Module 1 Videos Flashcards
Two Types of Research Approaches
Primary Data (user) Secondary Data (other than the user)
Secondary Data Resources
Authoritative Blogs (MOZ)
Google Trends & Correlate
Platforms (Adwords, Keyword Planner)
Syndicated Market Research (eMarketer)
White Papers
Primary Data Resources
Your own data by
Listening Labs Surveys Testing-split A/B Transactional Data Web Analytics
Google Trends
A website by Google that analyzes the popularity of top search queries in Google Search across various regions and languages. The website uses graphs to compare the search volume of different queries over time.
Used for SEO but mainly for PPC and social media-topics/keywords
- Seasonality information
- Filter by location
- Identify related keywords
Google Correlate
Shows many different variables that interact together commonly that can help you predict a certain outcome.
Careful–some correlations are just a coincidence or of no value. Split testing is what decides things best
Online Surveys
Vendors have templates w/typical questions for certain
situations, (e.g., customer satisfaction) they can be edited
Paid versions have advanced functionality such as skip logic, validation, chart analysis & statistics
(Keep an eye out for Response Error)
Some online survey vendors
www. qualtrics.com
www. surveymonkey.com
www. zoomerang.com
www. surveygizmo.com
www. polldaddy.com
Response error
Sometimes results might not be representative of your overall customers (e.g., if 80% of customer buy in brick-and-mortar stores but only online customers are surveyed)
Online Panels
Like an offline focus group or 1:1 personal interviews.
Specialized panels by Vertical or by Category (you can then design your own questions).
Main benefit is access to hard to reach target market (e.g., physicians).
ex: Research Now
Biggest source of primary data
Data from web & social media analytics data
Buyer Personas (using primary research)
Iconic representation of a target segment with a symbolic user-usually fictionalized aggregate. They are research-based (web analytics and transactional data), grounded in data, facts and actual interviews with customers.
Benefits:
- Personalize your messaging/offers
- Improve your web/social content
A buyer persona development process will typically take about 6-8 weeks to complete.
The Customer Journey
Awareness Research Decision Pre-Purchase Purchase Post-Purchase
Add a timeline for each phase-(e.g., 3 weeks)
This data comes from surveys of current customers
data from online tools (e.g., Google Trends) and your
own online/offline data-may need more than one
journey if different buyer personas.
Audit each step, see where problems are occurring,
decisions being made & where opportunities may exist
Awareness
How do potential customers find out about your product?
e.g. social media, TV, billboards
Research
Product websites, social media, search it on Google, ask friends, etc.
-consumer behavior comes in-
Decision
To make or not make the purchase.
Pre-Purchase
Where to buy? Check inventory. Engage in social sphere to help make final decision on product.
Purchase
Make the purchase.
Post-Purchase
Go back to social sphere: leave reviews, tell friends/family about the positive or negative experience.
Digital Experience Journey
Sometimes companies may create an “experience” only journey vs. buying stages
Company Strategy
Company’s Marketing Strategy > Digital Marketing Strategy > Digital Tactics
Strategy v. Tactics
Strategy Tactics
Who is responsible? Upper management | Mid to lower
level managers
Time frame Longer Shorter
Tools Market Research,
customer and market
data analysis. Then use of strategic
techniques such as Porter’s Generic Strategies,
Ansoff &GE matrix
Email, PPC, SEO,
Social media, UX, mobile,
web analytics,
conversion optimization, SMS,
etc.
Activity type Mainly planning/monitoring | More operational-
execution
Competitive risk Harder to copy Easier to copy
Traits More high level-the “what” Process oriented-the
“how to achieve”
Strategy
To be effective it cannot be “All Things to All People”. Need a lot of focus.
Should address these questions:
- What do we do?
- Who is our customer?
- What business are we really in?
–some look at the perspective of what they sell, but that can be too narrow. Trains don’t have to be railroad business, but transportation business.
Porter’s Generic Strategies -Company level
Low Cost (ie Walmart)
Differentiation between Real and Perceived
–have a unique way of delivering the product that sets it apart. (is a $100 perfume REALLY better than a $30 or is it just advertising?)
Focus Segmentation–to start off in ONE segment and branch out to other segments later.
What Strategy Is Not…
These are more like unrealistic tactics…
Let’s do a viral YouTube campaign…or everyone is using Instagram, Snapchat etc.-we need to jump in
Let’s be the Apple of toothpicks, staples, etc.
Launch now… we will figure out how to make money later
Make it free….
What is a tactic?
Tactics are how Strategic goals are achieved
Earned-not paid/organic Digital Strategy
Facebook, Twitter, Instagram,
Pinterest, YouTube, SEO, etc.
Paid Media Digital Strategy
Search or Display ads
(e.g., be it on AdWords, Facebook, Twitter, website
banner ads, etc.)
Owned Digital Strategy
E-mail,
Blog,
SMS,
Push notifications
What is the Digital Strategy End Game?
SEND THEM TO YOUR WEBSITE
Common mistake: Sent to your website, but they see a post to visit them on social media instead and engage there.
Best Payback in Digital-General Rule
- Usability-UX
- E-mail, SMS & Push
- SEO
- PPC
- Social Media
Considerations: Is it owned, earned or paid? … where do conversions occur?
