MKT494: Data Driven Marketing Final
Facts First, Gender Example
Gender is not a segment variable unless you would fundamentally change how you interact with a customer based on that fact. Why? Because a female persona is often the key influencer (less price sensitive; different needs, wants, pain points), a male persona often the key buyer (price sensitive). ex: Traditional jeweler would be able to segment gender, but Ahipoke bowl wouldnt
Leaders in Data distribution
Google, SAS and Adobe are the current leaders in the space according to Gartner (2018). This means as of now tracking a journey across channels, both on- and off-line, is both expensive and difficult.
Competitive Grid (3)
How do people feel and what do people think about the content a brand's competitors are crafting? Where are the holes or gaps in knowledge delivery that people want?
Usability
How easy is it to use our site? How do we know? --> Testing
Building a Hypothesis
IF access to deeper pages is more prevalent, THEN lead volume will increase DUE TO higher engaged users accessing more bottom funnel content and information on the site.
Conversion Rate Optimization
IF an action is done (independent variable) THEN a result will happen (dependent variable) DUE TO. an educated correlation (the relationship between the independent and dependent variables)
Paid Media
Intent informs targeting based on interest, behaviors and platform priorities.
Where does performance marketing fit?
Interest & Action Stages. Customer funnel: Awareness--> Interest--> Action--> Loyalty--> Advocacy
Action & Reaction Chart
Intro to a e-Commerce slide show Page 9/14, helpful to understand metrics in e-commerce
Native Platform Analytics (4)
Look at data and affinities within Facebook and Google to better understand the interests of a brand' customers.. -influences ad targeting, content topics, and potential partnerships
Decision-making through Correlations
Looks at top of the funnel activities to see if they impact mid-funnel call-to-action (buying) activities. -Many correlative decisions in marketing are made through regression, looking for correlation within relationships: -Does social media engagement lead to site visits? -Do email open rates indicate conversion rates? -Do Facebook "likes/fans" buy more in-store? From these correlations, decisions on how to market in the future become more clear and can be more easily defended. https://www.youtube.com/watch?v=-_o2obbjF3I
Promotion (e-commerce)
Loyalty Programs: The best loyalty program is a great experience -Hard to execute: Bad data is prevalent -Hard to understand: Customers don't easily understand Not differentiated: Everyone has them. -Discourage loyalty: They only work if people are loyal. If people are loyal, why do you need to do it. Instead you simply get people looking for a deal (your worst customers) Promotions (PMDs): competing on price rarely works
Product Development and Support
MVP, Development Roadmap, support Think --> Make --> Check -->(Repeat) Think: generate research, ideation, mental models, behavioral models, test results, competitive analysis Make: prototypes, dialogue design, value prop, hypothesis, delayed code Check: A/B testing, analytics, usability testing, funnel, sign-ups
Media Buying Terms
Media Mix: combo of channels you use CPM: cost per thousand CPA: cost per action CPC: cost per click Guaranteed inventory: planned ad space at a negotiated CPM (usually) RTB: real time bidding, non-guanrenteed inventory where advertisers bid against one another for the space
Making the Most of Content
Move from creating content for platforms to making content that works together, satisfies intent, and creates a holistic approach.
Old World vs New World (Social Media Approaches)
Old World Social Media Approach: -internal team *guessing* on content topics like miscellaneous holidays -Brand team or agency brainstorms with *only stakeholder data points* -Target a certain *quantity* of posts -Focus is entirely on category New World Data-Driven Social Media Approach: -Multiple points of data and analytics *informing* approach -Hearing directly from customers on their preferences and expectations -*Quality* of posts that are boost-worthy -Affinities uncovered across complimentary categories
Writing Rationale
Once you uncover a sound choice, use the efficiency of that choice in reaching your audience (and potentially in driving better conversion rates) as the "reason why" it is chosen. example: Ars Technica reaches the scale (number) of technophiles in major metropolitan areas we need and does so at the lowest CPM among targeted sites.
Predictive Analytics (example)
One easy example of a predictive analytic is your credit score, which is an algorithm designed to let a company know how likely it is you will be able to pay them back if they lend you money for a purchase.
One Metric That Matters Approach
One idea might be to start with the one metric a business needs to optimize its performance? -Great because everyone follows one path. Not-so-great if you have the wrong metric. -Plus it is hard to make something this "big" also be actionable.
Danger with Social Listening Tools
Over-reliance on Twitter for Social Listening Tools - only 24% U.S. penetration
Intent Informs Beyond Social
Paid Media --> influencer Strategy --> Customer Service --> SEO
Performance Marketing analogy
Performance marketing looks for low-hanging fruit, instances where a consumer is actively in the market, or seeking out what we have. -Our goal is to convince the consumer quickly that we offer the best option to fit their need and to get them to act quickly.
2) Find people with immediate intent
Performance marketing relies on search marketing as a primary tool because people signal their intent when they search a phrase -There are specific intent signals we can look for in their search -SKAGS, aka single keyword ad groups, are a trick performance marketers use. -w/ a SKAG strategy, a marketer only places one KW per ad group so there is a 1:1 relationship between the word searched and the landing pace to which they will be directed -SKAGS on avg improve click-thru-rates by about 25%
1) Segmentation
Performance marketing relies on very small segments, as close to personalization as you can get in a media setting.
Product Journey
Product Opportunity Audit --> Product Strategy --> Market + Concept Validation --> Product Development and Support
Quantitative vs Qualitative Data
Quantitative: -Contact/Information forms (online) -Email metrics -Site analytics -Sales/downloads/growth metrics -Product reviews (star ratings) -Sales frequency, recency, other behavior Qualitative: -Interests -SEO keywords -Comments from social media -Demographics, overlays -Product reviews (comments) -Customer service inquiries -consumers more likely to shop with brands they recognize, remember, and provide them with relevant offers and recommendations-- datas enables all of these
Invest in Remarketing
RSLAs (remarketing lists for search ads) allow you to target people who came to your site, left, and then searched Google or a partner site. -allows you to hit someone who was interested and for some reason left your site. they have signaled intent very specifically and remarketing gives a marketer the ability to convert that otherwise lost customer -hits on a key need we defined earlier: near-personalization of our advertising to drive an outcome.
Informs new service offerings (example)
Raley's Supermarkets used loyalty data to determine product attributes important to customers. Result = Launched labeling program that allows customers to choose foods based on nutrient value and environmental impact.
Reach (scale) vs. Frequency vs. Accuracy
Reach (scale): are you reaching enough people to hit a brand's goals? Frequency: are you reaching them enough times to elicit action? Accuracy: are you reaching them without a good amount of "waste"?
Marketing Investment Portfolio
Shaped as a triangle (1=top,4 = bottom) 1) Shaping Markets 2) Branding / 2) Customer Equity 3) Demand Generation 4) Infrastructure and capabilities high performers put more money into branding & infrastructure, less into demand generation and shaping markets
Demographic vs. Et. Al.
Some media is almost always bought via a demographic buy; that is, you tell a media outlet the demographics of the people you want to reach and they prove to you how effectively they reach that demographic. -With digital and data, that is changing to include many other variables that drive targeting.
Customer Journey Mapping
Step 1: Research // First Touch Step 2: Compare // Lead Step 3: Price Check // Opportunity Step 4: Purchase // Close
Optimization
Technology is used to gather data, and the data is consumed by more technology to automate and optimize every phase of the digital marketing lifecycle.
Its all about data
Test, test and test again. Measure everything. If it doesn't work, don't make excuses (like "let's give it more time") Let the numbers do the talking. Iterate upon what works. Iteration leads to exponential growth.
Target Audience
The "old school" says we find ways to reach our target audience, or the people we want to deliver our message to. -They may be existing users, non-users, heavy users or light users of the product -- and each of those have different strategies associated with them.
