# Full Load

> Copy the whole table again each time. Simple, but slow when the table is big.

**Field:** Data Engineering  
**Category:** Ingestion & Streaming  
**Also known as:** Full Refresh, Full Extract, Full Reload, Truncate and Load

## Think of it like

It is like tipping out a toy box and putting each toy back. Nothing is missed, but it is slow.

## Definition

A full load copies every row from a source each time it runs. It first empties the old copy, or replaces it. Then it copies the whole table again. It is the simplest way to load data, and nothing can slip past. But it gets slow and costly as the table grows. Small tables are often loaded this way every night.

## Example

A shop reloads its list of 300 stores every night, and it takes seconds. Its 50 million orders are loaded a bit at a time instead.

## Example code

```sql
TRUNCATE TABLE stores;
INSERT INTO stores
SELECT * FROM source_stores;
```

Empty the copy, then copy every row back in. That is a full load.

## Why it matters

It is the safe choice when deletes and edits are hard to track. Every run ends with an exact copy.

## Common mistake

Fully loading a huge table every hour. The bill and the run time grow with every new row.

## Three principles

- **Every row.** Nothing depends on what changed.
- **Simple.** No dates or markers to track.
- **Costly at size.** Run time grows with the table.

## Related terms

- [Incremental Load](https://learn.benibix.com/term/data/incremental-load)
- [Data Ingestion](https://learn.benibix.com/term/data/data-ingestion)
- [Change Data Capture (CDC)](https://learn.benibix.com/term/data/change-data-capture)
- [Data Backfill](https://learn.benibix.com/term/data/data-backfill)
- [Batch Processing](https://learn.benibix.com/term/data/batch-processing)

## Told apart from

- [Full Load vs Incremental Load](https://learn.benibix.com/compare/full-load-vs-incremental-load): People think an incremental load gives the same result as a full load, only faster, and it does not. A full load copies every row, so the copy always matches. An incremental load copies only rows past a marker, such as a time stamp. A row deleted at the source never passes that marker, so it stays in the copy. Many teams run both: small loads each night and a full one now and then.

![A source table of 10,000 rows, with the last 200 highlighted as new. One arrow leads to a full load box that empties the copy and copies all 10,000 rows again. A second arrow leads to a highlighted incremental load box that copies only the 200 rows past the last high-water mark.](https://learn.benibix.com/diagrams/dat2-full-vs-incremental.svg)

---

Source: https://learn.benibix.com/term/data/full-load
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-19
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
