Lehigh (10 13)

$X.99 per Month An introduction to the method and madness that underlies subscription pricing

Presented by: Brent Chudoba, SurveyMonkey Lehigh University: Principles of Marketing – MKT 111 OCTOBER 29, 2013

2

•  VP, GM of SurveyMonkey Audience • @SurveyMonkey since Apr. ’09 •  Previously:

-  Private Equity: Spectrum Equity Investors -  Investment Banking: Piper Jaffray

@bchudoba brentchudoba.com in/brentchudoba

Hi, I’m Brent

Presentation goals

3

Goals

4

•  Introduce several common pricing examples, discuss embedded marketing themes

• Discuss LTV (Lifetime Value), how it's used in practice and how tightly it connects to pricing

• Show how price points and and subscription terms interact with marketing and business goals

Scope

5

•  The content is focused on B2C (business to consumer) and B2B (business to business) subscription models -  Dealing largely with lower priced offerings

• Out of scope: lots of things -  International, licensed models, high value contracts,

services, auction pricing, price testing inventory constraints…

-  We are only scratching the surface of pricing considerations in the following material

DIRECTV What can you infer about DIRECTV’s marketing goals from the following page?

6

7

8

Packages 1 and 5 may be used as bookends, guiding people to packages 2,3,4

Avg. customer price probably around $70, avg. lifetime at least 2 years (LTV >=$1,700); CAC tolerance of $200 or more for a new paid customer

Disclaimer: Comments and estimates and are to be used for example/discussion purposes only. Estimations may be inaccurate.

This is where they want you, middle package, most popular

Spotify What can you infer about Spotify’s marketing goals from the following page?

9

10

11

They are really trying to drive to this package w/ a badge and a different color CTA

Using images to show the major differences, which is device portability, extremely helpful cue

This is a freemium product, if I downgrade I revert to free

SurveyMonkey What can you infer about SurveyMonkey’s marketing goals from the following page?

12

13

14

They want me here, “most popular” badge and “Gold” naming convention implies best of breed with minimal price increase

These are annual packages, but displayed with per month pricing, with one option for monthly billing

Marketing metrics

15

WTA?!? What the acronym?

16

I overhear conversations like this every day… Marketing: “This test failed, the LTV impact was too negative despite a higher AOV.“

Exec: “But first period churn may be better given the source, and with the higher AOV and ARPU, the LTR is going to look better.” Marketing: “The main AOV impact was related to higher conversion because of the source mix, but that also means CAC is higher for these users so LTV will be lower.”

Exec: “Ah, gotcha. Makes total sense, let’s keep testing.”

17

But the acronyms and jargon are pretty important

•  There are about a dozen inputs that go into many of the spend calculations for marketers

•  Most businesses rely heavily on core metrics that allow them to gauge performance

•  Advice: when you hear a metric you don’t understand, raise your hand, figure out what it means…

LTV (Lifetime Value) A metric that matters more than most for a subscription business

18

LTV: Lifetime Value

19

AKA: CLV (Customer Lifetime Value) AKA: The godfather of all metrics • Definition(1): a prediction of the net profit

attributed to the entire future relationship with a customer

• Formula:

(1) Source: http://en.wikipedia.org/wiki/Customer_lifetime_value

Calculating LTV

20

Let’s start with with a simpler LTV calculation: •  We are going to ignore discounting (time value of $), and

assume costs to acquire and supporting customers are negligible

LTV = Monthly ARPU * Average Customer Lifetime (in months) •  ARPU = Average Revenue Per User (per month)

How much do you pay me per

month?

How long will you keep paying me?