Optimization Techniques, In Closing
The Cycle: 1) Testing provides Data 2) Data enables Research. 3) Research uncovers Intent. 4) Intent enables Optimization. The more we can optimize our ads, our social media interactions, our sites, our apps and all customer touchpoints to the needs of the people we are trying to serve at that moment, the more the brand will convey a favorable experience, leading to brand loyalty and advocacy.
What Powers Voice Search Responses
The featured snippet. ● Google's attempt at delivering the most relevant response FIRST. ● We have to answer the query or "question" within the first paragraph ● Supporting information within the page BUT... ● Correlates to potentially less organic traffic
Share of Voice
The percent of content/ads/conversation about your brand relative to the overall content/ads/conversation about your industry. -In theory, if you increase your share of voice, more people are exposed to your company and your sales see an increase as a result. -keeps the media happy. you cannot go out of advertising or "go dark" - it now encompasses social media content, discussions, organic search results, etc.
current Problem in data
The problem now is not access to data generally, it is a lack on focus on the RIGHT data. To do this well requires a well-thought plan, intent and support from all areas.
Build vs Buy
Time to engineer // cost to acquire Internal resources // external support Cost/time to customize, maintain, test & fix // n/a
Measurement (SEO)
Today, voice search measurement in its current form, is a gray area. -However, we can use normal site measurement and analytic data to gauge its effectiveness. Keyword Tracking Applications: ● Keyword Position ● SERP Feature Triggering ● Device and Click-Thru-Rate Tendencies Site Analytics: ● Page Traffic to SERP Frequency ● Content Engagement and Site Actions ● SEO Measurement best practices
Proprietary Social Intent Analysis
Trend Identification -Socially Driven Customer interviews -Competitive Analysis (traditional and non-traditional) -Native Platform Analytics -Google Search Data the ones next to each other overlap. picture of it on page 49 of voice/social intent/optimization.pdf, worth a look
Media Efficiency
Uncovering the efficiency of media relative to one another helps us in determining what media to use and WHY we should be using it. -Paid media often represents one of the most substantial investments in marketing, so making sure it is efficient is important. Data and analysis of that data helps us uncover efficiency.
SEO
Uncovering the key topics your customers are searching informs opportunities for website changes.
Customer Service
Uncovering what people are actively saying about your brand and product sheds light on potential improvements, concerns and other opportunities.
Deciphering Audience Intent
Understanding user's needs as expressed in keyword intent lays the foundation for how to create, optimize, and piece together content effectively.
Ad money quote
"Half the money I spend advertising is wasted; the trouble is I don't know which half." - John Wanamaker
Neil deGrasse Tyson Data quote
"[Data] enables cosmology to thrive on the kind of foundation that so much of the rest of science enjoys, but it also constrains theories that people thought up when there wasn't enough data to say whether they were right or wrong."
Best Practices in using data (4 Key Considerations)
#1: Know the value you provide your customer. #2: Look at data to determine a target. #3: Data needs emotion. #4: Stay nimble.
ROI on Profit
((Sales growth - average organic sales growth ) * margin ) / marketing cost more accurate depiction of ROI from the period. lowest measure of ROI in this example
Fake Door Takeaways (more)
- Business Data is like movies that say 'Based on a real story' - All data comes from somewhere - As Analysts, you are going to have to make complex decisions based on imperfect data collection methods. - NEVER base your whole strategy around a singular metric. - Make sure you understand what fact your data is describing about your business. - Make sure everyone on your team has input on your metrics. You don't live in a bubble. - Understand your data is imperfect, and account for it in your measurement, analysis and processes.
Attribution Affects Analysis
- Conversion attribution is entirely arbitrary - There are many different ways to apply credit - Each attribution methodology Introduces bias for analysis - Pick the attribution model that fits your use case
Metrics to Measure against
- Impressions (cost per impression/view) - Clicks (cost per click) - Engagements (cost/engagement) - Form fill for example - Leads (cost/lead) - Sales (CPA - cost per action), also called conversion - Cost per customer acquired (COCA)
Qualitative Interviews (2)
-An analysis on what current brand followers and fans think of the brand, how they engage, when they engage and key themes that they are talking about. -Interviews with customers who do not currently engage with the brand on social media to understand what it would take for them to want to want to be there. -What do I want from you as a customer? -What would it take for me to become a customer?
Value of Attitudinal Insights, Key Message
-Attitudinal insights derive value well above the use of simple demographic and psychographic segments or overlays. -Real time tracking of customer experiences improves the validity of research, as does anything that encourages instantaneous recording versus future recall.
What can we hypothesize?
-Better engagement on the "Request a Demo" page -Text/Image CTAs on blog and content pages -Unique and Optimized CTAs based off of page intent and conveyed messaging -And Lots More
Key Findings of Optimization Example
-Content duplication is killer -CMS Page duplication -Product needs overlap -Only one page for each topic -General usage guides CRO is driven by data. many of the problems within websites are a result of the current content organization, and data can be used to figure out where the issues lie. -The goal is for this to be "never-ending;" that is, to continue testing new methods to improve conversion and making modifications that same day the tests are done. -Practically, most companies are able to make meaningful changes like the ones shown in the CRO Report on a quarterly basis and then easy, on-the-fly changes (like color adjustments for a button) on a daily or weekly basis.
e-Commerce KPIs
-Conversion Rate: what percentage of visitors convert? -AOV: how much do visitors spend? -Time to Checkout: how fast are sales completed? -Bounce Rate: do visitors move deeply into the site? -Checkout Abandonment: at what rate do people leave? -Return Rate: how frequently are orders returned? -Marketing Efficiency: ROAS/ROI
Media Buying pillars
-Data: right data, used in the right way. -Inventory: finding inventory to match our buyer. -Technology: using a transparent platform. -Expertise: understanding how all of this works.
Example Process
-Find your data source(s). -Create a hypothesis. -Clean data. -Load data to Tableau (or other software). -Run correlation. -Review outcome: Is the data significant -- seeing same results again? Is there a relationship? If so, how strong is it?
Ethics & Legalities of Big Data
-GDPR -California Consumer Privacy Protection Act of 2018 -Adblockers, No-Script Extensions and Word of Mouth
Finding rate cards
-In a typical media buying environment, rate cards will be easily accessible to you. -You have to enter in the unit type and the cost for that unit. examples on pages 34-36
SEO is how we uncover Intent
-It used to be a verb of sorts but has become a way that we uncover through behavior what it is people are interested in and how we can connect to that as a brand at the electronic moment of value for them. -So while it is a form of optimization, its purpose is to help us build better sites, better landing pages, better copy and better interactions with those we serve.
5) Buy to time of day, geography, etc.
-Learn when the people who buy are actually online. -Don't buy all day, every day, every week, every month. -Let data tell you when orders actually come in. For instance, don't market from 11p-5a if you are only getting non-buying "looky-loos". -Don't target locations that aren't converting. -Learn if you trend to urban, rural, coastal, etc.
Performance Marketing, Key Message
-Marketing in general is moving toward a model of proven ROI. Performance marketing is the modality we use today when quantified goals matter above all else. -For performance marketing to work, Awareness, Interest and -Action within the sales funnel need to be activated. -Performance mktg. involves lots of testing. there are a couple practice scenarios just before the key message on this slide show to work on performance marketing stuff
What to optimize for
-Moments of truth in the customer journey -Trends optimization is built by understanding intent
Metrics + Decision Making, Key Message
-Test. Learn. Evaluate. Repeat. The 4 steps of modern (data infused) marketing. -Always be testing! -Insights from one test become rationale for another. -Correlations help us uncover what might matter and what doesn't, with some **.
Media Planning and Buying, Key Message
-The more we know about our audience (data), the better we can buy media. -Look at reach, frequency and accuracy. -Sound rationale proves you know what you know - it should be indisputable, not conjecture. -Buy for audiences, not for publishers.
