How to Build an Import Mask in EDGEBIC, Field by Field

Building an import mask takes about five minutes and you do it once per recurring file. After that, importing this week's version of the same file is select the mask, click Do It, read the counts. This post walks the mask editor field by field.

For what a mask is and why EDGEBIC by User Solutions uses mapped files rather than fixed connectors, start at import masks explained.

Step 1: Open Import and Pick the Entity Type

Import lives under Settings, on the Import tab. The page shows an entity type list on the left and the masks saved for the selected type beside it.

Pick the entity type first, because it determines which fields exist to map:

Import sits inside Settings, which is permission controlled. If you cannot see the tab, that is an access question rather than a licensing one.

Step 2: Define the Mask and Read the File

Click Define New. The mask editor opens.

Name it after the routine, not the file. Weekly Product Sync outlives products-week29.xlsx, and the mask is what you will pick from a list next month.

Click Get File and choose the source. For a workbook, pick the worksheet in the Sheet box. EDGEBIC reads the column headings and lists them on the right, ready to drag.

One caution about sheets: if the workbook's sheet name later changes, the mask falls back to the first worksheet rather than failing. That is forgiving when a file is regenerated and dangerous when a workbook has several sheets, because you can silently import the wrong data. Re-pick the file and the sheet whenever the source workbook is restructured.

Step 3: Map the Columns

The field mappings grid has one row per available field, with columns for the target field, whether it is required, the source heading, the column number, and the conversion factor.

Drag a heading from the file-headers list onto the matching row's source-header cell. Rows ticked as required must be mapped before the mask can save and run.

Two rules make this faster than it looks.

Map only what you need. Unmapped columns are simply not read. A forty-column ERP extract usually needs a handful of mappings, and leaving the rest alone means next week's file can arrive untouched.

Absorb the mismatch here, not in the file. The mask is the place where your ERP's vocabulary meets the scheduler's. Doing it here means nobody has to clean a spreadsheet by hand on a Monday morning, which is the step that always eventually gets skipped.

Step 4: Set the File Options

For a fixed-format extract with no heading row, untick header present and fill the column number on each mapped row instead. Everything else works identically.

Step 5: Set the Conversion Factors

This is the field that prevents the most common bulk error there is.

The multiplication happens before the value is stored. A setup column reading 30 with a factor of 0.016667 lands as 0.5 hours. Non-numeric cells pass through untouched, and a factor applied to a zero correctly produces zero.

Get this wrong on a routing import and every setup value in the file is sixty times too large. The import will report success, because every row was structurally valid, and the error shows up as machines that appear to spend entire shifts changing over. Set the factor when you build the mask and check one imported value afterwards.

Step 6: Set the Behaviour Options

Select the saved mask and click Options. The available options depend on the entity type.

Two defaults are worth understanding rather than accepting.

Continue on row error is on. One typo does not block 199 good rows. Switch it off when a partial load would be worse than no load, such as a demand file where a missing quantity means the whole extract is suspect.

Update existing records is off. That makes re-running a file safe: matches come back Reused and untouched. Turn it on only for a run whose purpose is changing values. On an update, blank cells fall back to the existing value rather than wiping it, but zeros are real values and will be written. Map only the columns your run is meant to change. The three auto-create missing master options run the other way, on by default, so a sales order or routing file may create the products, customers, and work centers it names.

Step 7: Save, Run, and Read the Result

Click Save. The mask appears in the saved list with its name, last file and format.

To run it: select the mask, click Do It, and pick this week's file if the remembered path is stale. A progress window shows the run.

The result dialog gives you a summary line reading Created, Updated, Reused and Failed counts, plus a grid with one row per file row showing its status, key, message and the identifier of the record it touched.

Read it in this order:

1. Failed count. Any failure means a row did not land. Open the message.
2. The Created versus Reused split. If a re-run of a familiar file suddenly shows everything Created, your key column stopped matching. That is more serious than a failure, because it silently duplicates master data.
3. The log file. Click Open Log File for the full record: every row, including the bookkeeping rows the dialog hides, with exact reasons for each failure. The dialog gives you counts; the log gives you row numbers.

Log files live one per run and stay until deleted, which makes them a usable record of what entered the system and when.

Step 8: Prove It Before You Trust It

Two verification steps, both cheap.

Dry-run on ten rows. Cut the source down to ten representative rows and run that first. Ten rows expose a missing mandatory mapping (which aborts before a single row is read), a unit sixty times out, and a key column that matches nothing. Check the created records on their own screens rather than trusting counts.

Run a plant-wide diagnostics sweep after the full load. An import changes hundreds of records at once, and a single mapping error repeats across all of them. The sweep is the thing that catches the shape of a bad column before anyone schedules against it. The routine is in how to run and read the anomaly report.

Where to Go Next

Routing imports have their own mechanics, because steps must be linked to each other after every step exists. That is how EDGEBIC imports routings in two passes, and it also explains the wipe-and-recreate behaviour that surprises people on a second run.

The failures worth pre-empting are collected in import mask mistakes. For what the routing columns mean before you map them, read how to create a manufacturing routing. For per-ERP application, start at the ERP integration hub or the Fourth Shift integration guide. The wider discipline is ERP data integration best practices.

Full platform map: the complete EDGEBIC guide and the EDGEBIC hub.

Bring one real export to a demo and we will build the mask on screen. Contact US.