Pricing is at the core of how companies sell, support, and talk about their products. Yet most companies fail to architect a well thought out customer journey and decision-making process. In this session, Hazjier Pourkhalkhali, Global Director, Strategy & Value at Optimizely shares how the company changed its ways two years ago and saw a dramatic reduction in churn rates while improving win rates, deal sizes, and expansions. All due to a customer-first approach to pricing.
5. We’re Proud to Work With Great Global Enterprises
We work with 24 of the Fortune 100
6. 6
20182017
Our Pricing journey has taken two years
Q2 Q3 Q4Q1Q2 Q3 Q4Q1
Strategy
Implementation
BETA New Pricing & Packaging
Data Integrity, Tooling and Support
Enablement and Refinement
7. “People’s attitudes and feelings about
losses and gains are really not
symmetric. We really feel more pain
when we lose $10,000 than we feel
pleasure when we get $10,000.”
D A N I E L K A H N E M A N
N O B E L - P R I Z E 2 0 0 2
10. Pre-order encourages additional
research, distracts from purchase
Only valuable if people believe
they will purchase another product
Without a follow-up purchase in
mind, pre-order feels like losing
$20
12. Problem Solution Result
Identify the user’s
pain points
Validate pain points
through research
Prioritize the most
urgent pain points
Identify 10 solutions
Problem solve on the
most critical to test
Test multiple variations
Decide the primary metrics
Choose secondary metrics
Select monitoring goals
Problem-First Hypotheses
13. We conducted substantial research to inform our customer pain
points
Step 02
Financial
Modeling
External
Interviews
Internal
Interviews
Data Research
Academic Research
Client Benchmark 40 Customers
20 Prospects
10 Experts
Valuable segments
Churn regressions
Market analyses
Maturity models 2 Surveys
50 Interviews
Technical Problems
14. 14
1. Pricing grew too complex for customers and employees
2. Separate volumes caused friction and misaligned incentives
3. Ineffective packages led to churn
4. Feature selling hurt executive access
Under our our historical pricing, we faced several key problems
Excerpts
1
2
3
4
15. 15
1.1 Optimizely sold over 400 distinct product SKUs
Customers forgot feature
access and benefits
Account teams had to
research access per customer
Customers did not
understand our product
portfolio
Confusion around
products and
features
Excess feature gating hurt
market perception
Feature gating weakened
position in RFPs
Hurt best-in-class
messaging
Sales could not keep up with
SKUs
Marketing could not maintain
collateral
Product Managers would
ignore older plans
Difficult to support
SKU’s
16. 16
1.2 New Pricing simplified our model
Tiers
ProductPackages
Experimentation
Workflow
Experimentation
Experimentation
Feature Flags
Workflow
Enterprise Reporting
Experimentation
Personalization
Workflow
Enterprise Reporting
Full Stack
All Devices
and SDKs
Experimentation
Feature Flags
Workflow
Web
Desktop and
Mobile Web
Essentials EnterpriseBusiness
17. 17
2.1 Under our old pricing model, every product had separate volumes
Web Recommendations
Web Personalization
Full Stack Experimentation
Site traffic in Sessions / Visitors
Bands of MUVs
Web Experimentation
Site traffic in Sessions
3
2
2
Product Grouping Tiers Pricing Volume
IOT/OTT Experimentation
Mobile Experimentation
Exact MUVs or Exact Annual MUVs3
Exact MUVs
Exact MUVs
1
2
18. 18
2.2 Separate volumes led to misaligned incentives
Scenario
Year 1 Contract:
Two-product customer
Web Experimentation
2 Million MUV’s
Mobile
Experimentation:
2 Million MUV’s
Year 2 Contract:
Churn and expansion
Web Experimentation
1 Million MUV’s
Mobile
Experimentation:
3 Million MUV’s (+1M)
Net Impact
CSM Upset
Hurts compensation
AE Happy
Gains compensation
Many Hours Wasted
Not net value generated
19. 19
2.3 Universal impressions give customers flexibility and allow
account teams to refocus on the important conversations
Team can be agile. Teams can shift product usage as
needed without needing any commercial discussions
Volumes are sold in broad bands, which reduces pressure
to forecast exact consumption and gives more ease of mind
Expansions only needed for large increases. Only
substantial increases lead to contract changes, while smaller
increases can be handled on demand
20. 20
3.1 Our internal benchmarks showed that low volumes made
success difficult to achieve
NOTE: Winning is defined as reaching a statistically significant uplift on any metric. In our analysis, 98% of experiment metrics are seeking an uplift, and
not a reduction. Sample size is 103,000 experiments with at least 2 variations, at least 1,000 visitors per variation, and at least one non-baseline variation
Visitors per
variation
Chance to detect statistically significant uplift
13
18
22
24
26
25
21
1K -10K
10K - 50K
50K - 100K
100K - 200K
200K - 500K
500K - 1M
Over 1M
21. 21
3.2 An analysis of partial churns revealed important problems
Smallest
Plan
Product
Volume
Churn risk
Willingness to pay
22. 22
3.2 An analysis of partial churns revealed important problems
Smallest
Plan
Product
Volume
23. 23
3.2 An analysis of partial churns revealed important problems
Smallest
Plan
Product
Volume
25. 25
Smallest
Impressions
Package
Included with
all packages
3.2 Raising our minimum volume amount drastically improved
account health and team efficiency
Improves account health. Minimum volumes that
encourage healthy product adoption. Eliminates bad
plans
Refocuses underutilization talks from price to growth.
