We have reached an inflection point in the Sales Performance Management (SPM) space.
Despite supporting some of the world’s largest companies, legacy SPM vendors have struggled to innovate and meet the expectations of modern Enterprises due to technical debt and legacy architecture. Over the last decade, for example, as businesses increasingly need to act with agility and use real-time insights to quickly respond to changes in market conditions, real-time modeling has proved an essential need. The inability of legacy solutions to address these emerging challenges has made way for modern cloud-native commission platforms to blaze forward and redefine scale and performance expectations.
In Forrester’s recent report, “The Forrester Wave™: Sales Performance Management Platforms, Q1 2023”, they address this shift directly:
Legacy processing architecture is being replaced by new structures that greatly increase performance by processing in memory and in real time, limiting the need to consider order and sequence of imports. Although all vendors are progressing to these capabilities, they are in different phases of the journey. Some are in transition, while others started with these capabilities as their foundation.
At CaptivateIQ, we are proud to be at the forefront of innovation with our proprietary SmartGrid™ calculation engine, which offers our customers ultimate flexibility in calculating commissions how they need, when they need - pairing real-time modeling capabilities with the power to batch process large-scale data sets at ultra-fast speeds reaching millions of calculations per second.
As the last decade has shown, the business environment is increasingly dynamic, necessitating technology like CaptivateIQ’s state-of-the-art SmartGrid calculation engine to help Enterprises remain agile while accurately calculating complex commissions at scale.
How our commission calculation engine works
Our calculation engine operates on data in a 2D grid format (rows and columns). It provides formulas that can operate on any data across any grids in the model, as well as transformations to generate new grids by performing operations across any number of grids in the model.
Even though this part sounds simple, it is one of the most complex pieces of technology we have built. The performance, accuracy, and scale requirements are paramount, with trillions of calculations happening per month. Powering the type of calculations required for many of the models built in CaptivateIQ in real-time requires a significantly optimized engine.
How do we do it? Let’s talk through a few key innovations that allow us to unlock both real-time modeling and efficient batch processing for even the largest enterprise organizations:
Setting the Foundation
Let's start with the basics of CaptivateIQ. In contrast to the cell you see in spreadsheets, everything that SmartGrid does is based on columns - specifically data columns and derived columns:
- Data Columns: Hold raw data that can be entered manually or populated automatically from our data connectors (Salesforce, Snowflake, and many more)
- Derived Columns: Built off of data columns and use our extensive formula library to create complex operations, and can reference other columns
Columns are grouped together into worksheets, which are open grids for you to view your data, edit your formulas, and see your real-time results. When you want to combine worksheets together, you can use our no-code transformations to perform SQL-like operations on your worksheets, such as joining two worksheets together or stacking them in a union. With CaptivateIQ, you have ultimate flexibility to manipulate your data and create the models you need.
CaptivateIQ’s focus on columns instead of cells offers several major benefits:
- Reusability: When users operate on entire columns, they’re solving for all the rows in their data instead of specific ones, making your commission plans more reusable and more reliable
- Performance: When we store items in a columnar format, it means we can perform super fast vector-based operations on them, greatly improving SmartGrid’s performance when working with large datasets
Graph Clustering for Calculation Efficiency
SmartGrid takes worksheets, columns, and transformations and turns them into payout calculations for our customers' payees through the use of a graph. As you build out your commission plans in SmartGrid, the platform builds a graph representation of all the dependencies. In the most basic form of the graph, columns serve as nodes, and the edges represent references between the columns. When you open a worksheet or view a payout, SmartGrid uses our in-house formal grammar to parse each of the formulas for every node in the graph, and then evaluates them in order.
To ensure the evaluation is maximally efficient, SmartGrid intelligently groups nodes together through a clustering algorithm. As we build up our evaluation graph, we cluster columns together if they are independent of each other, since they can be evaluated in parallel. Thus, the bigger the cluster, the faster SmartGrid can run the evaluation. At CaptivateIQ we’ve put a lot of effort into optimizing how SmartGrid’s graph clustering algorithm works – the better our clustering is, the more efficient our computation can be, and the faster we can calculate results for our customers.
Smart Caching System for Optimized Performance
SmartGrid leverages caching to optimize performance. By checking if each node has been calculated before with the current data, SmartGrid can pull the cached node and insert it into the dependency graph, saving time on repeated calculations. To handle cache invalidation, SmartGrid updates its cache keys every time a node changes, relying on the cache's own TTL (time to live) for eviction.
This caching layer enables SmartGrid to work with larger datasets at peak speed, since we’re intelligently only recalculating what has changed. This sets up comp admins for success, as they can continue their work and see the impact in real time.
Automated Testing for Peace of Mind
Though we strive to provide best-in-class performance with SmartGrid, accuracy is still paramount when it comes to commissions. To provide confidence in SmartGrid’s accuracy, we’ve built automated testing into every part of the development process. For every change we make in SmartGrid’s core calculation engine, we run over 9 billion calculations and compare results to help ensure accuracy. Any deviations are detected and alerts will be raised for the team to dig in and follow up.
Decoupled Modeling Experience for Ease of Use
This calculation engine is decoupled from the UX and model building that happens on top of it. This has allowed us to develop a user experience that is very familiar and simple yet can perform complex calculations at scale.
Additionally, we’re able to optimize the underlying engine independently without worrying about the impact on the modeling experience, resulting in rapid iteration under the hood, and providing numerous improvements over time without disrupting our customers!
The modern "enterprise-grade solution" for calculating commission
We are delighted to have received the highest score possible in the calculation processing criterion in the most recent Forrester Wave report and to be recognized by Forrester as having an execution roadmap that “delivered an enterprise-grade solution” with “a complete incentive compensation product that is robust enough to support the most complex requirements.” We are proud that our state-of-the-art SmartGrid calculation engine enables large Enterprises to remain nimble while still processing complex commissions at scale.
While some legacy vendors will face a long road in re-architecting their calculation engines, Captivate’s customers are already reaping the performance benefits. With SmartGrid, not only are comp admins able to see changes and their impact in real-time while designing plans, but they can also run payout processing jobs that take just a few minutes for customers with thousands of payees and millions of transactions. This way comp admins aren’t stuck waiting around till the next day to review and publish their payouts, and can move with much greater velocity through their regular commissions process.
Come talk to us about enterprise sales commission
Curious to learn more about Captivate’s proprietary SmartGrid™ calculation engine, and how we’re building the commission solution fit for modern Enterprises? Take a call with someone from our team to learn more about how CaptivateIQ can help get commissions right for your business.
CaptivateIQ’s product makes the hard parts of compensation easy while providing a flexible, intuitive interface that works like Excel but with powerful data management capabilities and compensation-specific calculations.
(Source: The Forrester Wave™:Sales Performance Management Platforms, Q1 2023)
Download the complete Forrester report “The Forrester Wave™: Sales Performance Management Platforms, Q1 2023” to learn more about sales performance management and why Forrester names CaptivateIQ as “one of the best options for time to value.”