Mid-market CFOs are changing how they evaluate automation. The era of the pilot project, where success was measured by a proof of concept that processed a few thousand records or saved a handful of hours, is giving way to a more demanding framework. Finance leaders now want to know the unit cost of automation: what it costs to process one invoice, handle one customer query or generate one report. This is not a minor adjustment in reporting. It is a reset of the ROI calculus that determines which automation projects get funded, which vendors win contracts and how internal teams are held accountable.
For software vendors, the implications are significant. Products that once sold on the promise of 'digital transformation' or 'intelligent automation' are now being evaluated against hard unit economics. The sales cycle is longer, the procurement process is more rigorous and the proof required is more granular. For mid-market companies, the shift is equally consequential. It changes how they budget for automation, how they staff their finance and operations teams and how they measure the success of their technology investments.
This article examines the drivers behind this shift, the metrics CFOs are now using, the commercial impact on buyers and vendors, and the risks and unknowns that remain. It is based on observable trends in procurement behaviour, vendor pricing models and public commentary from finance executives, but it does not rely on invented statistics or unverified claims.
Why CFOs Are Moving Beyond Pilot Metrics
Pilot metrics were designed for a different era of automation. When robotic process automation (RPA) and AI tools first entered the mid-market, the priority was to demonstrate that the technology worked. A pilot that processed 10,000 invoices with a 95 per cent accuracy rate was considered a success. The problem is that pilots rarely scale. They are often run by a small team, on a limited dataset, with significant human oversight. The cost per transaction in a pilot is artificially low because the infrastructure, integration and change management costs are not fully loaded.
CFOs have become sceptical of these numbers. They have seen too many pilots that looked impressive in a slide deck but failed to deliver when rolled out across multiple departments or geographies. The result is a demand for unit economics: the fully loaded cost of processing a single transaction or completing a single process using automation, compared with the same cost under the previous manual or semi-automated process.
This shift is also driven by the maturity of the automation market. Early adopters have already captured the low-hanging fruit. The remaining opportunities are in more complex processes, where the cost of automation is higher and the margin for error is smaller. CFOs need a more precise tool to evaluate these opportunities, and unit economics provides that precision.
The New Metrics: Cost per Transaction, Cost per FTE Equivalent
CFOs are now asking for a standard set of unit economics metrics. The most common are:
- Cost per transaction: The fully loaded cost of processing one invoice, one purchase order, one customer service ticket or one compliance check. This includes software licensing, infrastructure, integration, maintenance, training and the cost of human oversight.
- Cost per full-time equivalent (FTE) saved: The total cost of automation divided by the number of FTE hours or roles that are no longer required. This is a more conservative measure than 'hours saved' because it accounts for the fact that not all saved hours are redeployed to productive work.
- Payback period: The time it takes for the cumulative savings from automation to cover the initial investment. CFOs are increasingly demanding payback periods of 12 to 18 months for mid-market automation projects, which is a significant tightening from the 24 to 36 months that were common in the past.
- Cost per process: For end-to-end processes, the total cost of running the process with automation versus without it, including exception handling and rework.
These metrics are not new in theory, but they are new in practice for many mid-market finance teams. The challenge is that they require data that is often not readily available. Most mid-market companies do not have a clear view of their current cost per transaction for manual processes. They may know the total cost of their accounts payable department, but they do not know how much it costs to process a single invoice. The shift to unit economics is therefore forcing CFOs to invest in process mining, activity-based costing and more granular financial reporting.
The Commercial Impact on Software Buyers and Vendors
For software buyers, the shift to unit economics has several commercial implications. First, it changes the negotiation dynamic. Vendors who price their products on a per-seat or per-robot basis are being challenged to justify that pricing against the unit cost savings. CFOs are asking for pricing models that align with the value delivered, such as per-transaction pricing or outcome-based pricing. This is a significant shift from the traditional software licensing model.
Second, it changes the internal business case. Automation projects are no longer funded as 'innovation initiatives' or 'digital transformation programmes'. They are funded as cost-reduction projects with a clear ROI target. This means that the project sponsor must be able to demonstrate, with data, that the automation will reduce the cost per transaction by a specific amount. This is a higher bar than the qualitative benefits that were often used in the past.
For vendors, the implications are equally significant. Products that cannot demonstrate a clear unit economic benefit are at a disadvantage. Vendors are responding by developing ROI calculators, offering proof-of-value engagements that measure unit costs, and moving to more flexible pricing models. However, this is not without risk. Per-transaction pricing can reduce revenue predictability, and outcome-based pricing requires vendors to take on more risk. The vendors that succeed will be those that can help their customers measure the right metrics and then deliver on them.
Why It Matters
The shift from pilot metrics to unit economics is not a passing fad. It reflects a broader trend in the mid-market towards more disciplined technology investment. CFOs are under pressure to deliver cost savings, and they are no longer willing to take a leap of faith on automation. They want to see the numbers. This matters for several reasons.
First, it will accelerate the consolidation of the automation market. Vendors that cannot provide credible unit economics will struggle to win deals, while those that can will gain market share. Second, it will change the way automation is deployed. Projects that do not meet the unit cost threshold will be shelved, and resources will be concentrated on the processes with the highest ROI. Third, it will increase the importance of data quality and process standardisation. Unit economics are only as good as the data they are based on, and many mid-market companies will need to invest in better data infrastructure to support this analysis.
Risks and Unknowns
There are several risks and unknowns associated with this shift. The most significant is the risk of over-optimisation. If CFOs focus too narrowly on unit cost reduction, they may miss the broader benefits of automation, such as improved accuracy, faster cycle times and better employee satisfaction. These benefits are harder to quantify but can be more valuable in the long run.
Another risk is that unit economics can be gamed. Vendors may structure their pricing or their proof-of-value engagements to produce favourable numbers, and internal teams may be tempted to manipulate the baseline cost per transaction to make the automation look more attractive. CFOs need to be aware of these risks and ensure that the metrics are independently verified.
There is also the unknown of how the technology will evolve. As AI models become more capable, the cost per transaction for automation is likely to fall, which could change the ROI calculus again. CFOs need to build flexibility into their evaluation frameworks to account for this.
FY Outlook
The shift to unit economics is likely to accelerate over the next 12 to 24 months. As more mid-market CFOs adopt this framework, it will become the standard for automation procurement. This will put pressure on vendors to provide more transparent pricing and more robust ROI evidence. It will also encourage the development of better tools for measuring unit costs, such as process mining and activity-based costing software.
In the near term, we expect to see more mid-market companies conducting formal ROI assessments before committing to automation projects. We also expect to see more vendors offering proof-of-value engagements that measure unit costs in the customer's environment. The vendors that adapt to this new reality will be well positioned, while those that continue to sell on hype will struggle.
For CFOs, the message is clear: the era of the pilot is over. The question is no longer 'does automation work?' but 'what does it cost per unit, and is that cost lower than the alternative?' The CFOs who answer that question with rigour will be the ones who deliver sustainable value from automation.
Conclusion
The automation ROI reset is a fundamental change in how mid-market companies evaluate and invest in automation. By moving from pilot metrics to unit economics, CFOs are bringing the same discipline to automation that they apply to other capital investments. This is a positive development for the industry, as it will lead to more efficient allocation of capital and more realistic expectations. However, it also creates challenges for vendors and internal teams, who must now prove their value in hard numbers. The companies that embrace this shift and build the capabilities to measure and manage unit economics will be the winners in the next phase of the automation economy.