Account encouraged to maximize usage, not reduce price
Saves substantial hours of negotiations. Easier
estimation for incoming customers, less hours spent on
partial churn
Refocuses GTM to high growth-potential accounts.
Account teams freed for high-growth accounts, smaller
accounts protected in healthier state
26. 26
4.1 Under the MUV model, executives had to evaluate lower-level
features to determine price
NOTE: Analysis of 138 partial churns from FY17 on web experimentation products
27. 27
Tiers
ProductPackages
Experimentation
Workflow
Experimentation
Experimentation
Feature Flags
Workflow
Enterprise Reporting
Experimentation
Personalization
Workflow
Enterprise Reporting
Full Stack
All Devices
and SDKs
Experimentation
Feature Flags
Workflow
Web
Desktop and
Mobile Web
Essentials EnterpriseBusiness
4.2 New Pricing simplified our model
29. 29
C-Suite
Senior
Executives
Managers
Teams04
03
02
01
4.3 Our new pricing decisions require the engagement of senior
executives
1 Should we experiment across more
channels? How do we scale up?
0 Are we missing features with our
existing experimentation stack?
2 How do we drive digital growth and
competitiveness?
30. 30
Historical Pricing: Final 13 SKU’s
Web Recommendations
Web Personalization
Full Stack Experimentation
CPM x 1M Site Traffic
(Visitors)
1M – 2M Band, Monthly
Visitors
Web Experimentation
CPM x 2M Site Traffic
(Sessions)
Product Grouping Tiers Pricing Volume
IOT/OTT Experimentation
Mobile Experimentation
CPM x 10M Annual MUV’sGood
CPM x 100K Monthly Visitors
CPM x 500K Monthly Visitors
Better Best
Good Better
Good Better Best
Good Better
Good
Good Better
31. 31
Customer example: Many moving parts
Web Recommendations
Web Personalization
Full Stack Experimentation
CPM x 4M Site Session /
Month
1M – 2M Band, Visitors /
Month
Web Experimentation
CPM x 2M Site Traffic
(Sessions)
Product Grouping Tiers Pricing Volume
IOT/OTT Experimentation
Mobile Experimentation
CPM x 10M Monthly Visits /
YearGood
CPM x 100K Monthly Visitors
CPM x 500K Visitors / Month
Better Best
Good Better
Good Better Best
Good Better
Good
Good Better
32. 32
ProductPackages
Full Stack
All Devices
and SDKs
Web
Desktop
and Mobile Web
Tiers
Experimentation
Workflow
Experimentation
Experimentation
Feature Flags
Workflow
Enterprise
Reporting
Experimentation
Personalization
Workflow
Enterprise
Reporting
Experimentation
Feature Flags
Workflow
Essentials EnterpriseBusiness
Volume Bands
10 MILLION
FREE
1 BILLION
UNLIMITED
Annual Impressions
1 BILLION
Experimentation
Workflow
Experimentation
Feature Flags
Workflow
Enterprise
Reporting
New Pricing