Law of Diminishing Returns
More Effort = More Return (conversions)
Less Effort = Less Return
7 P’s of Services marketing
People
(customers, employees, social interactions, roles & scripts, relationships)
Physical Evidence
(facilities, equipment, uniforms, symbols, signage)
Process
(service design, standardization, customization, operational efficiency)
Promotion
(internal marketing, direct marketing, advertising, etc.)
Product
(total service product, core product, supplementary services, facilitating services, supporting services)
Price
(cost-based, demand-based, operations-based, competition-based, relationship-based)
Place
(access, location, delivery services, electronic delivery)
Which “P” will play a larger role in today’s tech environment?
The “Process” aspect.
For example. Marketing automation and the
Internet of Things –Opportunity Identification
Marketer Skills Needed:
a) Data analysis-leverage A.I.- predict & understand needs
b) Understand processes & flowcharting
c) Creating customer & experience journeys
Mindset for tactical success in digital marketing
1) Always be conversion focused
2) Key Metrics–know your breakeven points, Cost to Acquire a Customer and Customer Lifetime Value
3) Split A/B Testing
Comparison Typical vs ROP (Return on Promotion Campaign)
Generic (waste of money) vs Specific (less waste of money)
Steps for a ROP Campaign
- Know your current customer defection rate and share of wallet–maximize and/or improve this first.
- Don’t choose your target segment based on instinct–conduct market research and look at your data then select a highly responsive and profitable segment.
- Know both your Customer Acquisition Cost (CAC) and Customer Lifetime Value (CLTV) to determine how much you should be spending to acquire a new customer.
The Allowable and Return on Promotion determines…
…what conversion rate you need to break even on that promotion. Allowable somewhere b/w GP and NP Formula is: Total selling price-Cost of order
CAC
Cost of Customer Acquisition
(same as CPA or Cost per acquisition used mainly for PPC while CAC is across platforms)
How much should I spend (be it with ads or other means to acquire a customer?)
1.) Calculate CAC=MCC/CA
MCC = Total marketing campaign costs RELATED to acquisition (not retention) CA = Total customers acquired
CLTV
Customer Lifetime Value
Many different ways to calculate. Some formulas are simple while some don’t look at key variables.
CLV=Margin($) x (Retention Rate(%)/([1+discount rate(%)]-Retention Rate(%))
At a minimum, make sure the CLTV model includes: average gross profit per customer (not revenue, could also use net profit), retention rate, present value over some long time (3+ years).
Snapshot customers who started Year 1 then follows them plus referrals.
Exclude delinquent or inactive customers.
CLV=Margin($) x (Retention Rate(%)/([1+discount rate(%)]-Retention Rate(%))
Ratio of CLTV to CAC
<1:1
1: 1
2: 1
3: 1
4: 1
Losing money at an accelerated rate
Losing money/maybe breakeven
Marginal returns
Optimal level
Could be underinvesting in promotions, might need to be more aggressive and be closer to 3:1 so your acquiring as many customers for healthy growth.
Google LTV
Google Analytics has an LTV (lifetime value) Feature so you can see which channels contribute more profitable customers.
PPC w/out web analytics conversion data =
you not getting the best ROI and might be a waste of money
Need to know CLTV to CAC
What is an average order size? $50 or $500?
What is my repeat business rate over the years?
What is my average profit? 10% or 40%?
Referral-Viral Coefficient
Viral Coefficient is a referral metric on steroids-but difficult to achieve.
It is the number of referred customers divided by the original customers
Used by SaaS (software as service), gaming and mobile app marketers
Coefficient must be greater than 1 for “viral growth”
Pros and Cons of Viral Coefficient
Pros: If little to no marketing budget and in one of these businesses–very important
Cons: Not changed by promotional efforts; driven mainly by products/service and how satisfied customers are.
How to find number of referred customers
of customers: 100
Avg # of invites/shares/referrals each customer sent out: 10
Referrals are customers*avg # of invites: 1,000
Net Promoter Score
Predictor of Referral Health
0-6: Detractors
7-8: Passives
9-10: Promoters
Net Promoter Score = %Promoters - %Detractor
NPS is an absolute number, not a %.
E.g. if 25% promoters and 15% detractors score is +10
How to design an Experiment-Split Test
Have a control and test A/B or multi-variate (need a lot of traffic for this type of test)
Randomization
Statistically-valid sample size
Tips when testing the offer (in split testing)
A/B Testing: test different versions of a web page, splitting the traffic evenly between both pages.
Multivariate: Test different elements within a single web page. Ideal for high-traffic websites. The marketer needs to be experienced and also have a large volume of data/traffic when different types.
- Test only one feature at a time.
- Code your tests so you can measure results (e.g. urls, special toll free number, match codes, different coupon codes
What do digital marketers test?
- Usability (webpage version A vs B)
- Market Segments (income levels)
- Product A vs B (Adidas vs Nike)
- Form submissions
- Call to actions (“act now” vs “3 days left…”)
- Offer A (10% off) vs Offer B (buy one get one free)
- Compare media vehicles (SEO vs PPC vs Twitter vs Facebook)
- Timing (weekday vs weekend, morning vs evening)
- Seasonality (summer vs fall)
10. Look for insights in web analytics data–then test your hypothesis.
