# ERP Data Dictionary Worksheet > Source: https://netray.co/tools/erp-data-dictionary-worksheet > Publisher: Netray (netray.co), AI and integration engineering for ERP and manufacturing systems. > Section: Free Tools ## Short answer An editable table to document ERP tables and columns with type, length, null rate, fill rate and owner. ## Category developer ## Audience - ERP analyst - Database administrator - Data steward ## Tool **Kind:** worksheet **Mode:** editable **Columns:** #### Table **Key:** table_name **Type:** text #### Column **Key:** column_name **Type:** text #### Data type **Key:** dtype **Type:** select **Options:** - int - decimal - varchar - nvarchar - date - datetime - bit - uniqueidentifier #### Length **Key:** len **Type:** number #### Key **Key:** key_type **Type:** select **Options:** - Primary - Foreign - Unique - None #### Null rate % **Key:** null_pct **Type:** number **Unit:** % #### Fill rate % **Key:** fill_pct **Type:** number **Expr:** 100 - null_pct **Format:** percent #### Business meaning **Key:** meaning **Type:** text #### Data owner **Key:** owner **Type:** text **Rows:** **Table name:** item **Column name:** item **Dtype:** nvarchar **Len:** 30 **Key type:** Primary **Null pct:** 0 **Meaning:** Item number **Owner:** Materials **Table name:** item **Column name:** description **Dtype:** nvarchar **Len:** 40 **Key type:** None **Null pct:** 1.5 **Meaning:** Item description shown on screens **Owner:** Engineering **Table name:** item **Column name:** u_m **Dtype:** nvarchar **Len:** 3 **Key type:** Foreign **Null pct:** 0 **Meaning:** Stocking unit of measure **Owner:** Materials **Table name:** po_line **Column name:** due_date **Dtype:** datetime **Len:** 8 **Key type:** None **Null pct:** 4.2 **Meaning:** Promise date on purchase line **Owner:** Purchasing **Table name:** po_line **Column name:** qty_ordered **Dtype:** decimal **Len:** 9 **Key type:** None **Null pct:** 0 **Meaning:** Ordered quantity **Owner:** Purchasing **Totals:** #### Average null rate **Key:** null_pct **Agg:** avg **Format:** percent #### Columns documented **Key:** column_name **Agg:** count **Format:** integer **Summary:** #### Average fill rate **Key:** avg_fill **Expr:** 100 - avg_null_pct **Format:** percent **Primary:** true **Notes:** Table and column names in the example are illustrative. Replace them with names from your own ERP schema. Null rates can be measured with a COUNT of nulls divided by COUNT of rows per column. ## Content **Intro:** Every ERP analyst has a private notebook of what each column really means. This worksheet turns that into a shared data dictionary you can export to Excel and keep current. **Sections:** #### How to use the worksheet **Body:** Add one row per column you actually use. Do not try to document an entire ERP schema in a weekend, since thousands of columns are unused. Start with the tables behind your top reports. Record the real business meaning, not just the technical name, because many ERP columns are reused for something their original name does not suggest. #### Measuring null and fill rate **Body:** Null rate is the share of rows where a column holds no value. Fill rate is the complement, 100 minus null rate. A column with a 4% null rate on a promise date means 4% of lines cannot be judged late, which affects any on-time delivery report. Profile each documented column with a simple query and enter the result here so quality is visible beside meaning. **Bullets:** - Null rate = null rows divided by total rows - Also check empty strings and placeholder values such as 0 or 1900-01-01 - Re-measure quarterly or after major loads #### Ownership **Body:** A dictionary without owners goes stale. Name a data owner per row, usually the function that creates or relies on the value, and ask them to confirm the meaning. Disagreement about what a column means is useful information, and is the starting point of a data governance conversation. #### Common mistakes **Body:** Teams copy the technical schema from the database and call it a dictionary. That lists type and length but misses meaning, allowed values, source and quality. Another mistake is ignoring user defined fields, which often hold the most business critical data in customized ERPs. **Cta:** A good dictionary also makes AI over your ERP far more accurate. If you want a governed assistant that understands your columns, we can help you build that on-premises. **Faq:** #### How many columns should I document first? **Answer:** Start with 50 to 150 columns behind your most used reports and integrations. It is more useful to have a small accurate dictionary than a large stale one. Grow it as questions come in. #### Where do I get data types and lengths? **Answer:** From the database catalog, for example INFORMATION_SCHEMA.COLUMNS in SQL Server or the data dictionary views in Oracle. Export them and paste into the worksheet, then add meaning and owners by hand. #### What is a good fill rate? **Answer:** It depends on the column. Keys should be 100%. Optional descriptive fields may be much lower. What matters is that the rate matches the business rule, and that unexpected gaps are investigated. #### Can this replace a data catalog tool? **Answer:** For a small team, a shared spreadsheet is a good start. Larger organizations eventually want a catalog tool with lineage and workflow, but the content you collect here transfers directly. ## Search terms erp data dictionary template, data dictionary excel template, database column documentation template, erp table documentation, data dictionary worksheet, column level data catalog template, sql data dictionary template --- Talk to an engineer: https://netray.co/contact?intent=erp_ai | Machine catalog: https://netray.co/agent/catalog.json