Attach a spreadsheet or document to dBot and it can check your list against the contacts you already have, tidy it up for importing, or read it and answer questions about what is inside. This guide covers all three options and what each one will and will not change.
Attaching a file
Open dBot and click the paperclip icon next to the message box. dBot accepts spreadsheets saved as CSV, PDF files, and Word documents. If your list is in Excel, use Save As and choose CSV first.
Once the file is attached, dBot asks what you would like to do with it. Pick one of the options below. You can also type extra instructions before you send, or just send it as-is.
Which option do you need?
These are three separate things you can do with the same file, not three steps in a row. Choose the one that matches what you are trying to accomplish.
- I have a list of people and I want to know who we already have → Check for duplicates
- My data is messy and I want to get it ready to import → Prepare for import
- I have a file and I want to understand what is in it → Analyze the data or Ask about it
Check for duplicates
Find out who is already in your databank
Attach your list and choose Check for duplicates. dBot reads the names and email addresses in your file and looks each person up against your existing contacts.
How dBot decides someone is a match
A person counts as already in the databank when their record matches on first name, last name, and email address, or on first name, last name, and ZIP code. This is the same rule the import tool uses, so what dBot reports here is what an import would decide about the same list.
Reading the results
- Already in databank: includes a link to each matching contact record so you can open it and check. If someone matches more than one record, dBot lists up to three and tells you how many more there were.
- Not in databank: no existing record was found.
- Could not check: the lookup did not complete for that row. Treat these as unknown rather than new, and try again.
Adding the new people
Everyone with no match is ticked for you. Click the button at the bottom to send just those rows on to the import review screen. You can untick anyone you would rather leave out.
You can also tick someone who did match. Because the importer matches on the same rule, it updates that person's existing record rather than creating a second copy; useful when your file has newer details than the databank does.
Nothing has been saved yet
A duplicate check only reads your contacts. Even after you click to add the new people, they are staged for review, no record is created or changed until you confirm on the import screen.
Prepare for import
Tidy a messy list into something importable
Choose Prepare for import when your data is not yet in the shape the import tool expects. dBot rewrites it into a clean spreadsheet and shows you the result before anything moves.
What dBot fixes
- Splits a full name such as "John A. Smith Jr." into separate title, first, middle, last, and suffix columns.
- Splits a single address line into Address, City, State, and Zip.
- Renames your column headings to the ones the import tool recognizes.
One thing dBot deliberately will not do
It never creates a Databank record ID column, even when your file has something that looks like an ID. Importing against an internal record ID overwrites whatever record currently holds that ID, and an ID from another system is not safe to trust. If you genuinely need to update existing records by ID, set that mapping yourself on the review screen.
Finishing the import
dBot shows the cleaned data and a Continue to Field Mapping button. That opens the normal import review screen, where you check that each column is pointed at the right field and complete the import as usual.
Nothing has been saved yet
Preparing a file only produces a tidied copy for you to look at. Your contacts are untouched until you finish the import on the review screen.
Analyze the data / Ask about it
Read a file and answer questions about it
These are two related options, depending on whether you want an overview or a specific answer.
Analyze the data
Best for spreadsheets. dBot summarizes what the file contains, points out patterns worth noticing, and can produce charts. You can save the analysis as a PDF or Word document.
Ask about it
Best when you have a particular question. Attach the file, type your question, and dBot answers from the contents, useful for long PDFs and reports.
For PDF and Word files, dBot reads the text. It cannot read a scanned page that is really a photograph of a document, because there is no text in the file to read. It reads the first 60 pages of a long PDF.
Nothing has been saved yet
Analysis only reads the file you attached. It never adds anything to your contacts or changes existing records.
Limits worth knowing
- File types: CSV, PDF, and Word (.docx). Save Excel files as CSV first.
- Duplicate check size: Up to 1,000 people per check. If your file is larger, dBot tells you how many it checked so you can split the list and run it again.
- Long PDFs: The first 60 pages are read.
- Permissions: Checking for duplicates needs permission to view contacts. Preparing a file for import needs permission to use the import tool. If an option is missing or refused, ask your Databank administrator.
If something looks wrong
- dBot read fewer rows than your file has. Check the row count dBot reports. For duplicate checks over 1,000 people, split the file and run it in batches.
- Columns landed in the wrong fields. Fix it on the import review screen before you complete the import , that screen is the final word, not dBot's suggestion.
- Someone was marked as new but you know they exist. Matching needs first name, last name, and either an email address or a ZIP code. A record with only an email and no name, or a nickname instead of a legal first name, will not match. Search for them directly to confirm before adding.
- Nothing was read from a PDF. It is most likely a scan. Retype the data into a spreadsheet, or use a copy that has real text.
Comments
0 comments
Please sign in to leave a comment.