GROWTH in Google Sheets: Exponential Fits, Defaults and Multi-Column Predictors
Use Google Sheets GROWTH with complete array outputs, documented optional defaults and exact exponential fixtures for one or two independent variables.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
Quick answer: GROWTH(known_data_y,[known_data_x],[new_data_x],[b]) fits an exponential trend and returns fitted Y values for selected X inputs. The last argument controls whether the multiplicative coefficient is estimated or fixed at one.
Availability and version history
- Introduced: first introduction date is not established by the reviewed official history; no desktop release number is inferred.
- Availability: documented in the rolling Google Sheets service; these examples use the web worksheet workflow.
- Checked: October 14, 2026; official GROWTH reference. Documentation and independent models, with no native Sheets execution or wider client certification.
Syntax and all three optional defaults
=GROWTH(known_data_y,[known_data_x],[new_data_x],[b])
Only known Y is required. Google documents these defaults:
- Omitted known X uses the sequence 1,2,3,… with the same length as known Y.
- Omitted new X uses known X, returning fitted values for the original inputs.
- The final argument defaults to TRUE, which estimates the coefficient b in
Y=b*m^X. FALSE fixes b at 1.
The Boolean argument named b controls estimation; it is not a place to type a desired numeric coefficient. All examples use positive known Y and exact exponential relationships. Native zero/negative-Y errors and noisy-fit numerical algorithms are not established by this compact reference.
Example 1: predict two new exponential values
Enter X 1,2,3 in A2:A4 and Y 6,18,54 in B2:B4. Put new X values 4,5 in D2:D3. Enter in E2, leaving E2:E3 clear for the expected result:
=GROWTH(B2:B4,A2:A4,D2:D3,TRUE)
The exact relationship is Y=2*3^X. Expected output is a two-row, one-column array:
| Cell | Expected value |
|---|---|
| E2 | 162 |
| E3 | 486 |
The table headers explain the locations; they are not returned cells. These are independently calculated exponential outputs, not a native spill capture or a guarantee that an actual process will continue tripling.
Example 2: omit the new target range
Enter this in G2 with G2:G4 clear:
=GROWTH(B2:B4,A2:A4)
Expected three-row output is 6,18,54. With new X omitted, Google says to return fitted values at the original X inputs. Here the observations already lie on an exact exponential curve, so the fitted and supplied Y values coincide.
For noisy data, that equality is not guaranteed. The function returns fitted estimates rather than simply copying the Y range by definition.
Advanced example: omit known X as well
=GROWTH(B2:B4)
The default X sequence is 1,2,3, matching this particular fixture. Expected fitted output is again 6,18,54.
That coincidence does not make omitted X safe for arbitrary observations. If the real X values have unequal spacing, replacing them with 1,2,3 changes the model. Supply the actual independent-variable values and their units when row position is not the intended scale.
Advanced example: fix the multiplicative coefficient at one
Use a separate dataset with X 1,2,3 and Y 2,4,8, following Y=2^X. For new X 4,5, the formula with the final argument FALSE has expected outputs 16,32.
For example, place this second X series in J2:J4, Y in K2:K4 and targets in M2:M3, then enter in N2:
=GROWTH(K2:K4,J2:J4,M2:M3,FALSE)
The relationship already has coefficient b=1, so fixing it to one is appropriate for this exact control. Applying FALSE to an arbitrary series imposes a constraint and can change the fit; it does not mean setting an additive intercept to zero in the original Y scale.
Advanced example: two independent variables
Google allows a one-column known-Y range with multiple independent-variable columns. In a separate area or sheet, enter:
| X1 | X2 | Y |
|---|---|---|
| 0 | 0 | 2 |
| 1 | 0 | 6 |
| 0 | 1 | 10 |
| 1 | 1 | 30 |
Assume this table occupies A1:C5. It follows Y=2*3^X1*5^X2. Put new predictor rows (2,0),(0,2),(2,1) in E2:F4. In G2 enter:
=GROWTH(C2:C5,A2:B5,E2:F4,TRUE)
Expected output G2:G4 is 18,50,90, a three-row, one-column array. Keep the two predictor columns in the same order for training and prediction.
Dimensions and remaining limits
For a two-dimensional known-Y array, Google requires known X to have the same dimensions or be omitted. The multiple-predictor examples above instead use a one-dimensional Y series, as the reference permits.
Unverified cases include invalid shapes, source errors, missing observations, nonpositive known Y and extreme exponential values. No exact native error labels or partial-output behavior are assigned here. FORECAST fits a straight line, a different model from this multiplicative curve.
References and verification
Examples use fictional data, English function names and comma separators. Expected outputs are independent mathematical calculations or explicitly documented rules, not native Sheets captures. Check the spreadsheet locale when reproducing punctuation-sensitive formulas.