# Reformatting Data

> By reformatting a field, you can create a more readable, and more usable format for analysis. In some cases, you can also reduce space. For example, convert a date field to a numeric field. You can reformat fields within the ElastiCube using a custom SQL expression.

*Source: https://docs.sisense.com/main/SisenseLinux/reformatting-data.htm*

---

Last updated: June 11, 2026

|  |  |
| --- | --- |
| [Tier](https://www.sisense.com/pricing/#pricing) | [Deployment](https://docs.sisense.com/main/SisenseLinux/introduction-to-sisense-cloud-managed-services.md#ComparisonofManagedCloudandSelfHosted) |
| Launch     Grow    Enterprise | Cloud     On-Prem |

By reformatting a field, you can create a more readable, and more usable format for analysis. In some cases, you can
also reduce space. For example, convert a date field to a numeric field. You can reformat fields within the ElastiCube
using a custom SQL expression.

**Note:**

When using floating points data types (Real, float, double), some decimal numbers may be rounded. If you need full
decimal accuracy, try using BigInt field types.

## Numeric Representation of Date Fields

### Business Case

Create a "Date" table using a numeric representation of a Date field instead of the Date field itself to improve
query performances. In addition, a numeric representation of a Date can provide the ability of filtering the data by
time range.

### Solution

To convert a Date ("4/21/2012 12:36:56 AM") to an integer (20120421), use the following syntax:

```
10000*getyear(Date)+100*getmonth(Date)+getday(Date)      
```

To get the hours and minutes too, use a bigint numeric representation. So, "4/21/2012 12:36:56 AM" will become:
201204210036, using the following syntax:

```
tobigint(100000000*getyear(DateTime)+getmonth(DateTime)*1000000+getday(DateTime)*10000+100*gethour(DateTime)+getminute(DateTime))      
```