↑↓ selectEnter openEsc close

    1. Elbuz

    How to merge duplicate attributes into one

    11 min readUpdated

    Elbuz merges attributes for one characteristic — “Diagonal”, “Screen diagonal”, “Display size” — into one and remembers the names so new price lists don't recreate them.

    When you need this:

    • Suppliers use three names for one characteristic — “Link and delete” moves values into the main attribute.
    • Each price list creates its own “Weight” — “Remove duplicate attributes” merges identical names directory-wide.
    • Price lists and the parser bring “Wt.” and “Mass” — saved names go into the main attribute.
    • An export needs one parameter, not two — a synonym swaps the name and joins values.

    Where to start: select the attribute to keep and open “Maintenance” → “Cleanup” → “Matching”.

    What merging duplicates does

    Synonyms: how to export several attributes under one name

    Sooner or later the directory has several rows for the same thing: “Diagonal”, “Screen diagonal”, “Display size”. This isn't an error or a glitch. It happens when the catalog is filled from different sources: every website and every supplier names a characteristic its own way, and Elbuz stores attributes uniquely by name, so each variant becomes a separate row.

    The fix for such rows is not deleting them but synonyms. You mark one attribute as the main one and link the rest to it: they stay in the directory but are no longer independent. You do this with the “Synonym” button. First select a regular attribute (not a synonym), otherwise Elbuz replies “Select a regular attribute (not a synonym) in the table first — then you can create a synonym”.

    How to make “Body color” a synonym of “Color”:

    1. In the “Attributes” window, select the “Color” row — a regular attribute, not a synonym.
    2. Click “Synonym” on the toolbar (tooltip “Create Attribute Synonyms”; the same item is in the row's context menu).
    3. In the “Create attribute synonyms for Color” window, tick “Selection” next to “Body color”. The link is saved right away; there is no save button.

    You don't type a name: an existing attribute becomes the synonym. “Color” becomes the main attribute, and “Body color” becomes its synonym.

    The “Create attribute synonyms for Color” window with a hint and a table of attributes
    1Hint above the table
    2Selection column
    1. The hint explains that checking an attribute adds it to the synonyms of the selected attribute.
    2. The Selection column holds a checkbox per attribute; each check is saved immediately.

    A synonym you create manually doesn't move values or delete attributes. The link changes the name the attribute is exported under. Values products already have under “Temperature” stay where they are, and the attribute itself doesn't go anywhere: the directory still has three rows. This is the most common misunderstanding: it seems that after creating a synonym “everything will merge by itself”, but only the name merges.

    What a synonym really does: on export, the subordinate attribute's name is replaced with the main attribute's name, and values from all linked attributes are collected into one parameter. A product with both “Temperature” and “Color temperature” filled in sends one “Color temperature” to the file, with both values separated by a comma, not two parameters with the same name. Nothing changes in the directory itself: each value stays with its own attribute until you merge them.

    You unlink a synonym from the main attribute where you created it: select the main attribute, click “Synonym” and, in the “Create attribute synonyms for …” window, clear “Selection” in the attribute's row. It becomes independent again; product values aren't touched. If you unlink the last synonym from a main attribute, the “main” mark is removed automatically: a single attribute doesn't need it.

    Three rows about one characteristic are merged into one not with synonyms but with the “Link and delete” button in the “Attribute matching” window.

    If attributes come from parsing, you can avoid creating duplicates instead of cleaning them up: the matching dictionary converts a name from another website into yours while data is collected. Read about it in the article on what exactly the parser takes from a product page.

    How to stop price lists from creating duplicates: the “Attribute matching” window

    The “Attribute matching” window does what synonyms don't: it merges several attributes into one, moving product values into the main attribute and removing extra attributes from the directory. It also blocks new duplicates: after matching, names from price lists and parsers go into the main attribute on import. You open it from the “Maintenance” → “Cleanup” → “Matching” menu in the same “Attributes” window.

    The main attribute is the one selected in the table when you open the window. If none is selected, the window opens, but instead of the linking block it shows the hint “No attribute selected. Close the window, pick an attribute in the catalog list and open “Matching” again — it will become the main one.”

    Attribute matching window: main attribute, subordinate attributes with ticks, and the two action buttons
    1Main attribute
    2Ticks on subordinate attributes
    3Save to dictionary
    4Link and delete
    1. Main attribute — the selected row; the other attributes get linked to it.
    2. Ticks mark the attributes that mean the same as the main one.
    3. Save to dictionary — safe: remembers the names without moving or deleting anything.
    4. Link and delete — irreversible: moves values into the main attribute and removes the subordinate ones.

    The window has two parts. The top one, “Link catalog attributes”, shows the main attribute's name in the header and the directory table with checkboxes below it. The bottom one, “Current matches”, is what Elbuz already knows about names from price lists and supplier websites.

