Links

Spend data integrations

Why connect to a finance system?

There's no single way to understand all SaaS application use, but analysing spend data is a good way to uncover applications that are being bought, and to check coverage of license data you might have entered.

Connecting to a finance system

Each finance system has a slightly different API, and configuration varies.
To find out more about configuration, look for the Integration help button on the integration page.
Finance systems contain a range of data so we minimise the access we have wherever possible and advise on the minimal set of roles or permissions required to enforce this.
We list the permissions that are required with an explanation of why each is required:

Limiting imports to specific expense categories

When you connect a source of financial data to Trelica by default we look for expense category codes that might relate to SaaS or software spend.
You can add or remove from this list.
We will only import transactions from the categories you choose.
The exception is if you click Scan other expense categories for miscategorized software spend. This will pull data from other expense categories and look for possible SaaS products.
You can also choose the date from which transactions will be imported.

Automatic updates

After the first import, Trelica will look for new or updated transactions once a week.
If you want to manually force a refresh, just click on the Refresh button:

Processing flow

This diagram describes the requests made when connecting to finance systems:
Spend data import flow
  1. 1.
    Trelica pulls back a list of account codes from the target application
  2. 2.
    The user configuring the connection selects codes related to software spend, and also the first date to pull historical data from.
  3. 3.
    Trelica explicitly queries just for those account codes, and from the specified date.
  4. 4.
    Trelica needs a small number of fields (see below); APIs typically return additional data. We discard additional data, and only store the fields we need.
  5. 5.
    Data is sent to our mapping engine to try to find matches against our catalog of 14,000 SaaS applications and vendors.
The integration is then refreshed automatically on a schedule (weekly by default), in order to pull transactions changed or added since the previous run.

Viewing and mapping transactions

Once your transactions have been imported you can view them by going to the Spend page and clicking the relevant data source:

Manually importing transactions

You can also manually import Excel or CSV file data. Go to Spend and choose Import Spreadsheet.
Upload an Excel or CSV file containing spend transaction data. We recommend Excel as this gives more consistent date and number formatting than CSV.
The file can optionally have a header row (recommended) and must at minimum contain columns containing:
  • Description
  • Date
  • Amount
Click Next to move to the next step which is to map the columns in the file.
If you are exporting a file from a finance system there's a chance your files might contain the same data, e.g. if a full report is run monthly from the start of the year to the current date.
In this case, make sure the export contains a unique ID for each transaction which can be mapped to the Transaction ID field in Trelica.
This means that each reload will update transactions previously uploaded with the same ID, rather than creating new ones..
Once you're done, click Import. As the file is processed you will see a summary of progress.
Click Review to check and change how transactions have been mapped to applications.

Fields stored by Trelica

Name
Status
Detail
Transaction ID
Mandatory
Unique ID for the transaction so that Trelica doesn’t fetch duplicate data
Billing Date
Mandatory
Date the transaction took place
Amount
Mandatory
Value of transaction
Currency Code
Optional
Required if the system reports data in multiple currencies
Description
Mandatory
Label for the transaction
Account Code
Mandatory
GL code or Account code
Transaction Status
Optional
Paid / Unpaid - mostly used for expense data so that we don’t retrieve data that is unapproved
Vendor Name
Optional
Some systems split out the vendor name from the description field
Cost Center
Optional
Associated cost center for the transaction
Period Start / End
Optional
Start and end (for amortization)
Note
Optional
Memo field
​