Pure CPQ: Fast, Flexible Quoting for Data Storage Solutions
Company
Pure Storage, Inc.
Design Team
Yue Wu, Jing Han
Timeline
June 2024 - Present
Organization
Digital Transformation Group
Product Management
Grace Fang, Dave Sloan, Ruchira Panda
Engineering
DTG Engineering Team
Project Sponsor
Mercury Core Team
Copyright / Patent
Pure Storage, Inc.
*Disclaimer: Information in the images does not represent real data.
Problem
Quoting data storage is slow, manual, and opaque.
Customers
Lack confidence that quotes reflect their actual needs, due to limited visibility and technical complexity.
Sellers
Rely on a manual, time-consuming process prone to errors in tailoring solutions, configuring details and creating quotes.
Industry Environment
Quoting data storage solutions requires deep technical expertise, creating higher barrier and little transparency for buyers.
Solution
Automation for all ◆, precision control for experts ❖.
1
Optimize Assets
◆ Simply click on recommendations to modernize your assets.
❖ Trade-in appliances, simulate capacity and performance.
2
Find the Best Fit
◆ Get accurate product recommendation by only 5 clicks.
❖ Control sizing details to find the most suitable products.
3
Configure Products
◆ Ready to go with the recommended configuration.
❖ Fine-tune product details through a guided, error-resistant flow.
4
Add Services
◆ Get recommended services with one click.
❖ Tailor services to specific needs of the customer.
5
Auto Complete Quote
◆ See pricing information immediately, and know the estimated approval time.
❖ Customize complex offerings including discounts and promos.
Impact
Slow, Manual, Opaque → Fast, Automated, Accessible
Customers: 14 Days → 1 Day
Customer waiting time is reduced. They gain confidence knowing their solution is based on clear, guided inputs.
Sellers: 40 Tools → 1 Tool
Sellers now quote data storage solutions through a single, guided tool, replacing the fragmented tools with one streamlined flow.
Industry Transformation: Expert Tribal Knowledge → Standardized Process
We’ve translated deep domain expertise into a scalable, standardized system, while still allow experts to flex and fine-tune when needed.
Feature Highlights
01
Find the Right Fit in Just 5 Clicks
Start with 4 simple inputs—storage type, performance tier, workload type, and capacity—and get recommendations that meet customer needs in 80% of the use cases.
02
Precision Planning for Complex Needs
Advanced users can simulate growth, tweak DRR assumptions, and customize workloads to match precise business demands—backed by visual tools for capacity planning.
03
Manage Fleets Intelligently
Users can take action with a click—upgrading products, renewing end-of-life assets, trading in for credits, or purchasing through available promotions.
04
Optimize Assets with Confidence
Expert users can select existing appliances and combine old and new workloads into optimized solutions—saving energy, reducing cost, and previewing capacity and performance visually.
05
Configure Products with Clarity and Control
Pre-populated defaults make setup easy, while visual feedback allows users to adjust product details and instantly see changes in specs and pricing.
06
One-Click Quote, Ready to Go
Once configured, the quote is instantly generated. Sellers receive estimated approval info, standardizing the process and reducing delays.
80s Video
Challenge
How do we translate complex, domain-specific knowledge into a fast and flexible quoting experience?
Limited understanding of user expectations for an ideal CPQ experience in enterprise context.
Initially the quoting process and tools seemed too complex to streamline without losing accuracy.
Legacy sales models in the data storage industry and diverse stakeholder opinions on how to modernize them.
Explore the Flow
Understand user expectations: one tool, three perspectives
Customers
Roles: Company/Organization Admin
Goals: Customers seek transparency into how the solution maps to their needs. They want to explore and customize options to manage budgets—without requiring technical knowledge.
Internal Sellers
Roles: Account Executive, Sales Engineer, Sales Account Manager, etc.
Goals: Internal sellers need end-to-end control over sizing, configuration, and pricing—while minimizing quoting time and reducing manual effort.
Partner Sellers
Roles: Account Executive, Sales Engineer
Goals: Partner sellers want to quote independently without always involving technical experts. They need to present solutions clearly and confidently to customers.
The quoting journey is not linear, but it involves 6 key steps. For internal sellers, partner sellers, and customers alike, the most time-consuming and manual parts are calculating needs (sizing), searching for the right products, and configuring them based on specific use cases.
Focus on calculating needs and configuring products: which flow is better?
If customer wants to buy a new product, here is the logic behind:
We explored 3 different flows to translate customer needs into product configuration:
Flow A was initially proposed by product experts. It follows a clear, task-oriented structure, assuming users already know what decisions to make at each step. However, this approach demands significant knowledge of Pure’s product portfolio. Customers would struggle to confidently select the right options without expert-level understanding.
Flow B simplified the entry point by introducing a few guided questions to drive recommendations, reducing the number of choices users had to make directly. However, it still required baseline knowledge of Pure’s offerings. It also limited exploration: users couldn’t easily see how different answers impacted results. In addition, configuration later became tedious, as users were prompted to fill in many additional fields after the initial questions.