    A checkbox in the top table is the same synonym as the “Synonym” button: it saves the link to the database right away. So if synonyms were created earlier, the checkboxes are already ticked when the window opens, and you don't need to tick them again.

    Both buttons start the same way, by writing names to the matching dictionary. Then “Save to dictionary” stops, while “Link and delete” moves the values and deletes the attributes
    “Link and delete”noyes“Save to dictionary”noyesnoyesButton in the “Attribute matching”windowWhich button?Confirmation “This cannot beundone. Continue?”Nothing changesSubordinate attributes ticked?Toast “Tick the attributes you wantto link to the main one”Subordinate names are written tothe matching dictionarythe main one is marked as main, therest become its synonyms“Link and delete” clicked?Toast “Matches saved:” with thecountvalues and attributes stay in placeProduct values move to the mainattributeSubordinate attributes are deletedfrom the directoryseveral values of a product arejoined with commasToast “Attributes linked and deleted:”with the count
    • Action or result
    • Decision: the path depends on the answer

    “Save to dictionary”: so duplicates don't appear later

    This button moves nothing and deletes nothing. It takes the names of the ticked attributes and writes them to the internal matching dictionary, each name pointing to the main attribute. Price list imports and the then read this dictionary: when a supplier file has a “Temperature” column, Elbuz puts its values into the main attribute instead of creating a new one. The ticked attributes also become synonyms of the main one: in exports, their name is replaced with the main attribute's name.

    If nothing is ticked, Elbuz tells you so: “Tick the attributes you want to link to the main one” appears at the bottom.

    • Product values don't change. They stay with their attributes.
    • Attributes aren't deleted. All three rows are still in the directory table.
    • Existing duplicates aren't cleaned up. The button only blocks new ones.

    You can undo it: clear the checkbox in this window or in the “Synonym” window, and remove the row from the bottom table with the trash icon.

    How to merge existing duplicates: “Link and delete”

    This action is irreversible. Elbuz asks you to confirm: “Product values from the ticked attributes will move to the main one, and those attributes will be deleted from the catalog. This cannot be undone. Continue?” There really is no way to undo it, except restoring the whole database from a backup.

    What happens, step by step:

    1. The names of the ticked attributes are written to the matching dictionary, just as “Save to dictionary” does. This matters: the names stay after the attributes are deleted, so the next price list import with a “Temperature” column fills the main attribute instead of creating a new one.
    2. Product values move to the main attribute.
    3. If a product ends up with several values of the main attribute, they are joined into one, comma-separated and sorted alphabetically: “2700K, 5000K”. Identical values aren't repeated: if both attributes had “5000K”, one “5000K” remains. Check this beforehand on one or two products: if a product really has two different temperatures, the attribute will hold a list.
    4. The extra attributes are deleted from the directory together with their descriptions, category links and settings.

    The main attribute's name doesn't change. Empty values aren't filled in: if a product had none of the attributes being merged, nothing appears in it.

    If a subordinate attribute was added as a column to the product grid, a field formula or an export template, that setting disappears along with the attribute. The main attribute doesn't get such a column automatically; you'll need to add it again.

    Names in the bottom table, “Current matches”, can be removed one at a time with the trash icon. You need this when a name is linked to the wrong attribute: once the row is deleted, price list imports treat that name as a separate attribute again.

    Current matches card of the Attribute matching window
    1Name in source
    2Catalog attribute
    3Delete match
    1. Name in source — how the characteristic is named in the supplier file or site.
    2. Catalog attribute — the attribute it is mapped to on import.
    3. Delete match — removes one mapping.

    Synonym or matching: which to choose

    Both deal with the same thing, different names for one characteristic, but they work in opposite directions.

    Synonyms and matching: the difference
    What it affectsSynonym“Attribute matching” window
    Product values Doesn't touch them Moves them to the main attribute, joins duplicates
    The attributes themselves All stay in the directory Extra ones are deleted, only with the “Link and delete” button
    When it works On export: the name is replaced with the main attribute's name On import: a name from a price list or website resolves to a catalog attribute
    Can you undo it Yes, by clearing the checkbox “Save to dictionary” — yes; “Link and delete” — no

    For the typical case, “three rows in the directory for one characteristic”: open “Matching” with the main attribute selected, check the checkboxes, click “Link and delete”. You don't need to create synonyms separately: the button does that too.

    How to remove identical names across the whole directory

    After several price lists, the directory often has attributes with exactly the same name: each price list created its own “Weight”. For this case there is the “Maintenance” → “Danger zone” → “Remove duplicate attributes” command. It asks for confirmation with the text “Duplicate attributes will be removed and only one attribute will remain. The search is conducted by attribute name. All product attribute values will be transferred to the one remaining attribute.” and goes through the whole directory at once.

