Category: Data quality

  • How to import a CSV into Salesforce without wrecking your data

    How to import a CSV into Salesforce without wrecking your data

    The short answer: To import a CSV into Salesforce safely, export a backup of the records you will touch, rehearse the load in a sandbox, fix the picklists, dates and required fields in the file, and load a small batch first. Use the Data Import Wizard for loads under 50,000 records on supported objects, and Data Loader for anything bigger. Updates need record Ids or an external ID.

    The file arrives with the subject line “quick one”. It has 4,000 rows, a column called “Status” that is nearly your picklist, and dates that someone typed by hand. The sender is already on to the next thing. You are the person who finds out, on Monday, that 600 accounts now say “Prospct”.

    Salesforce already ships two import tools that are good enough for most of this. The trouble is rarely the tool. It is what went into the file, and whether you could get back out.

    Which tool should you use to import a CSV into Salesforce?

    Use the Data Import Wizard for loads under 50,000 records on the objects it supports, and Data Loader when the load is bigger, the object isn’t supported, or the load repeats. Both are Salesforce’s own. Salesforce’s Help pages draw the line this way.

    Data Import WizardData Loader
    Where it runsIn Setup, in the browserClient application you install
    RecordsUp to 50,000 at a timeUp to 150,000,000 per CSV, per the “When to Use Data Loader” page
    ObjectsAccounts, contacts, leads, solutions, campaign members, person accounts, custom objectsObjects the wizard doesn’t support
    OperationsAdd, update, add and updateInsert, update, upsert, delete, hard delete, export
    Good atDuplicate prevention by account name and site, or contact or lead emailComplex mappings, scheduled loads

    Two cautions. Opportunities, cases, products and a few other objects aren’t on the wizard’s list, so check the list in your org (Setup, search Data Import Wizard in Quick Find). And Salesforce’s own articles don’t agree on where the wizard should hand over: one mass-update article suggests Data Loader from roughly 50,000 to 100,000 records and the Bulk API for hundreds of thousands, while the “When to Use” page gives the 150 million figure. Read the figure as a ceiling, not a target.

    Data Loader is available in Enterprise, Performance, Unlimited and Developer editions. Features vary by edition and license, so check what your org has.

    What should you check in the file before you load it?

    Check three things: picklist values, dates, and required fields, because those are where most loads fail or, worse, succeed wrongly.

    Picklists. Salesforce’s import FAQ says the wizards accept values that aren’t already in a picklist and warn before adding them. That is how a field ends up with “Prospct” as a permanent option. A restricted picklist behaves differently and rejects the value with an error such as INVALID_OR_NULL_FOR_RESTRICTED_PICKLIST. Before loading, get a distinct list of each picklist column and compare it to the field’s active values in Setup. Use API names where the field uses them.

    Dates. Salesforce’s Help explains that Data Loader converts dates in the CSV to GMT, so a machine in another time zone, or one that observes daylight saving time, can land a date a day off. Check whether the target is a Date or a Date/Time field, and spot-check a handful of rows after the first batch.

    Required fields and rules. A required field left empty, a validation rule that evaluates to true, or a duplicate value in a unique field will fail rows. That is the good outcome, because you see it. Run your validation rules in your head against the file before you start.

    One more, for updates: blank cells don’t clear existing values in the wizard, per Salesforce’s update guide. If you need to blank a field, plan a separate step for that.

    How do you update existing records without creating duplicates?

    Match on the Salesforce record Id, or on an external ID field, and never on a name you retyped. Salesforce recommends matching on the Id because it is unique and guarantees the right record is updated.

    For accounts and contacts, the wizard can also match on name, site or email. Salesforce’s update guide warns that if two accounts or contacts share a name, the wrong record may be updated, and that changing the Record Owner, Account Name, Account Site or Contact Name columns in the file creates duplicates. Data Loader’s upsert matches on a column holding either an external ID or the Salesforce Id, updating when it finds a match and creating a record when it doesn’t.

    Steps that hold up:

    1. Run a report of the records you want to change, with their Ids and the fields you will edit.
    2. Export it, edit only the columns you mean to change, and save as CSV.
    3. Match on Id (or your external ID) in the tool.
    4. For owner changes, use Mass Transfer Records instead, which Salesforce points to for ownership.

    If the load will create new records as well as update, run duplicate checks on the result afterwards. A fresh load is the most common way duplicates arrive, which is why merging duplicate accounts safely is the sequel to this post.

    Should you back up and test in a sandbox first?

    Yes to both. Salesforce’s mass-update article says “Always back up your data before performing any data operation” and recommends running a small test before the full update.

