NTT COMWARE English Site


NTT COMWARE Starts "PayMa" the AI-Based Payment Matching Support SaaS
- The more data you have, the more efficient AI makes the payment matching

NTT COMWARE CORPORATION (Head office: Minato-ku, Tokyo, President: Satoshi Kurishima, hereinafter NTT COMWARE) launches the service of "PayMa (Smart Payment Matching)" (hereinafter PayMa), a SaaS for payment matching, in November 2019. "PayMa" enables more efficient operation in payment matching by using AI to automate the process, which has been complicated due to a lot of visual inspections or manual tasks required in corporate accounting.

1. Background
In recent years, RPA tools have digitized finance and accounting operations to automate some repetitive manual tasks in checking data or account reconciliation. It still remains difficult to fully automate the procedures with conventional RPA, which need patterns in automation because it is unable to identify payees from the payer information from payment data, and is unable to check multiple invoices at one time. Therefore, these processes still require visual inspection or accounting personnel to undertake manual tasks. Furthermore, as market needs have changed from "ownership to use", subscription business models have rapidly become popular in the domestic market. However, companies are struggling with the burden caused by the increase in tasks to reconcile invoice and payment records in continuous use in such systems since it is necessary to reconcile repetitive payment on a regular basis.
Under the circumstances, NTT COMWARE has developed and is promoting "PayMa", the solution to support account reconciliation tasks by matching payment and invoice data using machine learning by combining its know-how and AI technologies to further promote digital transformation (DX conversion) of business for the accounting departments of companies.

2. Outline of the Service
"PayMa" provides the following three functions as a service specialized in the payment reconciliation process in accounting operations.

(1) Collect the name of the payer and payee (corporate name) from the payment data and automatically link them
(2) Extract billing data for payment matching based on the company name, payment amount or payment date, etc.
(Extract the data applied to either of the patterns: payment: billing 1: 1, 1: N or N: 1*1)
(3) AI-based machine learning of matching results between billing data and payment to improve accuracy

The settings can also be customized to create an amount range for payment differences payment and billing due to transfer fees and consumption tax.
This service is available regardless of industries or type of business. The customers can quickly start the service as SaaS without creating a system.

* 1) There are some payment patterns to set, e.g., a batch payment for multiple bills (1: N) or payment by installments for a single bill (N:1).

Figure: Usage image of

*2) PayMa can integrate billing data on NTT COMWARE "Smart Billing(R)" or billing data managed by customers' own billing system or tools

Figure: Usage image of "PayMa"

3. Effects after implementation
PayMa is a rule-based matching system in addition to using AI learning. The accuracy will be improved as customers use the system and reflect the results in their operation in following payment matching. The service is expected to save time and reduce the workload for payment matching.
The results of the trial service*3) show about an 80% increase in efficiency to automatically process payment data, which eventually reduced the time needed by about 90% for payment matching*4) *5).

Figure: Results of trial

Figure: Results of trial

* 3) Learning the actual payment matching results manually created by accounting personnel for the data in the previous year (approximately 20,000 payments/year and approximately 220,000 bills/year).
* 4) Source: NTT COMWARE inhouse research
* 5) The results are subject to the content of the target data.

4. Future Scenario
Preorders are accepted for "PayMa" in November 2019 to start the service in January 2020 or after. The functions will be upgraded for further efficiencies in accounting work through our continuous efforts to improve the service reflecting customers' voices and API integration to NTT COMWARE "Smart Billing (R)" or customers' accounting systems currently in use.

* "Smart Billing" is a registered trademark of NTT COMWARE CORPORATION.
* Company names and product names may be trademarks or registered trademarks of each company.

The information shown in this page is the latest as of the day of the release. It is subject to change without further notice. Thank you for your kind understanding and consideration. PageTop