Company Financials Forecast

Submit a Brazilian company's CNPJ to receive a full 2-year financial forecast.

The API automatically:

  • Retrieves the company's historical financial statements (or uses the ones you provide)
  • Identifies the relevant economic sector and selects macro variables that drive it
  • Runs an econometric model (OLS regression) for each of the 7 key indicators
  • Produces a complete projected Income Statement (DRE) with absolute BRL values
  • Generates AI-powered explanations in Portuguese for every projected indicator

Projected years are determined automatically based on the latest available financial period.
For example, if the most recent filing is 2024, the forecast covers 2025 and 2026.

Providing your own financial data

You can optionally send historical financial statements in the financial_data field.
When provided, the API skips the internal data retrieval and runs the forecast
directly on the data you supply. This is useful for:

  • Companies not yet in the Klooks database
  • Running forecasts on adjusted or pro-forma financials
  • Testing scenarios with modified historical data

The CNPJ is still required to identify the company's sector and select
the appropriate macroeconomic variables.

Recent Requests
Log in to see full request history
TimeStatusUser Agent
Retrieving recent requests…
LoadingLoading…
Body Params
string
required

Brazilian company tax ID (CNPJ). Accepts any format:

  • Formatted: 33.000.167/0001-01
  • Digits only: 33000167000101

Always required — used to identify the company's sector and select macroeconomic variables.

financial_data
object

Optional historical financial statements. When provided, the API uses this data
instead of fetching from the Klooks database.

Keys are year strings ("2022", "2023", "2024").
Each year must include at least the dre block. The contasDetalhadas and
balancoPatrimonial blocks are recommended for a complete forecast.

Minimum required: 2 years of data (3+ recommended for higher confidence).

boolean
Defaults to false

When true, includes additional statistical details in the response:

  • confidence_interval — 95% confidence bounds for each projected year
  • wide_confidence_interval — flag indicating when the interval is unusually wide
  • model block — R², adjusted R², observation count, significant variables and coefficients

Default is false for a cleaner response.

Responses

Language
LoadingLoading…
Response
Click Try It! to start a request and see the response here! Or choose an example:
application/json