    A backup here means the records you are about to change, exported with their Ids and the current values of every field you will touch. A report export or a Data Loader export will do. It is the only thing that lets you restore an update.

    The sandbox rehearsal is my advice, not a Salesforce rule, and it has a catch: a sandbox only tells you what production would do if the sandbox matches it. Picklist values, validation rules and automation drift. If you can’t get a recent copy, a very small batch in production is the next best rehearsal. Load 20 rows, open five of them, and read the fields as the owner would. Then load the rest.

    What goes wrong, and can you undo an import?

    I found no native undo for an import. What Salesforce documents is deleting what you added, and that only helps for inserts.

    According to the Help FAQ on deleting imported data, an admin can use Mass Delete Records in Setup to remove imported accounts, contacts, leads or solutions, which needs the Modify All Data permission. For imported custom objects, an admin can use Data Loader to delete them. Deleted records go to the Recycle Bin unless you choose to delete them permanently, and Data Loader’s hard delete skips the Recycle Bin.

    For updates there is nothing to delete. The way back is the backup you took: load the old values back by Id. This is the reason for the small first batch. A bad mapping on 20 rows is an afternoon. On 4,000 rows it is a week.

    What to do Monday:

    1. Write down which of the two tools your team uses, and the record count at which you switch.

    2. Add a “backup export” step to whatever process you have for incoming files, with the Id column included.

    3. Pick the object people load most often and list its picklist fields with their active values, so the comparison takes minutes.

    Where we come in

    Portage, by GTM Cloud, is in early access. It is a wizard inside Salesforce that checks every row of a CSV or Excel file against your live field rules and, for updates, shows the old and new value side by side before anything saves. It loads files a person brings to it, so it isn’t an integration, and updates need the Id column. If a preview step is what your process lacks, see Portage.

    Questions people ask

    How many records can the Data Import Wizard import at once?

    Up to 50,000 at a time, according to Salesforce Help. It supports a defined set of objects and CSV files. For larger loads, or objects the wizard doesn’t support, Salesforce points to Data Loader. If you are planning something big, a data health check before the load shows what you are about to add to.

    Do I need an external ID to update records?

    No. You can match on the Salesforce record Id, which Salesforce recommends because it is unique. An external ID helps when the file comes from another system and doesn’t carry Salesforce Ids, and Data Loader’s upsert can match on one.

    Can I import opportunities with the Data Import Wizard?

    No. Salesforce Help lists opportunities, cases, products and a few other objects among those the wizard doesn’t import. Use Data Loader for them. Check the supported list in your org under Setup, Data Import Wizard, because it can change between releases.

    Sources

  • How to merge duplicate accounts in Salesforce without regretting it

    How to merge duplicate accounts in Salesforce without regretting it

    The short answer: In Lightning Experience you merge duplicate accounts from the View Duplicates list: pick up to three, choose one as the principal record, choose the field values to keep, and confirm. Related records move to the survivor and the others go to the Recycle Bin. Salesforce documents no undo button, so export both records first, choose the survivor on purpose, and start with a few pairs.

    Search for a customer and two results come back: “Northwind Traders” and “NORTHWIND TRADERS INC”. One has the opportunity history. The other has the contacts and the open case. You know which is the real one. Nobody else does, and the person who created the second one has since left.

    Merging is the obvious fix, and the native tool is perfectly capable of it. The hesitation is reasonable. Here is what the merge does, what it can’t take back, and a way to do it that leaves you a margin for error.

    How does the native account merge work in Salesforce?

    Open an account, click View Duplicates if Salesforce has flagged any, select up to three records, choose one as the principal record, pick any field values you want to keep from the others, and confirm. That is the Lightning Experience flow described on Salesforce Help.

    Three details matter. First, you can merge business accounts or person accounts, but not one of each. Second, the merge list only shows accounts that your duplicate rules detect. For accounts they don’t flag, the same Help page points to the merge tool in Salesforce Classic. Third, if duplicates aren’t being flagged, review the Standard Account Matching Rule before blaming anything else.

    Salesforce also says you can’t merge accounts in the mobile app.

    What happens to related records and field values?

    The principal account survives, the non-principal accounts go to the Recycle Bin, and related records are linked to the merged account with some exceptions. That is the short version of Salesforce’s “Considerations for Merging Duplicate Accounts”.