The Landscape (Voice)
-Voice Search -Voice Products -Conversational Commerce
e-Commerce Measurement Tools
-Web Analytics: Google Analytics & Omniture (Adobe Analytics) -Dashboard Tools: Domo & Google Data Studio -Marketing: Google AdWords -Testing: Optimizely, Google Optimizer
Voice Searchers
-ask QUESTIONS instead of type short-tail keywords -use CONVERSATIONAL tones and words within their queries -want an ANSWER, not pages of possible answers -are more apt to TRUST the answer given versus bounce around and research multiple sources
Benefits of Design Thinking
-builds empathy with the end-user: the foundation of design thinking is empathy and the ability to understand customer emotions is empowering - focuses on the end-user -enables flexible value-based problem solving - accelerates innovative thinking
RET limitations
-can be hard to put into practice because you need a representative customer set to get good information. -you need to consider all possible answers in advance -provide a "reason why" for people to remain involved in an ongoing survey mechanism like this -it's only as good as what you do with it -attitudinal insights and RET are not mutually exclusive
Basis for Growth Hacking
-create a hypothesis about the people you serve (customers) -pore thru the data, mine as much as you can get: Look at internal data and third-party data. Understand your customers and your prospects, focus on early adopters, as they move the needle for others. Look to niche markets first, then expand. Look for key insights that will provoke action. Look at trends and purchase patterns. -now, look at platforms you need to hack. Social media? Video? phone? email? What combo and how? -important question: How do you build growth naturally into the actions your customers take?
Popularity of Performance Marketing
-forms of it continue to increase in popularity as data is increasingly available to measure success and as marketing in general is increasingly held to an ROI standard closely aligned with the business growth goals.
other e-Commerce Terms "worth knowing"
-hero image -product display page (PDP) -product list page (PLP) -call to action (CTA) -product description -feed -dynamic pricing -minimum advertised price -minimum selling price -attribution
4) Look well beyond Google and Facebook
-plenty of ways beyond the duopoly to drive traffic. For instance, a marketer can use Amazon if they are selling product that makes sense in that environment. Bing still powers Alexa and Cortana too.
3 Storytelling Principles
1) Identify the hero 2) Articulate your big idea 3) Utilize a story arc
5 Presentation Myths
1) Magic number of slides 2) Represent every point 3) Audience cannot read 4) Charts & tables are obvious 5) Everyone knows it's important
Tips to making a great presentation
1) Memorable dramatization 2) Repeatable sound bites 3) Evocative visuals 4) Emotive storytelling 5) Shocking statistics
6 core problems (example)
1) Reduce time spent on *decision-making* and instill patient confidence in decision. 2) Streamline the *registration* process. 3) Make patients feel their time spent *waiting* is valued. 4) Improve the *patient-provider interactions.* 5) Provide clear and consistent next steps during *discharge and referral* to ensure seamless transition. 6) Leverage MedStar PromptCare to foster a *lifelong relationship* with MedStar Health.
Ways to Win in performance marketing (6 differentiators)
1) Segmentation 2) Find people with immediate intent 3) Write awesome copy + headlines 4) Look well beyond google and facebook 5) Buy to time of day, geography, etc. 6) Invest in remarketing
Elements of Social Intent Analysis
1) Trend Identification 2) Qualitative Interviews 3) Competitive Grid 4) Native Platform Analytics 5) Search Intent Brief- Google Data
Journey Mapping (example)
1) pre-visit: first thought, trigger event, planning, deciding, online check-in 2) Arrival: way finding, arriving, check=in, registration, waiting 3) Treatment: triage, testing, diagnose, referral, prescription, discharge 4) Post-visit: follow up, prescriptions, billing, outcome 5) Referral/Transfer: referral, transfer Pre-visit --> Arrival --> Treatment --> Post-visit--> Referral/Transfer
Why is attribution modeling hard?
1. A few of us (accurately) understand the customer journey 2. Many of us don't have the data A. Data sets are limited B. Factors like multiple device usage and masked IP addresses 3. Most of us struggle with where to start
First Click Issues
1. Difficult to prove A. Depends on well maintained customer records B. Systems to track can be expensive 2. Relies on the assumption that all remarketing efforts are not important 3. Sales cycles can be too long A. Especially for big ticket items B. The media agency may be hired/fired before the sale is finalized
Before we Build/Buy
1. Get buy-in from senior leadership to invest time/money 2. Someone has to own the model (embrace objectivity) 3. Create a repeatable system for gathering data 4. Model, test and expand
Trend Identification (1)
Use third party research from folks like Mintel (you can access this as a library database for free at ASU) and Forrester to understand broader trends affecting your customers. Example: 70% of millennials are willing to pay more for a product/service that makes an impact on issues they care about.
Problems with Audience-Based Navigation
1. Takes users out of their task mindset and creates an additional step. 2. Confuses users about which group to choose. 3. Makes users feel anxious that the information they see might be incomplete or incorrect. 4. Often results in overlapping content, creating greater workload for users and content creators.
Why Voice?
100m installed home smart speaker base in 2018 Alexa achieved a 95% word accuracy rate for the English language "By 2020, 50 per cent of all searches will be voice searches." Voice is changing the landscape
Stages in the Customer Journey process
1: Empathy Mapping, see through their eyes 2: Audience Interviews, qualitative insights 3: Mapping Experiences, plotting real behaviors 4: Extracting Moments of Truth, or defining moments. key moments: highest high/low and endpoint
Voice Search Audit
A thorough, data-driven and prioritized approach to making the biggest impact in the quickest fashion. This audit focuses on bringing to light current site technical issues combined with the industry's voice opportunity landscape. Technical Considerations // Current Voice Landscape: -Domain Security // Voice Search Query Opportunity -Mobile Friendly Site // Current Organic Ranking -Site Speed // Competiton and Space -Site Structure // Featured Snippet Utilization -On-page Content Structure // Question Based Queries
Value of a Fan
Using Facebook Custom Audiences and shopper research, Facebook fans were more than twice as valuable than the retailer's average shoppers. -Major Retailer example: On average, individuals who became fans of the page increased their store visits from 24 to 62 visits within the six months following Liking the page. it pays to be engaging
Retargeting
Using the data we have on where people have been (usually on our site) to deliver a customized message to them in our advertising. - So if I went on a site looking for a flight to Tampa, future ads would reflect that for me -Many companies like Nespresso use data like this to, at a minimum, treat existing customers differently than prospective customers. -Let's face it, if you know you are in front of an existing customer, it would be good to enrich the relationship rather than try to create it again.
Look-alikes
Using the data we have to find our best customers and then try to duplicate them by finding other groups of customers with similar attributes.
Media Planning & Buying
Aligning your product or service with the people who might be interested in buying it...and doing so in the most efficient way possible. -It's all about getting your message out there to the people who matter without wasting energy with those who will never be interested.
Storytelling Flow
All about contrast. Content: what is vs. what could be Emotion: analytical vs. emotional Delivery: traditional vs. nontraditional
Ad units
An ad "unit" is the vessel that carries the brand message. For TV, the unit would be a :15 spot, a :30 spot, an on-field sponsorship, etc. example of how digital ad units look on page 32 of Media Planning and Buying.pdf
Performance Marketing Differentiator
Applied data science. we have to understand our data in order to do this. to use media within the fun that works in a complementary fashion to drive a person to take action quickly. We need to understand: -Affiliate marketing. -Lead generation techniques. -Pay-per-click/view. -Programmatic (display). -Native content. And understand how it all works together.
In the End (metrics)
At their best, metrics measure things that get a company to "why" so that strategy can be influenced. "How" matters too -- it's why we look at feature usage, time on site and such. But if we can get to the WHY we have metrics that change our trajectory.
