A member named Zoë Martínez walks up to the desk. Your cashier types “Zoe Martinez” into constituent search and gets nothing back. They try again, shrug, and create a new record. Now you have two Zoës, one with her membership and one with today’s ticket purchase, and nobody notices until her renewal reminder goes to the wrong one.
Accented and special characters (é, ñ, ü, ç, ø, apostrophes, hyphens) cause this kind of quiet trouble in any database, and Altru is no exception. Below is what to check, how to find people when the search comes up empty, and how to keep accented names from turning into duplicate records.
First, find out how your search actually behaves
Before you write any staff guidance, test your own database. Whether a search for “e” also matches “é” depends on how the underlying database compares text, and we’d rather you confirm it than take anyone’s word for it, including ours. In many setups, upper and lower case match each other but accented and unaccented letters do not.
Run a short test, ideally in a sandbox or on a test constituent you delete afterward:
- Create a test record with an accented last name, for example “Pérez-Test.”
- Search for it typed exactly, with the accent.
- Search again without the accent (“Perez-Test”).
- Search with only the part of the name before the accented letter (“P”) and, if your search supports wildcards, a wildcard in place of the accented letter.
- Repeat the plain and unaccented searches in a constituent query, using both “equal to” and “contains” or “begins with” criteria.
Write down which of these return the record. That small table becomes your staff instructions. If the results surprise you or differ between search and queries, ask Blackbaud support how your instance handles accent matching rather than guessing.
How to find someone when the name won’t match
When the unaccented spelling doesn’t return the record, staff need a fallback that doesn’t involve retyping the name five ways. In rough order of reliability:
- Search by something other than the name. Email address, phone number, ZIP code, or a membership card number or lookup ID are usually faster and avoid the problem entirely. Teach this as the default for returning members.
- Search on the part of the name you’re sure of. First name plus ZIP, or the first few letters of the last name up to (but not including) the accented character, will often surface the right person in a short list.
- Ask the guest how their name is spelled on file. Many people with accented names have had them stripped by other systems for years and won’t be offended. A quick “Is that with an accent?” is normal front-desk practice.
- Type the accent. Staff can enter accented characters directly: on Windows, the on-screen or international keyboard or Alt codes; on a Mac, press and hold the letter to pick the accented version. Post the five or six characters your audience uses most next to the register.
Make the record findable both ways
The long-term fix is to give each person’s record a version of the name that matches however someone is likely to search for it. Options to consider, depending on what your Altru setup offers:
- Alternate names or aliases. If your constituent records support an alias or alternate name, store the unaccented spelling there. Check whether constituent search looks at that field before you rely on it; test it the same way as above.
- Nickname or preferred name fields. Some teams put the unaccented version in a secondary name field. This helps queries even if it doesn’t help the main search, so be clear with staff about which one it helps.
- Keep the legal spelling as the primary name. Don’t “fix” a guest’s name by removing the accent from their record. It’s their name, it shows up on acknowledgments and membership cards, and people notice when it’s wrong.
Expect duplicates, and look for them on purpose
Accent mismatches are one of the more common sources of duplicate constituents, because the failure is silent: search returns nothing, so staff reasonably create a new record. Online purchases add to it, since a guest may type “Jose” on their phone and “José” on a laptop.
Build a constituent query you run monthly that looks for likely pairs: same email or same address with slightly different names. Accent variants tend to share an email, which makes them easy to spot. Merge them with Altru’s normal duplicate-merge process, keeping the accented spelling unless the person tells you otherwise.
Check what happens after the data leaves Altru
Searching is only half of it. Accented names can also break on the way out:
- Exports and CSV files. Opening an export in a spreadsheet can turn “Müller” into garbled characters if the file’s encoding isn’t read as UTF-8. Import the file instead of double-clicking it, and pick UTF-8.
- Email and mailing tools. Send yourself a test with an accented test record before a large mailing. A broken salutation in a renewal letter is more noticeable than a missed search.
- Printed cards and labels. Print one test card for a record with an accent and a hyphen before a big run.
A short version for the front desk
- No match on the name? Search by email, phone, or card number first.
- Still nothing? Search the first name plus ZIP, or the start of the last name.
- Ask how the name is spelled, and type the accent if there is one.
- Only create a new record after all three come up empty.
That list, plus a monthly duplicate check, handles most of the problem regardless of how your search treats accents. The testing step tells you how much of it you actually need.
