Purchase Order Invoice Reconciliation

Posted under case studies on May 06 , 2020 by Dheeraj Sangore


Reconciliation allows any entity to juxtapose their records against the records of another entity to reveal the variation in data.So, to avoid discrepancies, it becomes necessary to do reconciliation frequently. Reconciliation is used to make sure that the number of records is neither taken as an extra nor missing out any. The problem described here specifically relates to reconciliation between purchase orders given to internet providers and their corresponding invoices.

Scope of problem:

Here our client, which is a Bank, is reconciling invoices raised by the vendors (internet providers) with purchase order(PO) data of Procurement department. They have various branches in different cities which are identified by unique Sol ID. They are taking network services from various internet providers for its various branches in different cities. Hence the process of matching the records of vendor invoices and PO data referred is referred to as Network Reconciliation. It needs various checks and validations to be applied on the received invoice data and the data which is in PO master file. After satisfying all the conditions like bandwidth, invoice period, PO number, SOL ID, commissioning date, Invoice Amount, Invoice Period, Annual Recurring Charges (ARC) amount, one-time charge (OTC) amount match with PO data, the invoice is sent to the Finance department for payment activity

Challenges faced in manual processing:

  1. Due to larger data it requires large amount of time to check and validate invoices records with the PO master file. Various checks and validations are shown below:

    • Invoice period should lie between Purchase orderperiods
    • Invoice not already paid or Invoice payment in process
    • PO number, Vendor Details, Bank Branch Name, PO Amount, Service description (bandwidth- 1mbps ,2mbps etc.) should match for PO Data and the Invoice
    • Amount raised in invoice should be less than or equal to the PO Amount on pro rata basis and the Pending amount for the entire PO Period
  2. Data inconsistencies in PO Data and Invoice, making it difficult to match.
  3. Dependency on excel file for storing large number of records.
  4. Unable to identify that for which timeduration the invoice is paid from PO data.

Before Sheetkraft:

  1. Purchase order and Invoice data stored in excel.
  2. Data inconsistency which makes the reconciliation part cumbersome.
  3. No common field between invoice and PO data to filter already paid invoices.


After Sheetkraft:

  1. Automating Reconciliation which can cover all the case scenarios.
  2. Eliminate dependency of excel file for storing the large amount of PO data having lots of data inconsistencies.
  3. Storing invoice record in the database for ease in checking paid invoices and invoices in process.
  4. Paid Invoices and Invoice payment in process can be distinguished through UTR no. mapping against the Invoices thereby adding a check which did not exist previously, and users had to manually go in the Finance Payment Portal to check for individual invoices.



| Parameter            | Before Sheetkraft                             | After Sheetkraft                                 |
|--------------------- | --------------------------------------------  | ------------------------------------------------ |
| Processing time      | 8 hours                                       | 3 mins                                           |
| Frequency            | Depending upon the invoice received by vendor | Depending upon the invoice received by vendor    |
| Storing PO data      | Excel file                                    | Database                                         |
| Storing Invoice data | Excel file                                    | Database                                         |


Let's talk


7th floor, Unit No. 701, C Wing,
Kailas Business Park, S. Veer Savarkar Marg,
Park Site, Vikhroli (W), Mumbai-400079.


For General Queries & Technical Support



For Sale and Demos