Automation in media buying
Automation plays heavily into media buying. -"By 2022, 80% of the advertising process will be automated, the remaining 20% comprised of ... experiential tactics that will always need a human driver."
Consumer Purchasing Cycle
Awareness --> Evaluation --> Trial --> Loyalty --> (repeat)
Informs advertising and promotions (example)
Best Egg discovered people who visited third-party reviews pages signed up for loans at nearly 2x the average rate. Result = They prioritized the optimization of content to drive people to those pages.
Intent Analysis (Voice Search)
Beyond understanding the keywords and queries people use, categorizing by intent and associating search volume to the topics people are questioning most helps establish a voice friendly content strategy. we can then deduce: - Priority of creating content - Opportunity to optimize existing content - Potential reach in site traffic, OR - Potential reach in voice responses
A Marketer must mitigate:
Bounces (single request to server - so just 1 pg) - Abandoned carts - Unsubscribes - Pogo sticking (come, hit back, return to last page, in order to perform well
Programmatic Media Buying
Brands/Agencies DSP (demand side platform) --> Which impressions to buy and how much to pay for them <-- Space to sell and how to optimize what I receive for the finite space available <-- Publishers/Owners SSP (supply side platform) ultimate goal: Buy impressions across multiple publishers and platforms based on the audience we want to reach (not based on what the platform has to offer us).
Influencer Strategy
By understanding intent and preferences, gain insight into the types of influencers who will make an impact.
Attribution Considerations
CREATING A MARKETING ATTRIBUTION PROGRAM -This takes time: "2019 will be our data collection year" -This is a team effort. Everyone must buy into attribution. -Identify star contributors and others that need role refinement. -Not just digital. Offline, print, broadcast deserve a seat at the table. -Models must adapt with consumer trends and preferences make marketing investments with absolute confidence.
Customer Retention Rate (CRR) vs. Churn
CRR = % of customers you had at the start of a period that you still have at the end of period. Churn = measures the number of customers you add in a cycle/period. CRR is the opposite of Churn.
Why Growth Hacking?
Came into being for one simple reason: marketing of days past was simply too slow and plodding, not engaged in a relentless pursuit of growth. -Through growth hacking, a company creates an advantage that can be leveraged over all of its competitors.
Dashboard users
Chief Executive Officer Top revenue executive / sales Marketing Sales Operations Etc.
Site and Content Structure
Creating the landscape that promotes voice means considering the structure and information architecture where question and voice related queries flourish. Site Structure: The organization of content from a directory structure. Content harnessing an optimized site structure see higher rankings in search, and may see higher Featured Snippet utilization. Content Structure: Writing great content alone won't cut it for voice search results. Optimizing for page titles, proper heading nesting/utilization, anchor referencing, etc. help search engines understand content optimized under a conversational perspective.
#2: Look at data to determine a target.
Data determines targets. Targets aren't decided, then data accumulated about them. A company's data can really help in segmenting the right people to deliver the right message to at the right time.
#3: Data needs emotion.
Data needs humans as much as humans need data. Customer first should drive your decisions. Brands need to have empathy to the people they serve, so they need to uncover the right questions to answer, then rely on data.
Descriptive Analytics
Descriptive analytics summarize raw data, interpreting something that happened in the past. We can learn from past events and behaviors and try to apply that learning to the future. -Most of the statistics we use everyday fall into this category, including sales metrics, market share, percent of wallet, etc.
3 Types of Analytics (basics)
Descriptive- What is happening now // Easiest, most companies are here Predictive- What is likely to happen.//More difficult, fewer companies here Prescriptive- What to do next (data models that prescribe future decisions).//Difficult, category leaders here
The Truth (about Alexa)
Developers have jumped on the Alexa Skill bandwagon in hopes of creating sticky voice products that users WANT to use. Yet, there is a 97% drop off rate for Amazon Alexa Skills after only one week. -The success rate of these products is far less than what many want to admit.
e-Commerce Measurement
EVERYTHING is measured, why e-commerce is great. Measurement makes things black & white and removes instinct (to a degree) from decision making
Providing sound rationale
Efficiency is in and of itself a prime reason as to why one would use a specific medium. -Do your homework first: KNOW YOUR AUDIENCE. -Once you have this, then find your audience through media, don't find media then see how much of your audience is there.
Content Focused Ongoing Strategy
For voice search optimization and SEO in general, consistent optimization and work needs to be done. Page elements to focus on: - In-content linking and anchor usage - The content itself - Heading optimization - Media and externally referenced content
Cohort analysis (GA)
GA defines segments as "cohorts" and allows you to track cohorts. Often this is by the day the persona first entered your site. An analysis might look like this if looking at retention based on first day of exposure to an app.
#1: Know the value you provide your customer. (4 key considerations)
With too much data come issues with attention on what's important. Companies should determine that based on their value proposition. What do customers want and what can the company deliver better than anyone else. Then focus your data efforts and metrics around that.
Integrated Retail
a business operating both e-commerce and brick & mortar. -Historically, brands migrated from brick & mortar online -New trend is migrating from e-commerce to brick & mortar Key terms: Buy Online Pick up in Store (BOPIS), Ship to Store (S2S)
Performance Marketing
a form of demand generation in which a company has quantified goals (usually in sales, leads, downloads, etc.) and only pays for the marketing when a conversion event occurs. -moves risk from the marketer to the media, agency, etc.
Native Marketing (/advertising)
a form of paid media where the ad experience follows the natural form and function of the user experience in which it is placed.
Design Thinking
a methodology for innovation that combines creative and analytical approaches and requires collaboration across disciplines -Problem solving, not prettymaking
Buyer Persona
a semi-fictional representation of your ideal customer based on *market research* and *real data* about your existing customers. When creating your buyer persona(s), consider including customer demographics, behavior patterns, motivations, and goals. The more detailed you are, the better."
Model
a system for evaluating data chunks in a controlled setting to inform the marketing ecosystem at scale 76% of marketers who said they have or will have attribution capabilities this year.
Programmatic
automated bidding on advertising inventory in real time, for the opportunity to show an ad to a specific customer, in a specific context.
Why would a bounce rate be high for a particular referring source and not another?
because people are not finding what they expect to find easily. -miscommunication between the referring source and the page on which they are landing. -need to be highly integrated to be successful, and to always cater to the needs of the people you are trying to serve.
Experience vs Utility
consumer experience is much more important than site utility
Triggers
defines when and where tags are executed (GA)
History of Growth Hacking / Examples
early, classic example is Airbnb. Airbnb had a simple, elegant way to showcase rooms for rent but it needed volume of people interested in renting rooms and people who had rooms for rent quickly; otherwise, both would never come back. -Marketing would have set out a paid media budget to drive people into the top of the funnel, likely one budget for renters and one for rentees Instead, Airbnb understood the market and the fact most people were posting rooms for rent on Craig's List, so they "hacked" into those listings and drove people to list in both places. -Then they made their place much better -- they helped people photograph their spaces for instance. -It helped everyone and it grew exponentially as a result. After time, they added entire homes, etc https://growthhackers.com/growth-studies/airbnb. Dropbox also had a 'refer someone, get more space' offer that helped them grow more than 3200% in the first year -Wall Street Journal offered free wifi across NYC in exchange for email capture to put more qualified people in their sales funnel -Zappos surprised people with 2-day free shipping, which is now called 'surprise and delight'
Design Thinking steps
empathize, define, ideate, prototype, test (observe, reflect, make)
Marketing Tests
everything can and should be tested. -Messaging -Price -Creative Call-to-Action -Images / Videos Know your sources and track your customers: -Use source codes -Use tracking pixels -Use unique URLs and other identifies that do not rely on the consumer. Perform thorough post-analyses. Learn from everything you do (what went wrong, right, what could be improved). Called a "post-mortem" -Standardize the process so data coming from your efforts is consistent and employees can easily tell what works and what doesn't -Insights become rationale- use them as rationale for strategic and tactical adjustments, additions, or deletions to marketing efforts -Marketing intelligence should explain any outcome, good or bad
Fake Doors Inc
example from attribution guest speaker #2. goes thru a situation in establishing a new company and how to get it started, worth reviewing
Finding "Spot On" Content
find the intersection of consumer needs, brand goals, and areas of opportunities allows us to create content that resonates and delivers value to customers
Sales (growth) Key Metric
for most companies this is a key metric. -If you don't have sales and/or profits, a company ceases to exist. So it is practical that a key metric often has to do with sales, revenue, profit, growth, etc -imporant to not rush to add metrics and to add them accordingly, but thoughtfully.