    • What counts as a duplicate. Attributes with the same name in the catalog's main language (the default language), not the interface language. For the command, “Weight” and “Weight, kg” are different; merge them with “Matching”.
    • Which one stays. In each group of identical names, the oldest attribute, the one created first. You can't choose another one; if it matters which one stays, use “Matching”.
    • What moves over. Product values, values in price list rows and price list column mappings move to the remaining attribute. If a product had two values in different duplicates, they are joined into one, comma-separated.
    • What is lost. The duplicates' links in parsers aren't moved to the remaining attribute (value replacement rules set up for duplicates do move to it). Also lost: the duplicates' category links with their “Display in filter” checkboxes, duplicate columns in the catalog grid, field formulas that referred to them (copies stay in formula version history), and product Excel export columns with them. The remaining attribute keeps only its own categories, so check afterwards that it's enabled where the duplicates were.
    • There's no undo. Nothing can be rolled back. Before running it on a large catalog, merge the most important characteristics manually with “Matching”.

    Scenarios

    Suppliers called one characteristic by three different names

    What to do. In the table, select the attribute that should stay, open “Maintenance” → “Cleanup” → “Matching”, check the checkboxes in the “Link catalog attributes” table and click “Link and delete”.

    What you get. Product values move to the main attribute, several values of one product are joined with commas, and the extra attributes disappear from the directory. Their names stay in the matching dictionary, so the next price list import puts such a column into the main attribute instead of creating it again.

    What won't happen. There's no way to undo it. Grid columns, field formulas and export template fields built on the deleted attributes disappear with them and don't move to the main one. The main attribute's name doesn't change. If you don't want to touch product values and only want to block new duplicates, click “Save to dictionary”.

    After several price lists, the directory has five “Weight” attributes

    What to do. If it matters which attribute stays, merge the most important characteristics with “Matching”. Merge the rest with “Maintenance” → “Danger zone” → “Remove duplicate attributes”.

    What you get. In each group of identical names, the oldest attribute stays, with the values of all the duplicates.

    What won't happen. The duplicates' categories don't move to the remaining attribute. Check that it's enabled where the duplicates were (how to enable it).

    What to do when duplicates didn't merge

    Common problems
    What you seeWhyWhat to do
    “Select a regular attribute (not a synonym) in the table first — then you can create a synonym” You clicked “Synonym” with no row selected, or the selected row is an Select a regular attribute in the table: its name isn't highlighted in green
    You created a synonym, but product values stayed in the old attribute and the attribute itself didn't go away A synonym only replaces the name on export: it doesn't move values or delete attributes Open the “Attribute matching” window with the main attribute and click “Link and delete” (merging duplicates)
    You clicked “Save to dictionary” or “Link and delete”, and Elbuz replied “Tick the attributes you want to link to the main one” No subordinate attribute is ticked in the window, and the main attribute has no synonyms created earlier Tick attributes in the “Link catalog attributes” table and click the button again (merging duplicates)

    FAQ

    I created synonyms, but the values didn't move to the main attribute. Why?

    A synonym only replaces the name on export. To move the values and remove the extra attributes, open “Matching” with the main attribute and click “Link and delete” (merging duplicates).

    How do I unlink a synonym from the main attribute?

    Select the main attribute, click “Synonym” and clear “Selection” in the synonym's row. It becomes independent again, and product values don't change (synonyms).

    Suppliers use “Weight”, “Wt.” and “Mass”. How do I stop price list imports from creating duplicates?

    Select “Weight”, open “Matching”, tick “Wt.” and “Mass” and click “Save to dictionary”: the next price lists put these columns into “Weight”. “Link and delete” also moves the values already collected (matching).

    After several price lists I have lots of identical attributes. How do I merge them?

    Exactly identical names are merged by “Maintenance” → “Danger zone” → “Remove duplicate attributes”. Different names (“Weight” and “Weight, kg”) are merged with “Matching” (identical names).

    Which attribute stays after “Remove duplicate attributes”?

    The oldest one in each group of identical names; you can't choose another one. If it matters which one stays, merge these attributes with “Matching” (“Link and delete”).

    Where to find it and who has access

    Menu
    Directories → Base product catalog → Attributes
    Matching
    selected attribute → “Maintenance” → “Cleanup” → “Matching”
    Synonym
    selected attribute → the “Synonym” button or the row's context menu
    Identical names
    “Maintenance” → “Danger zone” → “Remove duplicate attributes”
    Access right
    the “Attributes” window in group permissions

    Related topics

    Was this article helpful?

    Missing a feature? Tell us.

    We will add it for free if other users need it too.

    If only you need it, we will agree on it separately.

    Article contents

    Suggest an improvement

    Try Elbuz free