    The detail, as the Help article states it:

    • Related items such as campaigns and activities are linked to the merged record, with some exceptions.
    • Chatter feeds come only from the principal record. Salesforce Files in the feed or the Files related list are kept.
    • Team members from the non-principal accounts are kept on the merged account.
    • Hidden and read-only data, such as sharing settings, comes from the principal record.
    • Created By and Created Date come from the oldest of the merged records, and the merge date shows as the Last Modified date.
    • You choose field values per field. Anything you don’t choose from the other records is lost with them.
    • In orgs with Experience Cloud sites, merged contacts must belong to the same site type. To keep site-associated contacts, make the account with site users the principal.

    There are also hard stops. Accounts tied to relationship groups can’t be merged in Lightning Experience, and team members with values in unique custom fields can block a merge until you clear them. If two accounts are indirectly related to the same contact, Salesforce asks you to remove the redundant relationship first. Read the full considerations page once before your first real merge. It is long, and it is the page that explains the error message later.

    Who is allowed to merge accounts?

    You need Delete access on accounts, and you must be an admin, the account owner, or a user above the owner in the role hierarchy. Reading the account requires Read access. This is from Salesforce Help for Lightning Experience.

    The Delete requirement is the thing to remember. Merging is a delete of the losing record plus a reparent of its children, so it needs the right that a delete needs. Don’t hand it out because a rep wants tidy data. In orgs where team members have greater access to child accounts than the owner does, Salesforce says the merge may require Modify All Data on accounts.

    Can you undo a merge in Salesforce?

    Not natively as a merge. The pages I could read don’t describe an undo, and the one thing Salesforce does document is the Recycle Bin.

    Help says the non-principal accounts “are moved to the Recycle Bin”, so a losing record can be restored while it’s still there. For contacts, a Salesforce Help article says that restoring the duplicate doesn’t change the master record, which keeps its “Contact Merged” history entry. It doesn’t say related records move back.

    The undelete documentation in the API guide, which I could read only through a search excerpt, says records removed by a merge can be undeleted but that child objects were reparented and that can’t be reversed. Treat that as how it probably behaves, and test before you rely on it. Expect a restored account to come back without the contacts and opportunities that moved, and verify that in a sandbox. Items stay in the Recycle Bin for 15 days, per Salesforce Help, after which they are scheduled for permanent deletion.

    So the honest answer is that a native merge is a one-way door unless you took a copy first.

    What do duplicate rules and matching rules do?

    A matching rule decides what counts as a duplicate, and a duplicate rule decides what happens when one is found: block the save, alert the user, and optionally report it. Salesforce’s duplicate rules page also says rules can run on creation or edit, and that duplicate rules don’t run on records that are manually merged.

    Matching rules have limits worth knowing. Salesforce allows up to five active matching rules per object, 25 active in the org, and 100 in total. Fuzzy matching supports only Latin characters, and Salesforce recommends exact matching for international data. Don’t mix fuzzy and exact methods in one rule, or duplicates won’t be detected.

    The Report option groups the duplicates it finds into duplicate record sets, which is the closest thing the platform has to a worklist. It’s useful. It isn’t a review screen. For where duplicates come from in the first place, see our post on importing a CSV into Salesforce safely.

    What is a safe process for merging duplicate accounts?

    Export first, choose the survivor deliberately, and start with a handful of pairs you can check by hand. That order matters more than any tool.

    1. Export both records and their children. Run a report on the accounts in the pair, with Id, owner, created date and the fields you care about. Export their contacts, opportunities and cases too, with their Ids. Salesforce’s mass-update guidance says to back up before any data operation, and a merge is one.
    2. Choose the principal on purpose. The principal’s sharing settings, hierarchy and feed survive, and the oldest created date wins regardless. Prefer the record with the site users, the parent in the hierarchy, and the integration ids. Opinion: when in doubt, keep the one more systems already point to.
    3. Decide field values before you click. Write down which phone, address and owner you want. The merge screen asks per field, and nothing you skip comes back.
    4. Merge three or four pairs. Then open the results as a user would: contacts, open opportunities, activity history, the sharing on the record.
    5. Scale up only after that. Batches of 20 or so, with a check after each.

    What to do Monday:

    1. Run your existing duplicate rules’ Report option, or the View Duplicates list, for accounts only, and count the pairs.

    2. Pick five pairs where you are certain, export both sides with children, and merge them.

    3. Check who in your org holds Delete on accounts, and whether they should.

    Where we come in

    Sluice, by GTM Cloud, is in early access. It builds on the duplicate rules you already have, lists duplicate groups for side-by-side review with a Result After Merge column, and takes a backup before each merge so you can undo it from History. It finds and resolves duplicates inside Salesforce. It doesn’t enrich records or match against outside data. See Sluice, or ask for a Data Health Check to size the problem first.

    Questions people ask

    How many accounts can I merge at once in Salesforce?