Attribution
from last click to sophisticated. -attribution modeling gives "credit" to media that drive people down a path to purchase, rather than assume the last point of contact was the one responsible for 100% of the sale.
Voice Search
gives advertisers the ability to optimize content and site visibility in a manner that future-proofs a brand for shifts in user search behavior as well as the evolution of new technologies.
Goals vs. Objectives vs. Aspirations
goal = a broad primary outcome objective = a measurable step you take to achieve said goa outcomes = tangible results that work toward the goal
Data in Media Buying Decisions
helpful. The more we know about our audience (the people we want to serve), the better we can target our funds and energy to interact with them where they are and where they want us to be.
Contact Research Organization (CRO)
iNVOLVES LEARNING THROUGH TESTING -In order to rank well in search engines and convert visitors into leads, a website must have unique, findable and useful information for its end users. -HUGE example starting on page 16 of Optimization Techniques.pdf. seems important
Facts First
important to sucessfully utilizing buyer persona. -When you work from a "facts first" mentality, you build a persona that sites, campaigns, content, etc. that resonates and drives the actions you want. -When building personas, you are creating them based on segments of customers. These segments should be materially different in how their needs are to be met.
Data enables less silos
in organizations. Key executives agree that eliminating organizational silos is critical to success and the biggest barrier to communication and information sharing. -Data should be part of the strategy.
Intention
intent = a determination to act in a certain way -our goal is to understand both active and passive intent and act on it Active intent = obvious what they are searching for, ex: "Mexican food 85281" -Influence action: by understanding why, you can then influence how they feel about a brand at the right time and place -You cant be everywhere, always: simply too expennsive -Its not about the brand: don't advertise brands, bring solutions to customers
Product Opportunity Audit
intersection between customer opportunity, market opportunity, and voice behavior to drive voice products and skills that will resonate with the people you serve and drive action. -customer interviews: Device usage / customer interaction -white space analysis -voice skill idea scorecard -product strategy workshop: product canvas
Growth Hacking
is a relentless focus on the growth of an organization. -It is not marketing, although marketers are increasingly asked to focus on growth. -True growth hacking does not engage in activities that do not directly lead to growth.
Growth Hacker
is part marketer, part product manager, part revenue officer and often someone adept at technology and engineering. -They obsess over customer acquisition and retention (no churn) and believe in exponential growth over incremental growth.
Marketing Optimization
is the process of improving the marketing efforts of an organization in an effort to maximize the desired business outcomes. -THE ACT OF ENABLING YOUR CUSTOMERS TO FIND WHAT THEY NEED, WHEN THEY NEED IT.
Data Goal (according to google/econsultancy)
is to connect with customers in ways that will grow the relationship, so loyalty is established and the value of the customer increases. -slow movers will be disrupted by a changing market. For the leaders, data is a priority and their ownership of that data is at the forefront of their success.
Fake Doors (my) Takeaway
it is hard to accurately attribute, as the customer goes through several websites multiple points at different times in their journey -leads to conversion duplication. sales may not view conversions the same as marketing. important to track the customer journey and get rid of duplications. -dont assume 1 conversion = 1 purchse -establish a source of truth with the team -dont take the metrics at face value
Common attribution methods
last click first click ("the hype man") linear ("the egalitarian") position based/u-shaped ("first and last") time decay ("faster is better") There is no perfect attribution model, and there never will be!
Design Innovation
lies between: Technology (feasibility), Business (viability), and Human Values (usability & desirability)
Companies using Performance Marketing
many types of companies use it in their marketing arsenal. examples: -Car dealers/manu to drive car shoppers. -Mortgage brokers to drive refinancers. -B2B companies to generally drive leads. -E-commerce companies to drive immediate sales.
Research often relies on _____
memory. When we ask people to recall events, particularly those that happened over an extended period, we risk degradation in our data simply because people don't remember everything they saw, where they saw it, what it said, etc. this is why we use ethnographic research - it allows us to intercept a person while they are behaving in the way we want to research
3) Write awesome copy + headlines
need to give a person a reason why -- a reason to read. -Nearly all studies show that good headlines give people a reason to click, to open, etc., thereby giving the marketer a chance to make the sale. -best way to write an awesome, compelling story that motivates is to test all ad copy and headlines, and replace the ones that perform as well continuously.
Intent Based Playbook
non branded intent -- branded intent -- brand focus, clear path to buy -- content topics by persona Value of Intent Based Playbook: Same major retailer. -Using Facebook Custom Audiences and shopper research, Facebook fans were more than twice as valuable than the retailer's average shoppers. image on page 58 kinda helps understand intent based playbook
Why isn't all marketing performance-based?
plenty of scenarios where it doesn't make sense. - New product/service launches, awareness - Trust-based (build reputation) - Metrics cannot be tracked/measured - Different embedded mechanism (TV=GRPs)
Successful Marketing
resonates and is relevant. Not from the boss or a hair-brained idea, instead a focus on the customer journey. -what makes sense to offer someone at the key moments in their journey? -uncover through intercepts, surveys, digital ethnography, empathy mapping, CRM data, etc. -Test, Learn, Evaluate, Repeat. Look at how & where, Isolate, and learn in an iterative approach.
Aquimo
skill-based competitions to fans with brand sponsored prizes, digital coupons, or cash prizes. Current Stadium Marketing: Commercial on the videoboard - 31k attendance But its only seen by 25% of the audience Stadium Marketing with Aquimo: -3X premium impressions in stadium via eSports experience -2X premium impressions outside the stadium with TV -10X overall brand marketing impact -Closed loop marketing with digital coupons driving customers to your products
Tags
snippets of codes added to a page. used in Google Tag Manager
History of Performance Marketing
started digitally in earnest in the early 2000s through affiliate marketing. -Affiliates were paid a 'bounty' for each person they got to act based on what a company might want. -It often was a lead or a sale and affiliates were paid a set amount for that lead/sale, regardless of what it cost them to acquire it.
Market + Concept Validation
testing prototypes with customers. -During this phase, take the product concept out to test with the market, moving into concept validation, which involves quickly building and testing a prototype 1) Understand- leverage data on user needs, leverage digital persona for testing 2) Diverge- envision, multiple solutions, ideate 3) Decide- choose the best idea, storyboard the idea 4) Prototype- prototype multiple voice dialogues, focus on usability not making it beautiful 5) Validate- show the prototype to real users, learn what doesn't work
Business > Marketing > Media Objectives
the company's business goal needs to be supported by a marketing objective that is then supported by media choices. -Doing this well takes time, as we are in a world where the people matter -- we go where they are, not vice versa. - From objective comes strategy and tactics. --> From tactics come budget. -Rationale is embedded throughout- proof points that show we know what we are talking about
Return on Investment (ROI)
the contribution of profit attributable to marketing divided by the investment in marketing.
Affiliate Marketing
the process of earning a commission by promoting other people's (or company's) products. You find a product you like, promote it to others and earn a piece of the profit for each sale that you make. (he didn't define these terms on the slides, but there are picture examples on the Performance Marketing.pdf starting on slide 13)
Attribution (from attribution guest slides)
the systematic process of quantifying contribution
Macro Conversion
they complete the main activity we desire (such as purchase).
