer Profit

●The profit a company makes off a customer or customer group over a period of time ●Seeks to acknowledge customer value

clusters

●There are no formal significance tests for the number of clusters, but there are informal tests that involve the amount of heterogeneity within each cluster. ●People typically decide based on managerial action: (a) How many clusters are feasible to target? (b) Are each of the clusters of sufficient size to be considered a market segment? (c) Is there sufficient differentiation between the segments?

marketing analytics metrics

●They are measures of a data-driven project to maximize marketing effectiveness. ●Understand marketing analytics metrics allows marketers to maximize their efficiency and minimize marketing costs. ●Marketing analytics metrics are ways to assess a marketing analytics strategy.

Other Search Volume Tools

●Tons of free, simple keyword research tools ○Searchvolume.io ○Serps.com ○Wordstream.com ●Bing and Yahoo offer services similar to Google Trends ●Moz is a popular commercial service that analyzes search volume and reports metrics and suggested actions

Customer Acquisition Cost (CAC)

●Total marketing costs over a period of time/ Total amount of new customers in that same time period.

Organizational Level

●Understand data, its size, and cardinality ○Cardinality: the uniqueness of data values in a column ●What do you want to visualize? What kind of information do you want to communicate

Artificial Design Intelligence

●Use of AI for ○Design of websites ○Written content and graphics ○Writing Reports ○Creating website dashboards ○Email communications ●If a company has numerically based content and data-driven info, the company could benefit

Dashboard

●Used to display multiple visuals on one page ●Can utilize interactive visualization allowing viewers to focus on one segment

Gantt Chart

●Used to represent project planning data ●Displays chronological tasks ●Can be updated to demonstrate progress on various activities

commercial

●User pays for access ●Linguistic Inquiry and Word Count (LIWC) ●Translates words and phrases into psychological states

Marketing Mix Modeling

●Uses historic company information, statistics, and regression to look for relationships, or correlations, between marketing activities and sales. ○Recommends where to invest for a future campaigns and strategies ○Measures the effectiveness of each media ○Focuses on ROI

Content Analysis

●Using acquired data in meaningful ways ●Studying digital text, photos, audio or visual formats of communication to further understand customers ●Sentiment vs. text analysis

Challenges of Big Data

●Validity of statistical inference ○Sample biases ○Model biases ●Privacy and public trust ○Disclosure threat due to mosaic effect ●Data integrity ○Missing, inconsistent and inaccurate data ○Volatile sources ●Data ownership and access ○Public good versus commercial advantage ○Value of private sector data

Commercial Data Tools

●Vast amounts of data (customer details, product information, trends, and more) ●Typically structured Costs $$$

Dex.io

●Web-based scraping application that doesn't require any download ●Browser-based tool that sets up crawlers to fetch data in real-time ●Has features that save the scraped data directly to Box.net and Google drive or export it as JSON or CSV files ●Supports scraping data anonymously using proxy servers

Document Frequency Matrix

●With tokenization complete, it is possible to construct a new dataframe (i.e., a matrix) where: ○Each row represents a document. ○Each column represents a distinct token. ○Each cell is a count of the token for a document ●This representation is extremely useful!

Hypothetical DFM

●Word ordering is not preserved! ●This is known as the "bag-of-words" model. ●The BOW model is a very common representation in text analytics

R packages

●a set of related functions written in R programming code and publicly available ●Steps: ○Install the package by searching and clicking the installation link ○Add it to the library using library()

within-subjects

●design is an experiment in which the same group of subjects serves in more than one treatment. Note that I'm using the word "treatment" to refer to levels of the independent variable, rather than "group".

between-subject

●design is an experiment that has two or more groups of subjects each being tested by a different testing factor simultaneously. ●Within subjects designs are advantageous because you get greater statistical power due to "internal matching" (you are your own control) ●However, in some cases, due to contamination, time constraints, and infeasibility, then "between subjects" designs must be used

Programming

●is the process of solving a given problem using executable computer algorithms, well-defined procedures for solving problems. ●Statistical programming is the process of solving data-related problems using executable computer algorithms.

Power View

●presents data through charts, graphs, maps, and other visuals like GIS.

site time

●the amount of time a customer spends on a site (e.g., important for media providers).

micro conversion rate

●the conversion rate at the campaign or platform level (e.g., the conversion rate for a particular marketing video delivered to newsite apps on a mobile phone).

conversion rate

●the total conversions divided by the total reach.

lead to close ratio

●total number of sales leads by the total number of sales (important for sales analytics)