Payhawk helps with expense automation by reducing the manual work when reviewing, confirming the details, or paying company expenses.
Payhawk utilizes optical character recognition (OCR) technology reading more than 65 languages including Cyrillic, an advanced AI data extraction algorithm, and multiple machine learning (ML) algorithms, to name a few, to leverage the data and automatically suggest as many expense fields as possible.
Automatic data extraction with OCR
Optical Character Recognition or Reader (OCR) is the electronic process of converting images of typed or printed text into machine-readable text, whether from a scanned document or a photo.
Payhawk uses OCR to automatically extract expense data from images and scanned documents captured through the mobile app. Once text is detected, the system proceeds to extract key details such as the total amount, invoice number, document date, and other information needed for expense categorization.
Next, Payhawk uses its AI-powered data extraction algorithm, which identifies and pulls out the relevant data from uploaded invoices and receipts, and automatically fills in the appropriate expense fields with the extracted information.
From expense and payment documents, the data extraction tool retrieves the following details and populates the corresponding expense fields:
Supplier (name, VAT number, country)
Document date
Document number
Due date
Total amount
Tax amount
Document currency
Document type (invoice or receipt)
For more information on the Payhawk OCR implementation, see the Smart data extraction made simple with OCR technology blog post by Trish Toovey.
Understanding expense categorization in Payhawk
Expense categorization in Payhawk refers to the essential master data your company requires for each expense. This information is organized into specific fields, either built-in by Payhawk (such as Document date or Category) or custom-defined by your company’s Payhawk administrators (such as Cost center or Project).
Some of this expense data can be automatically extracted by Payhawk's automation tool from attached receipts or invoices, such as the supplier name or document date. However, not all required information is available in the expense documents. In such cases, the missing data must be entered manually by the expense owner or reviewer.
Understanding suggestion strategies in Payhawk
To reduce manual input for expense fields where data isn’t available in receipts or invoices (and therefore cannot be extracted), Payhawk administrators can set up automated methods called suggestion strategies.
These strategies help the system suggest appropriate values for non-extractable fields such as Team, Category, VAT code, and other custom fields.
Locating the suggestion strategies settings
In Payhawk, suggestion strategies can be defined for the Category built-in expense field and any custom field created by the Payhawk administrators at your company.
The option to define a suggestion strategy is located in the Settings tab of the expense field. For example, to define a suggestion strategy for the Project custom field, go to Expense fields > Custom fields > Project > Settings > Suggestions.
Available suggestion strategies
Payhawk supports the following types of strategies for auto-suggesting expense field data:
None - Leave the expense field blank for employees to fill in manually.
One default value - Set the expense field to be automatically populated with a default value that’s rarely changed during review.
Payhawk AI - Enable the expense field to be automatically populated by Payhawk AI based on historical data for your account.
Supplier - Configure the expense field to be automatically populated based on the selected supplier, with specific values assigned to each supplier.
Team/employee - Assign the expense field to be automatically populated based on each employee or team, with inherited defaults and fallback options.
Tax rate - Automatically suggest the tax rate based on either the supplier or the category. This is configured in the Expense categorization tab for each supplier or within the properties of each accounting category.
Excluding auto-population of expense fields
If Suggestions is set to None, the expense field will not be automatically populated and will remain blank. In this case, employees will need to manually enter the required data for the expense field.
Auto-populating expense fields with predefined values
If Suggestions is set to One default value, you’ll need to define that default value, which will then be automatically populated in the expense field for every expense.
This auto-population strategy is ideal for custom fields that have the same value in over 90% of cases and are rarely adjusted by the Finance team during the review process.
Auto-populating expense fields based on historical account data
If Suggestions is set to Payhawk AI, the expense field will be automatically populated based on historical data collected by the Payhawk system for your account.
By default, the Payhawk AI suggestion strategy is applied to all custom fields. It uses an advanced machine learning model that takes into account various factors, such as the employee, supplier, historical field values, merchant category (for card transactions), and other relevant data, to determine the most appropriate value for each field on a per-expense basis.
This strategy is ideal for businesses looking to automate expense categorization, especially when there are no strict rules for specific fields. It’s particularly effective for fields where patterns exist but are too complex or variable to define manually.
Auto-populating expense fields for suppliers
If Suggestions is set to Supplier, the expense field will be automatically populated based on the selected supplier.
This auto-population strategy is useful when certain fields, like Category or custom fields, are directly tied to specific suppliers, for example, Uber might always be categorized as Taxi allowance, and Facebook Ads as Marketing Ads.
Once this strategy is selected, you can assign a specific value to each supplier:
In the Payhawk web portal, go to Suppliers.
Select the desired supplier and go to its Expense categorization tab.
Set the desired value for the expense field.
From then on, all expenses from that supplier will automatically use the configured value.
Setting a default auto-populated value for all suppliers
You can also set a default value for all suppliers from the Suggestions > Default value for all suppliers drop-down list in case a specific supplier doesn’t have one assigned. For example, if the Supplier location custom expense field isn’t set for a supplier, you can default it to Germany. The more supplier values you configure, the higher your expense automation rate will be.
Auto-populating expense fields for all employees and teams
If Suggestions is set to Team / employee and expenses are being auto-populated, you must assign a default suggested value for the selected field for each employee or team.
This auto-population strategy is useful for expense fields that are directly tied to the employee or team, such as Cost center, Department, or Location.
Once this suggestion strategy is selected, you can assign values to each team and employee:
In the Payhawk web portal, go to Employees.
Select the desired employee or team and go to their Expense categorization tab.
Set the desired value for the expense field.
By default, the employee inherits the default value from their team - in this case, you will see the Use team default option in the expense categorization section. If no value is set for the team, the system will check the parent team, and so on.
For example, if the Marketing - Website team is assigned the Project: Website Redesign, all team members will automatically inherit that value. This inheritance logic saves time, reduces manual setup, and ensures consistent categorization across the organization.
Additionally, you can set a default value for all teams and employees to be used when no specific value is assigned from the Suggestions > Default value for all teams/ employee drop-down list. For example, if no Cost center is set for an employee or their team, you can default it to Contractor expenses.
Auto-populating the tax rate expense field
The tax rate plays a key role in both the tax reclaim process and how expenses are categorized. In Payhawk, the built-in Tax rate expense field supports the following suggestion strategies.
Supplier-driven tax rate suggestion
You can assign a specific tax rate to each supplier, or leave it blank and manually enter the rate for each expense.
To set a tax rate per supplier, go to Suppliers > [Selected supplier] > Expense categorization > Tax rate. This works similarly to the supplier-based suggestion strategy for other fields.
Category-driven tax rate suggestion
Alternatively, you can define a tax rate for each category. In this case, the selected category will automatically suggest the corresponding tax rate.
Category-driven tax rate suggestions are not available if you have a live ERP integration where tax rates and categories are managed directly from your accounting system.