First Click (Attribution)
translation: "from a little spark may burst a flame"
Time Decay (Attribution)
translation: "short term memory illustrated"
Last Click (Attribution)
translation: "standing on the shoulders of giants" tons of steps stacked before the last click
Linear (Attribution)
translation: everybody gets a trophy
Position-Based (U-Shaped) (attribution)
translation: ice breakers and deal closers little interaction between first and last click
Experience
trumps all else. Experience = habit = profit Competiros are milliseconds away, switching costs are low
Counting Conversions
use tracking pixels to get more out of conversions, track their future habits
Common Metrics
used at the higher levels within an organization to measure the effectiveness of marketing in a holistic manner. -Return on Investment (ROI) (ROI on sales, ROI on marketing sales, ROI on profit.) -Customer Retention Rate (CRR) and Churn. -Customer Lifetime Value/Lifetime Value (CLV/LTV) -Conversion Rate -Response Rate -Customer Acquisition Cost (CAC) -Monthly Recurring Revenue (MRR) -Market Share -Share of Wallet -Share of Voice
Real-time experience tracking (RET)
used by large companies like Pepsi and Microsoft. It too relies on cell phones Focuses specifically on four key research areas: ● What is the brand involved ● What is the touchpoint (TV, call, social media) ● How the person felt about the experience ● How persuasive the experience was
Net Promoter Score (NPS)
usually a key metric. -Many companies measure customer satisfaction and the predominant way to do this is through NPS, then compare your results over time, among target segments and against your industry's baseline.
Intro to Social Intent Analysis
we are in a social commerce economy. Facebook accounts for 64% of total social revenue = 3rd most popular internet browser -however, we are flooded with noise, namely irrelevant content and ads from brands. -brands are in trouble: general decline in reach amongst major brands on facebook
Section Pogo-Sticking
when people jump from one page back to another while navigating a webpage
How does planning start?
with research. -research can come from data about your existing customers (1st party) or data from outside the company, often supplied by ad agencies, market research firms, data aggregators, etc. -through these research trends, competitive information, audience engagement and so forth is analyzed. -Ultimately, media planners look to unearth behaviors, information sources, media choices and influences to behavior. This can then be aligned to the company's objectives.
Advertisers Must... (Voice)
● CREATE more conversational content for websites, landing pages, and social media based on data telling them what people are looking for. ● ENSURE site content employs structural and markup based best practices - think of headings, metadata, alt tags, etc. ● OPTIMIZE visibility in Google My Business, Google Express, Amazon, Bing Places for Business and other native voice backing applications... even Wikipedia!
Data drives transformation
● Helps companies become customer centric ● Helps make informed decisions ● Enables "one-to-one" relationships that are otherwise not available with large customer bases ● Informs product ideas and launches, new service offerings, advertising + promotions
Tips to achieving exponential growth
● Let data be your guide. -Don't listen to gut instinct. Speak with FACTS. ● Think differently. -Be wild not mild - don't follow what everyone else is doing. ● Learn. Learn. Learn. And that means lots of tests. ● Eschew hierarchy. -Everyone is a "do-er" and knows how to do things.
Ways to grow CLV/LTV
● Use data well to personalize your approach with customers. Know your customer well! ● Provide a level of customer service that stands out, where people really take notice. ● Make every interaction with the customer as easy as possible for them. Simplify!
Voice Search (Process)
Voice Search Audit --> Intent Analysis --> Content Structure --> Content Creation and Optimization
Renuzit example
We examined over 1,800 unique keywords centered around Renuzit as well as the generic "air freshener" topic. -Pulling average monthly search volume over the last 12 months and then breaking out into differentiated categories, we are given a broad understanding of how people are searching for information and content relevant to the Renuzit. -Air Quality by Modality: to better understand what those users using generic terms are eluding to -Air Freshener Types: insight into how popular certain products are -Generic Categories: highlight a few key Needs within air care related search queries -Generic Categories, Renuzit only: same categorical breakouts w/ a filter only showing keywords associated to Renuzit-- Renuzit's branded search intent is largely built off these primary categoiries -Deficiency vs. Opportunity: shows Renuizts organic reach compared to key competitors. discovers new ways to begin optimizing for and becoming more competitive within digital space -Areas of Opportunity: 1) a growing trend in oils. diffusers, and DIY 2) air fresheners vs. odor removers/neutralizers 3) Safety, allergy related information, MSDS access 4) Sales and conversion focus, bottom funnel CTAs 5) Featured snippets and long-tail optimization 6) content buildout and optimization best practices
Correlation Coefficient
We want our Correlation Coefficient r to be farther from 0 on a -1 to +1 scale. A -0.7 to -1 and a +0.7 to +1 represent strong relationships, either positive or inverse. Is "moderate" good enough (+/- 0.5)? In marketing, PROBABLY NOT. closer to 1, positive or negative, means strong correlation
Data Significance (p value)
We want our P Value to be low; otherwise, there might not be enough data to make the outcome significant -- or, is it likely to repeat this outcome with other data sources. We want low -- under 0.05 or hopefully lower.
Search Intent Brief- Google Data (5)
What are the key topics a brand's customers are searching for and how does that shape social strategy? -You are trying to uncover common trends within search behavior, keywords, and inquiry phrases to provide insights on potential topics for the social channels. -This data is used to validate what came out of the interviews.
Identifying Key Metrics, Considerations
What is it that actually drives business? How can that be measured? Are we measuring things that ALL stakeholders agree are critical? Are we measuring only 'mission critical' items or 'nice to haves' as well? Do we rely on and believe in the data we have? Or will we dismiss it? Can we act upon the trends we see?
Growth Hacking (essential questions to be answered)
What is the value to my customer? My prospect? How does this align with my brand promise? How does this improve things for the people we serve? For us? How do I put in a continuous loop of testing so this improves upon itself?
Testing
What should we test? Everything! common tests: colors, images/buttons, $ vs %
e-Commerce stats
-$4.5T by 2021 -23% annualized growth -57% of shoppers have completed international transaction -China accounts for twice US e-commerce sales (~$700B) -Amazon is 44% of e-commerce sales Brick and mortar is still 92% of retail sales, but e-commerce has unparralled growth
Data & Customer Journey, Key Message
-Customers go through many on- and off-line touchpoints with a company, and their experience largely rests on how well a company addresses each touchpoint (and in particular the most important and last ones). -Customer journeys, personas and customer segmentation all benefit greatly from data -- or proof points -- that drive the story (rather than the story driving the facts to include)
Mandatory Skills for Today's Growth Leader
-Data and Intelligence -Market Insights and Knowledge -Holistic View of the Customer Journey -Brand Building and Development -Storytelling in a Digital World
Where Data Comes From + How It's Used, Key Message
-Data needs to be a part of company culture, meaning decisions should be made with data from the top down. -Companies need to stop letting other organizations "own" their data and create direct relationships with their customers. -Data should be centralized without being a silo. Easy to say, hard to do. -Data drives the customer experience because it leads to personalization, empathy (understanding) and relevant conversations/communication.
5 Types of Goals (GA)
-Destination: a specific location loads ex: Thank you for registering! web page or app screen -Duration: sessions that lasts a specific amount of time or longer ex: 10 minutes or longer spent on a support site -Pages/Screens per session: a user views a specific number of pages or screens ex: 5 pages or screens have been loaded -Event: an action defined as an Event is triggered ex: Social recommendation, video play, ad click -Smart goals: when enabled, GA automatically evaluates your website or app visits and assigns each a score, with the "best" visits being translated into Smart Goals.
