Ivalua vs Medius vs Esker for AP Automation

For a $120M services company moving off manual email-chain AP, Ivalua delivers KPI tracking through an AP-centric dashboard embedded directly in its AP Automation module (Invoice Hub). The platform captures workflow event data across invoice receipt, matching, approval, and payment, and surfaces those signals as real-time metrics. Ivalua explicitly names invoice cycle time, touchless processing rate, exception rate, early payment capture, and discounts as tracked metrics within built-in analytics and real-time dashboards: "Built-in analytics and real-time dashboards can help you track key metrics such as invoice cycle time, touchless processing rate, exception rate, early payment capture, or discounts" (Ivalua, Automated Invoice Processing blog). The P2P automation guide further specifies cost per transaction as a tracked metric alongside exception rate by category or supplier, with Ivalua benchmarking 70%+ touchless rate, under 10% exception rate, and under $6 cost per invoice as performance targets visible within platform dashboards. Discount capture is tracked through Ivalua's early payment and dynamic discounting capabilities, with discount capture rate named as a day-one KPI in Ivalua's P2P best practices content. The Analytics and Insights module provides an extensive catalog of standard reports plus the ability to craft targeted analyses and dashboards, with KPIs pulling automatically from Source-to-Pay workflows without manual data exports from Sage Intacct.

For a $120M multi-location services company replacing email-chain approvals with no current KPI visibility, Medius delivers these five metrics through its dedicated Medius Analytics module, which sits within the AP Automation platform and surfaces pre-defined dashboards and KPIs drawing from the invoice-to-pay workflow event database. The module provides a full, real-time view into spend delivered through pre-defined reports, dashboards, and KPIs, including a dedicated Capture dashboard that surfaces touchless and extraction rates with drill-down to the individual supplier level. Medius surfaces comprehensive, real-time analytics on invoice status, exception rates, cycle times, and cost per invoice, so AP leaders can report with confidence, track against industry benchmarks, and continuously improve performance. Touchless rate is tracked as a named, configurable dashboard gadget: AP managers discover the 'Touchless Metrics for Invoices by Suppliers' gadget, add it to their dashboard, and sort by invoice volume to see each supplier's corresponding touchless percentage. For average days to approve, the AP Benchmark Report names total invoice processing cycle time and average approval time as discrete, tracked KPIs for both PO and non-PO invoices. The KPI dashboard in Medius Analytics also delivers insight on lead times by sub-process and invoice handler, enabling identification of specific approval bottlenecks. Discount capture rate is addressed through the analytics and payment scheduling layer: Discount Capture Rate is a documented KPI within Medius Analytics, indicating the extent to which available early payment discounts are being captured, with optimization leading to cost savings over time. Customers confirm that access to process KPIs within the tool allows them to easily measure and track results, enabling proactive settings optimization to drive higher automation rates.

For a $120M multi-location services company with a 3-person AP team processing 1,800 invoices monthly across two Sage Intacct entities, Esker delivers KPI tracking through embedded, role-based dashboards built directly into its AP Automation module. As invoices move through the workflow: receipt, validation, matching, approval, and posting, the platform captures timestamped event data and surfaces it in real time without requiring manual exports or IT involvement. The Esker on Demand datasheet enumerates distinct dashboard views by role: CFOs see AP process metrics and Days Payable Outstanding; AP Managers see payment KPIs, process efficiency, and accrual reporting; Cost Center Owners see spend analysis and requests pending approval. The platform gives teams access to dashboards, KPIs and actionable insights that support better decisions at scale. All five buyer-requested KPIs are specifically documented across Esker's published materials: intelligent dashboards with real-time KPIs allow finance teams to customize AP metrics in easy-to-read graphs and reports, including average processing time by month. Customisable dashboards display live analytics and can track any KPI you can imagine. The exception rate and touchless rate are explicitly documented as platform-tracked metrics: Esker customers report 80%+ touchless invoices and 70%+ fewer exceptions as measured outcomes of the platform. Cost per invoice is named as the first of Esker's 10 key AP KPIs in their published eBook, and early payment discounts captured is also enumerated as a tracked KPI, with the system logging and flagging critical payment due dates to enable discount capture. Reporting tools empower AP managers to monitor volume of invoices processed by FTE per day, average time to process an invoice, and other Key Performance Indicators. The Esker Anywhere mobile app extends this visibility: it enables managers to review and approve supplier invoices and track KPIs while on the go directly from Apple or Android devices.

For a 3-person AP team manually keying 1,800 invoices per month with no prior automation, Medius Capture addresses Stage 1 of the pre-processing journey (invoice capture and data extraction) using a multi-stage proprietary AI pipeline: Siamese CNNs for document classification, Markov models for line-item extraction, and a CNN-based coding engine called SmartFlow. The critical mechanism for this buyer's requirement is that SmartFlow is 'trained on your historical actions' at the tenant level, not solely on a generic global dataset: Medius explicitly states that touchless processing rates 'improve systematically with time and volume as Medius's models learn from each customer's unique coding patterns and correction signals.' This means every time an AP user corrects a field or validates a coding suggestion, that correction is fed back into the per-customer model, progressively reducing manual touches on that supplier's invoice format. Medius also states that its machine learning 'uses pattern recognition to capture invoices, code them correctly, and route them for processing, all based on patterns specific to that company,' and the vendor claims 95%+ coding precision is reached after as few as two processed invoices from a new supplier. The global pre-training dataset (2.4 billion+ invoice field data points, including 393 million+ real-world human corrections) provides a strong starting baseline, so accuracy on new vendor formats begins high and continues to improve as your specific correction history accumulates.

For a 3-person AP team processing 1,800 invoices/month across a diverse vendor base, Esker Synergy AI operates two complementary learning mechanisms inside its invoice capture stage. The first is passive auto-learning: each time an AP user corrects an extracted field during the validation form step, the system silently updates its extraction model so that the next invoice sharing that document structure is populated with higher accuracy without any manual intervention. The second is active teaching: a trained power user can open a teaching request form and define explicit extraction rules for a specific recurring supplier template; those rules then apply automatically to future lookalike documents. Both mechanisms work at the header and line-item level. An automation trends dashboard tracks per-vendor improvement signals including number of changes by vendor and automation rate by month, giving the AP team direct visibility into whether accuracy is improving for a given supplier format over time. Esker confirmed in its U.S. patent press release that this machine learning capability applies specifically to vendor invoices, not only to order documents.

For a multi-location services company processing 1,800 invoices per month across a varied vendor base, Ivalua's Invoice Hub handles stage-1 capture using what it calls Hybrid Invoice Data Capture (IDC), a combination of OCR and cognitive AI that operates at the pre-processing journey stage before matching or approval routing. The cognitive layer uses image segmentation neural networks to locate and extract invoice fields rather than relying on fixed per-supplier templates, and it applies a statistical supplier identification model to recognize suppliers even when their names or logos vary across formats. A 2023 platform release added 'unique management of the OCR and AI-generated invoice data capture rules,' giving AP administrators the ability to view, inspect, and manually refine the underlying extraction rules rather than treating the system as a black box. This means accuracy can improve over time, but the documented improvement path combines system-generated AI rules with human-managed rule tuning by the AP team, rather than a fully autonomous correction-feedback loop that automatically updates per-vendor recognition models each time an operator corrects an extracted field.