    Up to three accounts per merge in Lightning Experience, according to Salesforce Help. One becomes the principal record and the others go to the Recycle Bin. For a long list, you repeat the merge pair by pair, which is why a worklist and a clear survivor rule save so much time.

    Why doesn’t Salesforce show my duplicates in the merge list?

    The Lightning merge list only shows accounts your duplicate rules detect. Salesforce Help says to review the Standard Account Matching Rule if duplicates aren’t flagged, and points to the Classic merge tool for accounts the rules miss. If your duplicates arrive through file loads, our guide to loading a CSV safely covers that side.

    Which account should be the principal record?

    The one you can least afford to lose parts of. The principal’s hidden and read-only data, such as sharing settings, carries over, and its feed is the one kept. If site users or a hierarchy parent sit on one record, make that record the principal. That is the Help guidance for site users, and my view for the rest.

    Sources

  • Who owns the records of people who left?

    Who owns the records of people who left?

    The short answer: The departed person does, until you move the records. Deactivating a Salesforce user doesn’t reassign what they own. To find those records, report on each object with the owner’s Active field set to false. To reassign them, use Mass Transfer Records for accounts, leads and custom objects, and Data Loader for opportunities, cases and activities.

    Say a rep leaves on Friday. IT switches off the account, someone ticks the box on the user record, and the leaver email goes round. On Monday an automation that updates opportunities starts failing on 40 of them, a dashboard that “runs as” the rep comes back empty, and a lead routed to a queue last month is still sitting under a name nobody recognises.

    None of this is a bug. It is what the platform documents. The gap is that nobody owns the follow-up.

    What does Salesforce do when you deactivate a user?

    It stops the person logging in and frees their license, and that is nearly all. According to Salesforce Help, the user stays in the list marked inactive, and records they own stay with them until you transfer ownership.

    What else the Help pages say happens:

    • Deactivated users lose access to records shared with them manually or through team membership, effective immediately. Users higher in the role hierarchy keep access until the shares are removed asynchronously.
    • They are removed from other users’ default account and opportunity teams, but their own default teams aren’t removed.
    • Files they own aren’t deleted, and the deactivated user remains the owner until an admin reassigns them.
    • Chatter groups they own stay theirs until reassigned.
    • Inactive users can still hold permission set assignments, so Salesforce suggests removing those afterward.
    • Deactivating needs the Manage Internal Users permission.

    Reassigning first is not required, but if a user owns more than 10,000 records of one object, Salesforce says to transfer them beforehand to avoid performance issues.

    Why do records owned by inactive users matter?

    Because things that run on ownership keep running on a name nobody answers to. Four places it bites.

    Automation. Salesforce Help says “processes can’t update records that are owned by inactive users”, so a flow or process that edits those records fails until they have an active owner. There is an admin setup that lets users create and edit records with inactive owners. In Setup, User Interface holds the org-level checkbox, and a permission, Update Records with Inactive Owners, covers users. Treat it as a workaround, not a plan.

    Reports and dashboards. Salesforce Help warns that Lightning dashboards owned by or running as a deactivated user may return unexpected results. Change the dashboard owner.

    Sharing. Per Salesforce Help, the person loses access shared with them manually or through teams. Managers higher in the role hierarchy keep access until the shares are removed asynchronously, so check what the rest of the team can still see before you assume nothing changed.

    Assignment. Records stuck on an old name don’t get follow-up, and some defaults block deactivation outright. Salesforce lists default owners for leads and cases, the default workflow user, the default lead creator, workflow email alert recipients, custom hierarchy field selections and Customer Portal administrators. Approvers must be removed from approval processes. If one of these blocks you, freeze the user to stop logins now and finish the cleanup before deactivating, because freezing doesn’t release the license.

    How do you find records owned by inactive users?

    Build a report on the object with a filter on the owner’s Active field. Salesforce Help describes this for leads: create a custom report type with Leads as the primary object, add the Lead Owner’s Active field through the layout editor, then filter “Lead Owner: Active equals False”.

    The steps from Setup:

    1. Create, then Report Types, then New Custom Report Type, with the object as primary.
    2. Edit Layout, add fields related via lookup, and add the owner’s Active field.
    3. On the Reports tab, create a report from that type.
    4. Add the filter “Owner: Active equals False” (the label follows the object, such as Lead Owner).
    5. Group by owner, and save it in a shared folder.

    Repeat for accounts, opportunities, cases and any custom object that matters. Salesforce documents the leads case, so read the same pattern across to the other objects as my adaptation. If you want a quick estimate first, open the Inactive Users list in Setup, then group your main objects by owner and look for those names. Count the result as a share of the total. It is number five in seven data quality numbers every admin should know.