How Data Informs Marketing, Key Message
-Getting the right data in front of you is important. -Making sure the data right in front of you is accurate is equally important. -Uncovering insights and turning them into actions (provocations) is the most important step that often never happens.
Key Message from Intro Slides
-Getting the right data in front of you is important. -Making sure the data right in front of you is accurate is equally important. -Uncovering insights and turning them into actions (provocations) is the most important step that often never happens.
KPIs and Metrics, Key Message
-Leading and lagging indicators help us in measuring how marketing is achieving KPIs. -It is critical to have the right goals. Make sure you are measuring the right things! -GA and Tag Manager provide a great solution for tracking at little expense, just don't "set it and forget it." Test, learn, test again.
Critical Skills for Tomorrow's Growth Driver
-Real-Time View of the Customer -Understanding of Entire Product Portfolio -Customer Experience Strategy Ownership -Finance-Based View of Engagement Strategy -Field Sales Knowledge and Experience
Internal Data Sources and Metrics, Key Message
-There is a proliferation of data -- access is now less the problem. Now focus is the issue. -Choosing the right key metrics to measure takes time, consensus and a sense of trust (in each other and the data). -ROI is important, but CLV and other long-term metrics might be more important.
Google Tag Manager Example
-Track where a person came from (the referring source), where they went on the site, what they did pre-purchase and how they got to the action page. -It could be as easy as tracking an "add to cart" action when looking at why your ecommerce sales are off. Years ago, a developer would need to write code to track that; now, anyone can add a tag through platforms like Tag Manager.
Market Share
-shows competitiveness of a company. -shows the strength of a brand/company relative to its competitive set in an industry. ex: Sales of $10,000 in a market with total sales of $20,000 means you have 50% market share. If you increase sales to $12,000 that looks like growth. But that is not the case if total market sales grew to $30,000. Now your share went from 50% to 40% even though your sales grew. -important to define the competitive set to get share numbers that are meaningful.
e-Commerce Values
1) Be Fast: Speed doesn't go in slumps. Site Speed. Browsing & navigation. Fulfillment. All need to be lightning quick 2) Be Beautiful: Aesthetics matter. They are subconscious indicators of quality. Half of people leave sites they call "unattractive" 3) Be Consistent: Habits are made possibly by consistency and simplicity
Data enables educated change (2 types)
1) The adoption of a new capability or a response to consumer behavior 2) Enduring transformation and/or large-scale reorientation (causing restructuring and rethinking of entire companies)
Steps to set up a test
1. Come up with a hypothesis to prove/disprove. 2. Determine the goal to measure it. What do you want answered? 3. Create a report that will answer the question. 4. Put the appropriate code (tags) on pages to measure. 5. Analyze the data as it comes in. 6. Take action and test anew once you have enough data. 7. Continue to monitor and iterate.
Dashboard
A dashboard provides a visual way for people to see how we are doing relative to our KPIs or our success metrics. The dashboard is clean and generally has limited information -- only the key information is included! Busy dashboards don't get used!
Industry-Specific Metrics
An e-commerce company, for instance, likely has different metrics than does a brick-and-mortar store. For instance, time on site might be important for ecomm while same-store sales might be important for a physical store. At the same time, add-on sales, repeat sales, complementary sales, etc. might be important to both. -a subscription-based model like Netflix and Birchbox likely measures churn and net recurring revenue (NRR) while non-subscription companies are less likely to consider these.
the Promise of Big Data
As a company, you can uncover secrets hidden within your data that lead to insights that provoke actions to give you an unfair competitive advantage. -the promise of big data has not been realized by most marketers. The ones who realize the promise are getting far ahead, and are true leaders. As more companies have the funds to acquire and understand AI, this realization changes (increases?) year to year.
CLV metric
CLV is the present value of all of the profit streams a customer will generate over the term that customer does business with a company.
CRR Formula
CRR = ((E-N)/S*100) E = # of customers at End of period N = # of New customers added in period S = # of customers at Start of period
Data Informs Product Ideas/Sales/Launches (example)
Coca-Cola analyzed Facebook data to determine if there was a relationship between check-ins and likes and propensity to purchase Coke products. Result = The more check-ins and likes, the better the prospect, so Coca-Cola spent more time on those bars and restaurants than others.
Conversion Rate
Conversion rate is measured by a "visitor" who takes "action," which often means they buy something. It can be used on- and off-line. Online (measured thru GA): Actions/Clicks Offline: Actions/Store Visits 2018 avg rate = 2.3%, top 25% of sites = 5.3% so a high-performing site like Amazon gets over 5 people who visit to buy while an average site only gets 2 people to buy. It's why the leaders continue to pull away from the fray.
Organization of Data Assets
Data Engineer (store, manage and clean data) --> Data Analyst (crunch data with machine learning, then deliver new insights)--> Data Communicator(take analysis and insights and translate to business operations) -Data should be centralized so there is one source of "truth" in the organization
KPI
Key Performance Indicator. (Hopefully) a leading indicator that tells the marketer how things are going. In general, the Fewer the better. KPIs ladder up to success and are often measured on a dashboard. -many companies track lagging indicators, which show how things went in the past. These aren't as helpful as leading indicators, which predicts how well you will do in the future based on current information
Leading Indicator
Measure an action that leads to a KPI. Should tell you in advance if you are not tracking to your KPI so you can take action. -A good leading indicator tells a marketer what to expect in the future in terms of a key success metric. It is different from a lagging indicator in that the lagging indicator simply reports on what already happened.
One Big Reason (why to continue marketing despite bad looking ROI)
Most of this is transactional, looking at a snapshot based on or assuming one sale per person. We should, instead, consider the value of a customer (or group of customers) over a period of time -- in marketing we call that their lifetime value.
Data in Peril?
No easy answer here. Clearly some of the data we once relied upon from external sources might not be as readily available as it once was. -facebook data eliminated partner categories targeting but still has its own targeting partners from its own data (but there is still usable data from facebook for advertisers) -EU: General Data Protection Regulation (GDPR) = If a US company sells into the EU in any way. it must be compliant or face penalties. different from US, in Europe there is substantial concern from public over privacy and how data is accessed Key components:-A person has the right to be "forgotten," all data erased. -Portability: opt-in/opt-out notices must be clear and precise -A business needs to have a well-thought response to a data breach and employees trained on this. _California Consumer Privacy Act of 2018, goes into effect 2020
Predictive Analytics
Predictive analytics provide actionable insights based on data by estimating the likelihood of something happening in the future. These are used to forecast based on data you have, filling in the holes with educated guess. -Examples are customer behavior, purchasing trends, etc
Prescriptive Analytics
Prescriptive analytics offers advice and turns insights into provocations. They quantify the effect of future decisions to advise on outcomes before decisions are actually made. Can fundamentally improve the customer experience. - generally predict multiple futures so multiple outcomes can be reviewed. This is a likely area for AI, machine learning, etc. -goal of prescriptive analytics is to have the right thing to the right person at the right time, all the time. If achieved, customer satisfaction improves, which generally improves the lifetime value of that customer as well.
Lagging Indicator
Report on progress toward a KPI from one point in time to another. These are a historical perspective of actual performance.
Lagging vs. Leading example
Sales (lagging) vs Pipeline (leading) Website visitors (lagging) vs Content Syndication aka number of mentions (leading) - The more we can find leading indicators that are an accurate reflection of a future state, the better we can use KPIs to manage the growth (or success) of a business.