Flow C introduced real-time exploration. Users could adjust simple inputs and immediately see how recommendations evolved. This minimized the knowledge barrier while giving users more transparency and control. Based on their inputs, default configurations were automatically derived, requiring minimal manual adjustment unless users wished to fine-tune further.
We selected Flow C because it created a more intuitive and confidence-building experience. However, a new challenge emerged: Was Flow C even feasible? Could such complex sizing and configuration parameters be streamlined without compromising accuracy? This critical question set the direction for the next phase of collaboration between design, product, and engineering teams.
Key insight: minimal inputs can still make the most magic
16
User interviews for generative research
10+
Participatory design workshops with experts, internal sellers and partners
12
Concept testing with technical users and other users
Was Flow C feasible? During a participatory design workshop, a breakthrough emerged: by building a product recommender, we could cover 80% of user needs with just four simple inputs—storage type, performance tier, workload type, and total capacity—things customers already know. With those inputs, we could narrow down to a reasonable set of options and present recommended configurations to minimize hardware, to optimize performance, or to align with common use cases.
Earn User Confidence
How do we build trust in the recommendation results?
Once we proved that Flow C was technically feasible, a new challenge emerged: How could we help users trust that the recommended solutions were truly tailored to their inputs?
Balance Ease/Control
Consider another use case: how do we help users optimize assets with less manual effort?
When customers optimize their fleets, they often consolidate multiple existing appliances into a new hardware product or service. While new customers typically start by purchasing new products directly, existing customers usually look for solutions that build on their current fleet. This use case called for a different user flow.
Focus on analyzing usage and simulating
To minimize user effort, we designed smart recommendations that provide actionable insights based on existing fleet data. Users can easily take actions like renewing subscriptions, optimizing utilization, or applying promotions—by simply clicking on the recommended insights.
However, real-world customer behavior revealed a gap: smart recommendations alone could not address all needs. Many customers wanted the flexibility to selectively choose specific appliances for optimization—something internal systems could not predict. To support this, we needed to design a sub-flow that allowed users to manually analyze usage and simulate new configurations.
How do we balance simplicity and control to support decision-making?
Through user research, we decoded the typical decision-making process into three clear steps:
Selecting multiple existing appliances for consolidation.
Adding any additional workloads to the optimization plan.
Simulating the combined usage on a new destination appliance.
This flow preserves a guided experience while offering flexibility for users who need more control—helping them make informed optimization decisions without losing clarity or ease of use.
Iterate on Interactions
Build an intuitive configurator through continuous user testing
The project unfolded alongside shifting business priorities, evolving tool integrations, and internal system constraints—demanding agility and constant alignment. We began with storyboards to align cross-functional teams, then built a proof of concept prior to the MVP to validate core ideas through user testing.
User testing consistently revealed new behavioral patterns and needs. For example, we learned that users change the "site" structure of products more frequently than expected. In our system, a site groups products based on installation location. Two overlooked use cases surfaced:
Reassigning products to different sites.
Duplicating products with slight changes and placing them in another site.
To address this, we explored three design options and tested them with users:
Option 1: Dropdown Field on the Card
Pros: Matches familiar user behavior—selecting saved sites from a dropdown, similar to CRM patterns.
Cons: Becomes overwhelming when managing more than 4 cards or 2+ sites; difficult to track mappings visually.
Option 2: Drag and Drop Cards
Pros: Intuitive and visual. Users can duplicate and move products between sites easily, and adjust product order within each site.
Cons: In a long list, users may need to collapse intermediate sites (e.g., dragging from Site 1 to Site 3 past Site 2).
Option 3: Dropdown Field in Main Content
Pros: Keeps card UI clean and focused; inputs are centralized in the main section, reducing distractions.
Cons: Users must click into each product to change its site—tedious and repetitive where there is a long product queue.
Based on user feedback, we moved forward with Option 2: Drag and Drop Cards, which best balanced clarity, flexibility, and ease of use for common workflows.
Learn More
Pure Storage: The Unified Enterprise Data Platform
We store, manage, and protect the world’s data. Visit our website: purestorage.com
Acknowledgement
Design: Yue Wu, Jing Han
Product Management: Grace Fang, Dave Sloan, Ruchira Panda, Shobhit Sharma
Project Management: Stacy Wen
Engineering: Ramu Muppavarapu, Aditya Sonam, Allamaprabhu Munjan, Krishna Guduru, Ayan Pan, Aishwarya Krushna Chaudhary, Gnana Varadaraju, Priyanshu Patel, Raghavendra Patamata, Ayush Kumar, Madhusudhanan Varadarajan, Ayush Bhargava, Britney Harrison, Gaurav Srivastava, Ikjae Park
Business: Vince De Paul, Remko Deenik, Sri Chittineni, Steve Gordon, Jonathan Tan, Angad Narang, Vipul Bhatnagar, Kristen Caskey