Rod lift is the most widely deployed artificial lift method in the world. Depending on whose numbers you use, rod pumped wells account for between 60% and 80% of all artificially lifted wells globally. The engineering software market that serves these wells has been remarkably stable for decades - a small number of desktop tools, developed by a small number of companies, used by a relatively small community of specialist engineers.
That stability is starting to change. Not dramatically - this is not an industry prone to rapid disruption - but meaningfully. The changes are driven by three factors: the shift in well geometry from vertical to deviated, the reduction in production engineering headcount per well, and the emergence of cloud-based platforms that challenge the desktop-only model that has defined the market since the 1990s.
The established players
RODSTAR, now under ChampionX following the Dover acquisition of Theta Oilfield Services, remains the tool most rod lift engineers learned on. The simulation engine is validated across thousands of wells and the broader ChampionX ecosystem - XDIAG, XBAL, XROD, XSPOC - provides coverage from design through diagnostics and production optimization. XROD introduced AI-assisted optimization for vertical wells, which represents one of the first applications of machine learning to rod pump design. The limitation is architectural: RODSTAR is a Windows desktop application with file-based workflows and per-seat licensing that separates vertical and deviated capabilities.
SROD, from Lufkin (now NOV), carries nearly a century of rod lift manufacturing knowledge. The v9.2.0 release introduced a Multi-Case Generation Wizard that automates parametric design sweeps - a feature that addresses the time cost of manual iteration that production engineers have accepted as unavoidable for decades. SROD's equipment libraries are built around Lufkin hardware, which is an advantage for operators standardized on that equipment and a limitation for those who are not.
QRod from Echometer remains the free option. It pairs with Echometer's fluid level measurement hardware and provides basic rod pump design and performance prediction. It will not replace RODSTAR or SROD for complex deviated well design, but it removes the financial barrier for smaller operators and consulting engineers who need a design verification tool.
The architectural shift: desktop to cloud
The most significant structural change in the market is the emergence of cloud-native rod pump simulation. PetroBench, which launched its V2 platform in early 2026, runs entirely in the browser with no local installation. This is not a desktop application wrapped in a web interface - it is a platform built from the ground up for multi-user, multi-device access with centralized data storage, automatic versioning, and API-based integration.
The technical capabilities that differentiate the cloud approach are not primarily about the simulation engine - the wave equation is the wave equation regardless of where it runs. The differences are in the workflow infrastructure: cubic spline interpolation with configurable step length, multi-scenario comparison views, automatic design versioning, structured well management, integrated reporting, and multi-language support with global unit systems.
Whether this architectural difference matters depends on how your organization works. A single engineer in one office running simulations on mostly vertical wells has little reason to change from a proven desktop tool. A team of engineers spread across multiple basins, working on deviated wells, collaborating on designs, and needing access from the field - that team feels the limitations of desktop architecture daily.
The well geometry problem
The rod lift software market developed during an era when most rod-pumped wells were vertical or near-vertical. The simulation tools, the design methods, the rules of thumb, and the API design curves were all validated primarily against vertical well data. The Gibbs wave equation (1963) and the API RP 11L design method accommodate deviated wells, but the tooling and workflows around them were optimized for the simpler case.
The well population has shifted. In the Permian Basin, the Eagle Ford, the Bakken, and most other active producing regions, new rod-pumped wells are predominantly deviated or horizontal. Wells with 40 to 70 degree maximum inclination, multiple doglegs, and measured depths of 10,000 to 15,000 feet are now common rather than exceptional. The rod-tubing interaction in these wells - side loads, contact forces, bending stresses, friction effects - is fundamentally more complex than in vertical wells.
This shift has exposed limitations in simulation tools that were adequate for vertical wells. Linear interpolation at 50-ft step lengths produces reasonable stress predictions in a 5,000-ft vertical well. In a 12,000-ft deviated well with a 6-degree-per-100-ft build section, the same approach can miss stress concentrations that determine where and when rods fail. The tools are not wrong - they are being applied to a class of problems they were not originally designed to solve at this level of complexity.
The workforce equation
The number of production engineers per well has declined steadily over the past decade. Operators that had one engineer per 50 wells in 2015 now have one engineer per 150 to 200 wells. Some of this reflects genuine efficiency improvements - better surveillance, exception-based monitoring, automation. Some of it reflects cost reduction that has pushed more work onto fewer people.
For rod string design specifically, this means the engineer has less time per well for simulation setup, iteration, and analysis. A design workflow that takes 45 minutes per comparison cycle is more problematic when you have 200 wells competing for your attention than when you have 50. The overhead of file management, manual data entry, and report formatting - time that produces no engineering value - becomes a larger fraction of the available design time.
This workforce pressure creates demand for tools that reduce the non-engineering overhead of simulation work. Automated comparison, template-based design, integrated reporting, and centralized well management are not luxury features in this context - they are responses to the reality that production engineers do not have time for workflows that were designed when well-to-engineer ratios were a quarter of what they are today.
Remaining gaps in the market
Several capabilities that production engineers need remain underserved across all available tools. Automatic calibration of simulation models against measured dynamometer data is still largely manual - the engineer compares predicted and measured cards visually and adjusts parameters until they match. Automated calibration using optimization algorithms would save significant time and reduce the subjectivity of the matching process.
Integration between design tools and production surveillance systems remains fragmented. An engineer using one tool for rod string design and another for well monitoring has to manually transfer data between them. The design tool does not know that the well's production has declined by 30% since the rod string was installed, and the surveillance tool does not know what stress margins the current design was operating under.
Probabilistic design methods - designing rod strings that are robust across a range of possible operating conditions rather than optimized for a single assumed condition set - are available in academic literature but not well implemented in commercial tools. Given that well conditions change continuously from the moment the rod string is installed, designing for a single operating point is inherently limited.
Where this is heading
The trajectory is toward platforms that combine simulation, surveillance, and optimization in a single environment with shared data. The desktop model of isolated simulation tools connected to nothing else will continue to serve individual engineers working on small well counts, but it does not scale to the well-to-engineer ratios and well complexities that define modern production operations.
The simulation engine itself - the wave equation solver - is mature. The physics have not changed since Gibbs published the foundational work in 1963. What is changing is everything around the solver: how data gets in, how results get compared, how designs get documented, how knowledge gets retained, and how engineering decisions connect to the operational systems that execute them. That is where the next decade of improvement in rod lift engineering software will be concentrated.