ROI on Sales
Sales growth in a period / Marketing cost in same period ex: if sales growth is $42,000 and marketing costs are $12,000, the ROI is $42,000/$12,000, expressed as a % or a ratio. -350% ROI, or a 1:3.5 ratio (for every dollar spent, 3.5 dollars come back in sales)
Establishing Success Metrics and KPIs example
Step One: Agree on the business goal(s) Step Two: Establish Success Metrics - these should be limited, specific and time-bound. Step Three: Establish KPIs based on Step Two. These measure what needs to be in place for the metric to succeed. You can have multiple KPIs for one Success Metric. Step Four: Identify the Measurement (Analytics). Is it a %, a ratio, an average, a rate of increase, etc Business Goal: Increase sales Success Metric: Increase the quantity of customer leads by 17% compared to total customer leads generated last year KPI: Percentage of site visitors who complete an inquiry form on our site (email submission, more info submission, etc.) Measurement analytic: Prospects/Total Visitors (monthly)
Lagging Indicator Sources
There are many sources for lagging indicators. ● Sales performance data ● CRM data ● Analytics data ● Financial data ● Marketing automation data ● Advertising data (from Facebook, publishers)
Iceberg Analogy
There is far more to flawless execution than meets the eye i.e.: Merchants: What products will we sell? -Pricing: How much will they cost? -Operations: How will we sell them? -Supply Chain & Distribution: How will people get their stuff? -Marketing: How will people know about us? -Call Center: What if there's a problem?
#4: Stay nimble.
Things change quickly and a company should be prepared to pivot in all things, and data and metrics are no exception. A brand risks being disrupted if it doesn't stay ahead of the data curve.
Maximizing the Customer Journey
Using data points in designing customer experience, we can address questions like: ● What percentage of customers take a given path relative to other paths. ● What archetype of customer takes each path and what sets them apart. (several more questions) -this data can be hard to access, 60% of data across social platforms is dark. other sources like google and GA provide data, as well as search data and internal factors such as your analytics package, CRM software, and so on.
Importance of Owning Data
VERY important. Owned data = first-party data. -It is information that a brand or company collects about its customers through increasingly digital transactions. -If a brand relinquishes data ownership to someone else, they are unable to learn about their customers and their behaviors. Happens all the time (ex: when you buy Pringles at Frys, only Frys knows that it is you who likes Pringles) -Many service/product based businesses rely on other media to promote their goods and services (Google, FB,TV networks, etc.) Those media companies gather the data about their users and will not pass this information along for free
e-Commerce Models
Volume: Wal-mart, Target Premium: Louis Vuitton, Chanel Membership: Costco, Jet, Amazon The beauty of e-Commerce is that there aren't the typical differences among these: real estate, employees, aesthetics, etc
Metrics depend on the 'who'
Who within the organization dictates what should be measured intently. For a product manager, looking at product features usage and heatmaps of eye movement might be important, while those are not important for a CEO looking to make overall business decisions or a CMO looking to budget media buys.
Customer Acquisition Cost (CAC)
considering the total cost to the company in actually acquiring a customer EQUATION: (the cost of paid media, agency fees, internal salary costs, commissions, bonuses, overhead for that part of the business) / number of new customers. -You can look at this by segment to determine your more profitable customer segments and then focus more growth strategies on them.
Customer 1st Mentality
data informs efforts towards this. used to be company first, now conversations abound around: ● Having a fluid conversation across touchpoints ● Understanding customers better ● Personalizing every communication -the customer journey "meanders over multiple touchpoints" and those can be on- and off-line, start with mobile or in-store, etc. -This starts by allowing all areas of a company access to important data so they can uncover insights and take the necessary action.
"Mainstream' company view on data
data is simply an abstract goal rather than something engrained in their culture.
Buyer Persona effectiveness
done correctly, they work. 900% increase in site visit length, 171% increase in market-generated revenue, etc. Even if every case does not produce the results of this study, they are proven if they rely on FACT.
Big Data
google: Extremely large data sets that may be analyzed computationally to reveal patterns, trends, and associations, especially relating to human behavior and interactions.
Conversion
how goals are achieved. when a person does what we as a marketer want them to do.
Response Rate
is measured as the number of people exposed to your message/advertisement (called impressions) relative to the number who click on it. For an in-person store it might be the number of people who saw an ad versus the number who visited a store. -often stated as click-through rate digitally and often very low
Dashboard usage
many small companies do not have dashboards and as a result have a hard time understanding how they are doing. visualization of data separates haves and have nots. dashboards include different components depending on the company, but most dependent on the audience. there are likely multiple dashboards displays within one company.
ROI on marketing sales
more complex than sales because sales doesn't account for organic growth. (Sales Growth - Average organic sales growth) / Marketing cost still doesn't account for margin (doesn't look at COGS)
Variables
receive and store information to be used by tags and triggerss Tags-->Triggers-->Variables, Google Tag Manager
Successful Brands Focus
should be intently on data. -First party data impacts "all significant functions of the enterprise, including product development, customer value analysis and pricing." -Nike/Nordstrom both recently bought out data based companies (consumer data analytics/AI personalization) and are both high in their industry looking to get an upper hand -Data is a strategic differentiator
Day-in-the-life of a Data-Driven Marketer
support from key executives, elimination of silos between departments and is inclined to insights/facts over gut instinct. Analysts take large volumes of data and make sense of them, they need to be able to trust and understand the data collection and management, and they don't just run regression analyses day after day. - They measure success and build from that success
Customer Journey
the journey that a customer goes through in the process of discovering, finding, and purchasing a product -a company's biggest differentiator is rooted in customer experience (CX). What journey does the customer embark upon and how do we as companies make it easier, better, faster. -drives three types of loyalty: retention, enrichment, and advocacy -should always focus on how to be a loyalty company, not how to run a loyalty program. good brands retain customers because they love them and enrich their lives, so they advocate on the company's behalf
Success Metrics
the metrics a business uses that ladder up to the ultimate success of the organization as a whole. What success looks like for you as a marketer that contributes directly to the success of the goals of the organization. As a marketer, you should always know how you are being measured.
Bounce Rate
the number of people who come to your site but leave without taking any action. -shown by referring source, so you can look at those with high bounce rates and try to determine why they are high, then test ways to reduce them.
CLV
the only metric that matters. sports fan analogy, visible proxy of loyalty to a brand. -that loyalty is measured through LTV, or the amount of profit you can expect to generate from a customer, from their first interaction to their last. -CLV can look at customers in general or, more effective, can look at customer segments to help determine who a company's best customers are, then create marketing for look-alike audiences to that customer. -the more loyal a customer becomes, the less you spend on acquiring them -top-of-funnel are most expensive for a brand, so keeping customers is where profit is derived. 80/20 rule
Share of Wallet
the percent of a person's wallet s/he spends on the company's brand relative to other brand's in the competitive set. how to get the information: ● Intercept studies (ethnographic studies) ● 3rd party data (larger industries) ● Surveys
Micro Conversion
they complete an activity on the road to the main conversion event (such as signing up for an email newsletter before buying).
Goal of Customer Journey Mapping
to uncover points along a customer journey that causes pain or pleasure/delight, and then back those recall events with data to prove/disprove them
Monthly Recurring Revenue (MRR)
used in subscription-based and retainer-based companies (often start-ups) as a key metric in determining a burn rate on cash -shows forward momentum. companies in subscription-based models want to continually improve MRR to cover fixed then variable overhead costs -Many start-ups have MRR as a key metric as it shows them how much they can not only expect this month, but also next month and so on, accounting for churn. This is why we want low churn.
Data
webster definition: facts about something that can be used in calculating, reasoning, or planning. Wikipedia: Data is measured, collected and reported, and analyzed, whereupon it can be visualized using graphs, images or other analysis tools. Data as a general concept refers to the fact that some existing information or knowledge is represented or coded in some form suitable for better usage or processing.
Tips to establish great KPIs
● Make sure your entire team agrees on what is "key" to be measured. ● Make sure if the indicator is met that your goal is also met. ● Make sure your KPI can be easily measured with a data source you have control over. ● Make sure you have a team in place to monitor how you are doing against KPIs. ● Make sure you limit your KPIs to only a "key" ones.