Monthly ARPU

21

22

23

24

Monthly ARPU

25

How much do customers pay every month? • Netflix: $7.99 • Spotify: either $9.99 or $4.99 • SurveyMonkey: $17 or $25 or $65 For products with different package pricing or term discounts (annual vs. monthly pricing), you basically need to find the weighted average, or calculate each package type separately

Average customer lifetime

26

Average customer lifetime

27

How long do customers stick around? Method 1: Half life •  Simple, but can grossly underestimate LTV Method 2: Sum of monthly retention % estimates •  Accurate, but requires healthy historical data and/or good

assumptions These are just 2 of myriad ways to find avg. lifetime

Half life

28

•  Assuming stable churn rates (e.g., 5% per month), the average life time of a customer is when retention reaches 50% and half of customers have churned

•  In the example below, right after month 13

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

100%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Ret

entio

n R

ate

Month * The data used above is used for example/discussion purposes only.

Sum of retention

29

•  The sum of the area under a retention curve can estimate average retention •  Useful with irregular churn patterns (e.g., high churn in early months) •  In the example below, the sum of the first 36 months of retention percentages

is 13.4 months

0% 10% 20% 30% 40% 50% 60% 70% 80%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Ret

entio

n R

ate

Month * The data used above is used for example/discussion purposes only.

LTV Calculation

30

LTV = Monthly ARPU * Average Customer Lifetime (in months) Using examples from the previous sections: •  If ARPU = $10/month •  And Average Customer Lifetime = ~13 months

Then, LTV = $130 per new customer

Question: Why does LTV matter to a marketer?

31

32

It often governs how a marketer spends their time and $ •  If the marketing spends $140 on paid acquisition

channels (e.g., AdWords) to acquire a customer that is only going to generate $130 (using our prior example) in their lifetime, they will run out of money to spend very quickly, if they still have a job…

LTV can govern…

33

• How much businesses are willing to spend to acquire a new customer

• Whether a business model can support having sales people

• What marketing channels are available (e.g., online, social, TV, radio)

• What areas in the customer lifecycle are the most important (acquisition, retention, engagement)

So where does pricing fit in to all of this?

34

35

•  Price is the 1 variable in the LTV equation that marketers and businesses have 100% control over

•  So its one of the most important levers that marketing can use to achieve business goals -  Initial price setting -  Price testing -  Packaging/bundling -  Discounting -  Upselling

36

So if a LTV governs marketing and sales possibilities… And pricing is a key component of LTV… And higher prices correlate to higher LTVs… Why not just raise prices to the maximum that people would pay?

Price often impacts conversion

37

•  Price impacts conversion rates which impact revenue •  Marketing needs to know what it is optimizing for: conversion or

revenue, or something else

0%

2%

4%

6%

8%

10%

12%

14%

16%

$0

$10

$20

$30

$40

$50

$60

$70

$1 $2 $3 $4 $5 $6 $7 $8 $9 $10 $11 $12 $13 $14 $15 $16 $17 $18 $19 $20

Con

vers

ion

Rat

e

Rev

enue

per

100

Vis

itors

Pricing

Revenue & Conversion Rate @ Various Prices

Revenue per 100 Visitors Conversion Rate

* The data used above is used for example/discussion purposes only.

Quick pricing detour Let’s look at how pricing changes can interplay w/ conversion and LTV

38

39

Question: How many people have access to a Netflix subscription?

We surveyed 514 people and asked: Do you currently have access to an active Netflix subscription? •  Yes •  No •  I don’t know


40

Question: How much does Netflix cost?

Survey Question: How much does Netflix cost per month? •  $4.99 •  $7.99 •  $9.99 •  $14.99 •  I don’t know * The data used above is used for example/discussion purposes only. N=246

41

42

Question: What would you do if Netflix raised its price to $9.99?

Survey Question (for those who answered $7.99 to the initial pricing question): What would you do if Netflix raised its price to $9.99 per month? •  Nothing •  Cancel my subscription •  Move to a competitor •  I don’t know •  Other (please specify)

* The data used above is used for example/discussion purposes only. N=144

43

Question: What would you do if Netflix raised its price to $14.99?