    How do you reassign records from an inactive user?

    Use Mass Transfer Records for accounts, leads, service contracts and custom objects, a list view for small batches, and Data Loader for opportunities, cases and activities. Each has a documented scope.

    MethodWhat it coversDocumented limits
    Mass Transfer Records (Setup)Accounts, leads, service contracts, custom objectsUp to 250 records at a time
    List view Change OwnerLeads in the Help exampleA list view shows 2,000 records
    Data Loader update on OwnerIdOpportunities, cases, activities, anythingYou export, edit and reload the file

    Mass Transfer Records. Setup, Quick Find, Mass Transfer Records. For accounts it also moves contacts (business accounts only), attachments, notes, open activities, and open opportunities owned by the current owner. Closed opportunities and other users’ open opportunities are optional. For leads, open activities move, except when the lead goes to a queue. You need Transfer Record and Transfer Leads for accounts, and Transfer Record for custom objects.

    One thing to know first. In Professional, Enterprise, Unlimited, Performance and Developer editions, transferring accounts removes access given by manual sharing, Apex managed sharing and sharing rules, then applies the new owner’s rules. As reported by Salesforce Ben, Winter ’27 adds a Keep Manual Shares When Transferring Records option under Sharing Settings, off by default and org-wide when on. Its status isn’t stated in that article, so check the release notes. We cover it in manual shares and owner changes.

    List views. Salesforce Help describes selecting leads in a list view and choosing Change Owner. The new owner needs at least Read access on the object, and in Lightning the button may need adding to the list view button layout.

    Data Loader. Salesforce Help says Mass Transfer can’t bulk-transfer opportunities, cases or activities. Export the records with their Owner ID, replace it with the new owner’s User ID, and run an Update. Check the success and error files afterward.

    What goes in a leaver checklist?

    Report first, resolve blockers, transfer, then deactivate. In that order, because the report disappears the moment ownership is gone.

    1. Run the owned-records report for each key object and save the export, with Ids.
    2. Check the blockers: default owners, approvals, workflow email alerts, hierarchy fields.
    3. Move shared-record dependencies: dashboards the person owns or runs as, folders, Chatter groups, files.
    4. Transfer records, with a note of who inherits what and why.
    5. Deactivate (Setup, Users, Edit, clear Active), or freeze if something still blocks you.
    6. Remove permission set assignments and handle default team memberships.
    7. Rerun the report. It should read zero.

    What to do Monday:

    1. Build the inactive-owner report for accounts and opportunities and run it. Write down the count.

    2. Add “run the report, then transfer” to the leaver ticket, ahead of the deactivation step.

    3. Ask who currently owns the dashboards and default assignments in your org.

    Where we come in

    The Data Health Check is our two-week fixed-scope audit, and records owned by inactive users is one of the checks it measures, with the records behind each figure. It reads and recommends. It doesn’t change a single record, so the transfers are yours, or ours as separate work. The 20-minute scorecard gets you a rough number first.

    Questions people ask

    Does deactivating a Salesforce user transfer their records?

    No. Records they own stay with them until you transfer ownership, according to Salesforce Help. Transferring first isn’t required, but Salesforce recommends it when a user owns more than 10,000 records of one object, to avoid performance issues.

    Can you edit a record owned by an inactive user?

    Not by default in every case. Salesforce Help says processes can’t update records owned by inactive users. An admin can enable editing of accounts, opportunities and custom object records with inactive owners through the User Interface setting and a permission, but reassigning is the cleaner fix.

    Is there a limit on Mass Transfer Records?

    Salesforce Help says the tool handles up to 250 records at a time and points to Data Loader for larger transfers. It also covers only accounts, leads, service contracts and custom objects, so opportunities and cases need Data Loader or another route.

    Sources

  • Seven data quality numbers every Salesforce admin should know

    Seven data quality numbers every Salesforce admin should know

    The short answer: The seven Salesforce data quality metrics worth tracking are duplicate rate, key-field completeness, stale-record share, orphaned records, records owned by inactive users, completeness of newly created records, and picklist sprawl. Standard reports can produce all seven. Each one is a share of a total, so you can repeat it every quarter and see which way it is moving.

    Somebody asks whether the data is good. You say “mostly”, because the only evidence you have is the complaints that reached your inbox. A complaint count measures how loud people are, not how clean the data is.

    A number you can repeat does more. It doesn’t need to be precise. It needs to be the same measurement every time, so that “7%, down from 9%” means something. These seven are the ones I would set up first. Every reading below is a rule of thumb, which means a starting point for a conversation and not a standard. No industry body publishes the right figures, and I’m not going to invent some.

