Unlike other P2P markets, in Airtm it is the customer who posts a request, taking the first step to start a peer-to-peer exchange. The Matching Algorithm is the set of automatic rules used in the process of matching customers and cashiers according to an optimization criterion. Customer requests are presented to Cashiers, who are advanced, verified, well-performing customers, who have been empowered to service these transactions, helping customers complete the exchange, while earning a commission for themselves. 

In order to optimize transaction distribution and marketplace performance, transactions are not presented simultaneously to all cashiers; instead, only cashiers who have permissions on a particular method and who are active online are eligible to view customer requests.

When a new request is entered, the Matching Algorithm randomly selects a number of cashiers, to whom the customer's request is presented. If the transaction remains unaccepted, after a certain number of seconds has passed, the algorithm selects a new group of cashiers at random to serve the transaction. This dynamic continues until the transaction is accepted by a cashier, canceled by the requester, or expires.

The number of cashiers contained in each group and the time between batches of cashiers are private business variables that are monitored and adjusted according to the performance of the marketplace.

Frequent questions

When the transaction is matched with a second group, is it hid to the first group?

NO. Matching is done incrementally. As an example, suppose that the algorithm shows the operation to 10 cashiers at random, after 30 seconds, if none of these cashiers has accepted it, it proceeds to show it to another 10 cashiers randomly, that is, the transaction is already being viewed by 20 cashiers, the algorithm will continue to incrementally display it to new random batches of 10 cashiers until all active cashiers with permission on that payment method have been matched with that transaction.

Why is the transaction shown to incremental groups instead of everyone simultaneously?

By distributing transactions in this way, all cashiers are more likely to accept requests because on each transaction they see, they are competing against a limited group of cashiers, rather than competing against hundreds or thousands.

Are the cashier groups pre-constituted? Am I always part of the same group?

No. There are no “subgroups” of cashiers, the algorithm takes a random sample of the total number of cashiers active at the time, who have permission to accept requests with that payment method.

Does the algorithm take into account how many transactions I have previously taken to decide whether or not to send me transactions?

NO. The Matching Algorithm does not take into account how many transactions a cashier has completed or how much money they have earned. The sample is completely random from all the cashiers found online.

Does the algorithm favor new cashiers over experienced cashiers?

NO. The algorithm logic does not contain information on the registration date, the number of completed transactions, or any other similar metrics.

Why does the volume of transactions I see vary from day to day?

The volume of requests within Airtm fluctuates according to many internal and external factors. Additionally, the number of active cashiers also varies and plays an important role.

Are there “VIP” or somehow privileged cashiers to whom more transactions are distributed?

NO. Airtm distributes transactions completely randomly, there are no special groups or cashiers with advantages or preferences in the distribution of transactions. If you have strong evidence that a cashier has an advantage or mechanism that allows them to view more transactions, you can report it anonymously here.