Survey Question (for those who answered $7.99 to the initial pricing question): What would you do if Netflix raised its price to $14.99 per month? •  Nothing •  Cancel my subscription •  Move to a competitor •  I don’t know •  Other (please specify)

* The data used above is used for example/discussion purposes only. N=144

Netflix pricing considerations

44

• On 10/21/13, Netflix reported that it added 1.3M net subscribers additions (new minus churned)

• Let’s assume it added 2M new subscribers and 700k cancelled, so 1.3M net additions

• And let’s use some estimates for our LTV variables -  ARPU = $8/month -  Average customer lifetime (months): 24


LTV using current pricing

45

• New customers: 2M • ARPU = $8/month • Average customer lifetime (months): 24 • LTV per user: $192 • Overall LTV = $384,000,000


LTV using $9.99 pricing

46

Assumptions: •  Netflix does not increase pricing for existing customers (churn stays the

same) •  The % of people who said they would cancel are a proxy for the % drop in

new user conversion •  The conversion drop has no impact on avg. customer lifetime Updated metrics: •  New customers: 1.3M (2M * (100%-35%)) •  ARPU = $10/month •  Average customer lifetime (months): 24 •  LTV per user: $240 (+25%) •  Overall LTV = $309,984,000 (-19%)


LTV using $14.99 pricing

47

Assumptions: •  Netflix does not increase pricing for existing customers (churn stays the

same) •  The % of people who said they would cancel are a proxy for the % drop in

new user conversion •  The conversion drop has an equal impact on lifetime Updated metrics: •  New customers: 292k (2M * (100%-85%)) •  ARPU = $15/month •  Average customer lifetime (months): 24 •  LTV per user: $360 (+88%) •  Overall LTV = $105,048,000 (-73%)


Scenario Review

48

•  $7.99: New customers: 2M, New customer LTV: $384M •  $9.99: New customers: 1.M, New customer LTV: $310M •  $14.99: New customers: 1.5M, New customer LTV: $105M

•  My hypothesis before running the experiment was that LTV may actually be higher for the on of the higher priced scenarios (it wasn’t, which makes for a less difficult conclusion)

•  With this example we see that individual user LTV must be combined with the impact of new user conversion

•  And even if the higher price scenarios had provided higher overall LTV, we must answer questions like: -  How do we feel about total subscriber count and overall user growth? -  How do we factor in viral marketing from more happy customers? -  How important is market share and competitors picking up more customers?


Business goals help optimize pricing strategy

49

What can pricing be optimized for?

50

May optimize for: •  Conversion

•  Cash

•  LTV (long term revenue)

•  Profit

•  Supply/Inventory mix

Business scenario:

•  Boostrapped startup

•  VC backed startup

•  Established business

•  Unknown CLV

•  Physical product retailer

Price point implications

51

Subscription term implications

52

What’s your LTV? A peek into the complexity of a multi-product, multi-service calculation

53

LTV can get really, really complex

54

Does anyone have an iPhone?

How would Apple estimate your LTV? How valuable did you become to Apple when you bought an iPhone?

After your iPhone purchase, you became: •  Much more likely to spend money on iTunes •  Much more likely to buy a Mac, which makes you

-  Much more likely to spend more money on iTunes -  Much more likely to upgrade to the latest iPhone device within 1 year of

launch

•  More likely to buy an iPad •  Much more likely to upgrade to new versions of all these products…

55

Thank you!

Appendix

56

Survey respondent sourcing

57

For the consumer survey about Netflix pricing in this presentation, we used SurveyMonkey Audience survey respondents •  We surveyed 514 respondents on October 27-28, 2013

-  Screening/skip logic was used, so not all respondents answered all questions

•  This survey was conducted for example and discussion purposes •  For more information on SurveyMonkey Audience

-  SurveyMonkey Audience Overview video -  SurveyMonkey Audience Help Center

Lehigh (10 13)

Business

ltv lifetime value

month pricing

monthly pricing

ltv method

subscription pricing

ltv impact

customer price

years ltv