    1. What is your duplicate rate?

    Duplicate rate is the share of records that have a twin. It tells you how much history, pipeline and ownership is split across records that should be one.

    To get it, build a summary report on Contacts grouped by Email, sorted by record count, and see how many records sit in groups of two or more. For accounts, a tabular report sorted by name and a hand count of twins in a few stretches of rows gives you an estimate. If you use duplicate rules with the Report action, Salesforce Help describes a custom report type with Duplicate Record Sets as the primary object, which lists the sets your rules found.

    The caveat is that those sets only contain what your matching rules detect. Records the rules never flagged won’t be there, and sets can linger after records stop matching. Rule of thumb: under 2% of accounts with a twin is comfortable, and over 8% is a project.

    2. How complete are your key fields?

    Key-field completeness is the share of records where the fields you rely on actually have a value. It tells you whether routing, segmentation and reporting have anything to work with.

    Choose five fields per object, say Title, Email, Phone, Mailing Country and Lead Source on Contacts. Build a report, note the total, add a filter on one field with the operator “equals” and the value left empty, and note that count. Repeat for each. The share filled, averaged, is your number.

    The hard part is the choice of fields, not the report. Pick the ones a process reads, not the ones that look important. Rule of thumb: 90% or more filled is healthy, and under 70% means people are working around the gaps.

    3. How many records are stale?

    Stale-record share is the portion of records nobody has touched in a long while. It tells you how much of your database is memory and how much is working material.

    Salesforce’s LastActivityDate is a system field that holds the date of the most recent logged task, event, call or email, and standard objects such as Leads, Contacts, Opportunities and Accounts include it in reports. Filter on it being older than your cutoff, say 90 days for leads, and compare with the total. For accounts with no logged activity at all, Salesforce Help describes an Account report with a cross filter that returns accounts without Activities.

    Two limits. LastActivityDate only reflects logged activity, so teams that work records through other channels look quieter than they are. And for shared activities, the date appears only on the primary contact. Cross filters are available in Enterprise, Performance, Unlimited and Developer editions. Rule of thumb: for open leads, under 10% untouched for a quarter is good, and over 40% means the list has become a museum.

    4. How many records are orphaned?

    Orphaned records have lost the parent they should belong to. A contact with no account drops out of account views, roll-ups and most reports.

    For contacts, report on Contacts and filter on Account Name equals blank. For the other direction, accounts with no contacts, use a cross filter with Accounts as the primary object, “without”, and Contacts as the secondary object. Salesforce Help documents WITH and WITHOUT cross filters and gives “contacts without activities” as its own example of the pattern. An account with no contacts isn’t always a mistake (a prospect may be exactly that), so read the list before you act.

    Rule of thumb: under 2% of contacts with no account is fine, and over 10% means something upstream is creating them.

    5. How many records belong to inactive users?

    This number shows how much of your data is sitting with someone who can’t see it, answer it or be routed to. Salesforce Help describes a custom report type on Leads with the Lead Owner’s Active field added, then a filter of “Lead Owner: Active equals False”.

    The same idea works for accounts, opportunities and cases. For an estimate without setup, list the users on the Inactive Users view in Setup, then group your main objects by owner and look for those names. There is a whole post on this one, including what to do about it: records owned by inactive users. Rule of thumb: zero is the target, because every record here is one someone could have reassigned in a minute at the time.

    6. How complete are new records at the moment they are created?

    New-record completeness is key-field completeness measured only on records created recently. It separates a cleanup problem from a process problem.

    Run the report from number 2 again with a Created Date filter for the last 30 days. If the old backlog is patchy but the new records are full, you have cleanup to do and the door is fine. If new records are as patchy as the old ones, cleaning is pointless, because the form, the integration or the import is still letting them in incomplete. The fix there is a required field, a validation rule or a better file template, not a campaign. A load that skips these checks is the usual cause, which is why importing a CSV safely matters upstream of every number here.

    Rule of thumb: new records should score at least as well as your overall figure. If they score worse, the problem is current.

    7. How badly have your picklists sprawled?

    Picklist sprawl is the number of distinct values in use, compared with the values you meant to allow. It tells you whether a grouped report can still be trusted.

    Group a report by the picklist field and count the groups, using the Add group menu in the Lightning report builder. Compare them with the active values in Setup, and look for “Other”, blanks and near-twins like “Tech” and “Technology”. Salesforce’s Help explains why this happens: an unrestricted picklist stops users entering new values in the user interface, but the API, automation and other apps can still add them, while a restricted picklist keeps them out. Admins with permission to edit the field can still add values to either kind.

