1) Alternatively, you might prefer to use a modified indicator: "the number of new jobs created by the supported businesses".
2) Decide what type of work counts as "employment" (for example: How long does a contract need to last? Is part-time work acceptable? Does the employee really has to have a formal contract?).
3) If possible, measure the indicator in at least two points in time (for example, 2-3 months and then again 8 months after the support provision). The second measurement will enable you to assess the extent to which the new positions were sustained.
4) Minimize the incidence of "fake reporting" by verifying the employment information from several different sources, such as the employer, other employees, documentation (contract, payroll), the employee and other key informants.