diff --git a/tests/test_plotly_utils/test_utils.py b/tests/test_plotly_utils/test_utils.py
new file mode 100644
index 0000000000..a39b3741ee
--- /dev/null
+++ b/tests/test_plotly_utils/test_utils.py
@@ -0,0 +1,140 @@
+"""
+Tests for the pure string and utility helpers in _plotly_utils.utils.
+
+These functions have no external dependencies and back the "did you mean"
+suggestions and path-parsing used in validator error messages. The expected
+values below were taken from the function docstrings and confirmed against the
+actual output.
+"""
+
+import pytest
+
+from _plotly_utils.utils import (
+    chomp_empty_strings,
+    display_string_positions,
+    find_closest_string,
+    is_skipped_key,
+    levenshtein,
+    split_multichar,
+    split_string_positions,
+    _natural_sort_strings,
+)
+
+
+@pytest.mark.parametrize(
+    "s1,s2,expected",
+    [
+        ("kitten", "sitting", 3),
+        ("", "abc", 3),
+        ("abc", "", 3),
+        ("abc", "abc", 0),
+        ("flaw", "lawn", 2),
+    ],
+)
+def test_levenshtein_distance(s1, s2, expected):
+    assert levenshtein(s1, s2) == expected
+
+
+def test_levenshtein_is_symmetric():
+    assert levenshtein("kitten", "sitting") == levenshtein("sitting", "kitten")
+
+
+def test_find_closest_string_exact_match():
+    assert find_closest_string("range", ["rang", "range", "orange"]) == "range"
+
+
+def test_find_closest_string_picks_nearest():
+    assert find_closest_string("colour", ["color", "colored", "scale"]) == "color"
+
+
+def test_find_closest_string_ties_break_lexicographically():
+    # "ac" and "ad" are both distance 1 from "ab"; the lexicographically
+    # smaller one wins regardless of input order.
+    assert find_closest_string("ab", ["ad", "ac", "xy"]) == "ac"
+
+
+def test_split_multichar_splits_on_every_char():
+    ss = ["a.string[0].with_separators"]
+    assert split_multichar(ss, list(".[]_")) == [
+        "a",
+        "string",
+        "0",
+        "",
+        "with",
+        "separators",
+    ]
+
+
+def test_split_multichar_no_chars_returns_input():
+    assert split_multichar(["abc"], []) == ["abc"]
+
+
+def test_split_string_positions():
+    ss = split_multichar(["a.string[0].with_separators"], list(".[]_"))
+    assert split_string_positions(ss) == [0, 2, 9, 11, 12, 17]
+
+
+def test_display_string_positions_single_index():
+    p = [0, 2, 9, 11, 12, 17]
+    assert display_string_positions(p, 4) == "            ^"
+
+
+def test_display_string_positions_offset_length_char_no_trim():
+    p = [0, 2, 9, 11, 12, 17]
+    assert (
+        display_string_positions(p, 4, offset=1, length=3, char="~", trim=False)
+        == "             ~~~      "
+    )
+
+
+def test_display_string_positions_all_indices():
+    p = [0, 2, 9, 11, 12, 17]
+    assert display_string_positions(p) == "^ ^      ^ ^^    ^"
+
+
+@pytest.mark.parametrize(
+    "strings,expected",
+    [
+        ([], []),
+        ([""], [""]),
+        (["", ""], ["_"]),
+        (["", "", "", ""], ["___"]),
+        (["hey", "", "why", "", "", "whoa", "", ""], ["hey_", "why__", "whoa__"]),
+        (["", "hi", "", "I'm", "bob", "", ""], ["_", "hi_", "I'm", "bob__"]),
+        (["hi", "i'm", "a", "good", "string"], ["hi", "i'm", "a", "good", "string"]),
+    ],
+)
+def test_chomp_empty_strings(strings, expected):
+    assert chomp_empty_strings(strings, "_") == expected
+
+
+def test_chomp_empty_strings_reverse():
+    strings = ["", "hi", "", "I'm", "bob", "", ""]
+    assert chomp_empty_strings(strings, "_", reverse=True) == [
+        "_hi",
+        "_I'm",
+        "bob",
+        "__",
+    ]
+
+
+@pytest.mark.parametrize("key", ["geojson", "layer", "layers", "range"])
+def test_is_skipped_key_true(key):
+    assert is_skipped_key(key) is True
+
+
+@pytest.mark.parametrize("key", ["x", "y", "geo", "Range", ""])
+def test_is_skipped_key_false(key):
+    assert is_skipped_key(key) is False
+
+
+def test_natural_sort_strings_orders_numbers_naturally():
+    assert _natural_sort_strings(["x10", "x2", "x1"]) == ["x1", "x2", "x10"]
+
+
+def test_natural_sort_strings_reverse():
+    assert _natural_sort_strings(["x10", "x2", "x1"], reverse=True) == [
+        "x10",
+        "x2",
+        "x1",
+    ]