Something like below. I don't have test data, so haven't tested it yet.
Edit: The WHERE conditions on your original query caused the LEFT JOIN to become a de factor INNER JOIN, which threw off the totals. I did more of a query re-write than that to perhaps improve the query overall.
SELECT
B.SYNCHMESSAGE as 'td', '',
MAX(B.row_count) AS [COUNT],
A.IMPORTMESSAGE as 'td', '',
MAX(A.row_count) AS [COUNT]
FROM (
SELECT
, SYNCHMESSAGE, COUNT(*) AS row_count
FROM DataDEV.dbo.SYNCHTERMS
WHERE
DATEADDED >= DATEADD(DAY, DATEDIFF(DAY, 0, GETDATE()), 0) AND
DATEADDED < DATEADD(DAY, DATEDIFF(DAY, 0, GETDATE()) + 1, 0)
GROUP BY
, SYNCHMESSAGE
) AS B
LEFT OUTER JOIN (
SELECT
, IMPORTMESSAGE, COUNT(*) AS row_count
FROM Data.dbo.Data_Errors
WHERE
IMPORTDATE >= DATEADD(DAY, DATEDIFF(DAY, 0, GETDATE()), 0) AND
IMPORTDATE < DATEADD(DAY, DATEDIFF(DAY, 0, GETDATE()) + 1, 0)
GROUP BY
, IMPORTMESSAGE
) AS A ON
A. = B.
GROUP BY
B.SYNCHMESSAGE, A.IMPORTMESSAGE
ORDER BY
B.SYNCHMESSAGE DESC
SQL DBA,SQL Server MVP(07, 08, 09) A socialist is someone who will give you the shirt off *someone else's* back.