    Mind that the Lightning builder lets you group by up to two row fields and two column fields, so check one field at a time. Rule of thumb: under 5% of records in odd buckets is fine, and over 20% means the field has stopped being a field.

    What to do Monday:

    1. Pick two objects, Accounts and Leads for most teams, and run numbers 2, 4 and 5 on them.

    2. Write the three figures in a note with today’s date. That is your baseline.

    3. Put a reminder in the calendar for the same day next quarter, and run the same reports.

    Where we come in

    Our 20-minute Salesforce data health scorecard walks through ten checks with standard reports and gives a score, so you can try this today. It is an estimate. If you want every record counted, with the records behind each figure, the Data Health Check is a two-week fixed-scope audit. It measures and recommends, and it doesn’t fix anything.

    Questions people ask

    How do you measure data quality in Salesforce?

    Pick a handful of measurable checks, such as duplicates, completeness, staleness, orphans and ownership, express each as a share of a total, and run them on the same schedule. Standard reports do most of it. The trend matters more than the first reading.

    What is a good data completeness rate in Salesforce?

    There’s no published standard, so treat any figure as a rule of thumb. For fields your processes depend on, 90% or more filled is a reasonable target. Under 70% usually means people are compensating by hand. Measure only the fields something actually reads.

    How often should you check Salesforce data quality?

    Quarterly is enough for most teams, with new-record completeness checked more often if you’ve recently changed a form or import process. Keep the reports saved, so each run is the same measurement as the last.

    Sources

  • Running low on Salesforce data storage: your options before you buy more

    Running low on Salesforce data storage: your options before you buy more

    The short answer: The Salesforce data storage limit for most editions is 10 GB per org plus a per-user amount, and most records count about 2 KB each, whatever they hold. When you run low, check Storage Usage in Setup, find the objects that fill it, and work down this list: delete what’s truly disposable, archive, use Big Objects, move data outside Salesforce, then buy more.

    Storage Usage says 96%. Your account manager has already offered to fix that for you, which is kind of them. Before you say yes, it’s worth asking what’s in there.

    Most orgs don’t run out of space because of today’s data. They run out because of closed cases from 2016, tasks that logged themselves, and emails nobody will reread. Nobody decided to keep them. Nobody decided to delete them either, which is the part that usually needs fixing first.

    How does Salesforce calculate data storage?

    Salesforce counts most records at a flat size of about 2 KB, regardless of how much data the record holds. Salesforce’s article on estimated record sizes says “Most Salesforce records consume approximately 2 KB of data storage,” and adds that these are approximations that vary with custom fields, rich text content and related records.

    The exceptions on that list matter more than the rule:

    • Person Accounts count 4 KB, because each is an Account and a Contact joined together.
    • Campaigns count 8 KB. Campaign Members count 1 KB each, but a large list adds up.
    • Articles count 4 KB, and email messages vary. Salesforce’s example is a 100 KB HTML email taking 100 KB.
    • Quote, Order and Work Order line items each carry their own 2 KB.
    • Files aren’t part of data storage at all. They count separately, at their actual size.

    At 2 KB a record, 10 GB is roughly five million records. That sounds like plenty until you count the activities.

    How much storage does your edition get?

    For data, Salesforce Help lists a minimum of 10 GB per org, plus an amount for each user license. Professional and Enterprise add 20 MB per user, and Unlimited and Performance add 120 MB per user. Essentials and Starter get 10 GB with no per-user addition. Salesforce’s example: a Professional org with 10 users gets 10.2 GB.

    For files, Salesforce lists 10 GB per org for Professional, Enterprise, Unlimited and Performance, and 1 GB for Essentials and Starter. The minimums apply to Salesforce and Salesforce Platform user licenses, and orgs with custom user licenses need to ask Salesforce for exact figures.

    One thing catches people out: Salesforce says file and data storage are calculated asynchronously. Delete a million records and Storage Usage may not move for a while. Don’t delete a second batch because the first looks like it didn’t work.

    How do you see what’s using your storage?

    Open Setup and go to Storage Usage. Salesforce’s file storage article points admins there to see current usage and the major contributors. Salesforce’s record size article refers to the same area as “Monitor Data and Storage Resources”. Read two things:

    1. Which objects hold the most records. Multiply by the sizes above for a rough guide to where the space is.
    2. Whether the problem is data or files. They have separate allocations, and the fixes differ.

    For files, the same article notes that Content Bodies typically contribute the largest portion of file storage. Reduce it by deleting the related ContentDocument and ContentVersion records, not the Content Body directly.

    What usually fills Salesforce data storage?

    Activities, cases, email messages and campaign members, mostly. These are high-volume, low-regret records that pile up quietly, and most of them are never opened after the quarter they closed. That’s my observation, not a Salesforce statistic, so check your own Storage Usage first.

    Email messages are the one to watch. A plain-text message is small, but an HTML one with a signature, logo and disclaimer can be much larger, and a busy case queue generates a lot of them.

    What are your options, in order?

    Work in this order, because each step is easier to reverse and less work to run than the one after it.

    1. Delete what’s truly disposable. Test records, junk leads, spam cases. Salesforce’s deletion article says deleted records stay in the Recycle Bin for 15 days, and that after you empty it, “once purged, they cannot be recovered.” Delete through the UI for small volumes and Data Loader for big ones.

    2. Archive. Move old records out of data storage into somewhere you can still read them. Salesforce has its own product for this, covered below, and some third-party apps do it too. The point of archiving is that the records are kept, with a retention period, not just removed.

    3. Big Objects. Salesforce’s Big Objects guide says they’re built to hold hundreds of millions or billions of records, and describes “historical archive” as a use case. The trade-offs are real: triggers, flows and processes aren’t supported on big objects, they support only object and field permissions rather than sharing rules, and the guide says to build a custom object as a working dataset if you want reports. Salesforce’s allocations page lists big object storage for 1 million records per org on certain editions. In my opinion it’s a developer’s solution, not an admin’s.

    4. External storage. Keep the records, or the files, in a store outside Salesforce and show them through a connector. It moves your retention and security questions to someone else’s system, so know who owns them before you commit.

    5. Buy more. Sometimes it’s the right answer: the data is live, legally needed in the org, and can’t be shrunk. Ask your account team what’s available under your contract. It’s a decision, not a default.

    What is Salesforce’s own archive product?

    Salesforce Archive is Salesforce’s data archiving product. Salesforce’s page says it lets you build policies that offload inactive records, files and attachments from production on a schedule, view archived records in the context of Salesforce records, restore them one at a time or in bulk, and place records on legal hold. The page gives no edition or availability details, and it doesn’t say where archived data is stored, so ask your account team before you plan around it. Salesforce Help also documents an older Archive managed package and a migration path to the app, so check which one you have. We haven’t used either.

    What can go wrong when you delete records?

    Three things, and the last one hurts most. Parents take children with them, links break, and the undo window is short.

    • Cascade deletes. Deleting a parent can delete its children. Salesforce’s recovery article notes that cascade-deleted children don’t show in the Recycle Bin, only the parent does, and that undeleting the parent brings them back. If a child was deleted before its parent, it’s permanently deleted and undeleting the parent won’t return it.
    • Broken links. Records that pointed at what you deleted lose that link. Check what refers to an object before you clear it.
    • No undo after 15 days. After a purge, the recovery article says you can only recreate records from an export, with new IDs, and you then have to relink everything yourself.

    Export what you plan to delete, with Ids, first. Rehearse in a sandbox. Do a small batch.

    What to do Monday:

    1. Open Setup, Storage Usage, and write down the top three objects and whether data or files is closer to the limit. A Data Health Check can do this for you if you’d rather not.

    2. For the biggest object, run a report by age: how many records are older than two years and untouched? Our post on data quality metrics covers staleness.

    3. Ask who would object if those records vanished. If nobody can say, the answer is archive, not delete.

    Where we come in

    Lazarette, by GTM Cloud, is pre-release and isn’t publicly listed yet. It archives old records on a policy, previews every run including the lookups it would clear, keeps archived records readable on the parent record, and restores them. See the Lazarette page.

    Questions people ask

    How much data storage do I get in Salesforce?

    Salesforce Help lists 10 GB per org for most editions, plus a per-user amount: 20 MB per license for Professional and Enterprise, and 120 MB for Unlimited and Performance. Essentials and Starter get 10 GB without a per-user addition. Custom user licenses need confirming with Salesforce. Check your edition’s row before you plan around a number.

    Do files count toward data storage?

    No. Files are tracked separately and consume their actual uploaded size, per Salesforce’s record size article. Data storage counts records at about 2 KB for most objects. Images in Article rich text fields are stored in file storage. Check both allocations in Storage Usage, because the fixes differ.

    Why didn’t my storage drop after I deleted records?

    Salesforce says data and file storage are calculated asynchronously, so large deletions can take time to show. Wait before concluding it failed. Also confirm you deleted the right object, because activities are often the bulk, and check Storage Usage again the next day before you decide anything else.

    Sources