Lookup and Dynamic Array Functions Interview Questions for Advanced Excel

VLOOKUP can only look to the right of the key column, so it cannot return a value from a column on the left. The column index is a hard-coded number that breaks when columns are inserted. It returns only the first match, defaults to approximate match if the last argument is forgotten, and can be slow on big ranges. XLOOKUP or INDEX/MATCH solve all of these problems.

Check that the last argument is FALSE, then compare the two values directly with =A2=Sheet2!A5. Common causes are trailing or non-breaking spaces, one side being a number and the other text (like employee codes), and the lookup value sitting outside the first column of the range. Use LEN to spot hidden characters, and fix with TRIM, CLEAN, VALUE or by making both columns the same data type.

Situation: a monthly incentive sheet used VLOOKUP with the last argument left blank, so it ran an approximate match. Task: I was asked to verify payouts before release. Action: I spotted employees getting a neighbour's target, changed every lookup to exact match with XLOOKUP, and added a count of missing IDs. Result: twelve wrong payouts were corrected before payment, and exact match became our team's standard rule.

INDIRECT converts text into a reference, so =SUM(INDIRECT(B1&"!C2:C100")) sums the sheet whose name is typed in B1. It is useful for summary sheets that switch between months and for dependent dropdowns. The risks are that it is volatile, it breaks silently when sheets are renamed, it does not work with closed external workbooks, and formula auditing tools cannot trace it. Use it sparingly.

In Microsoft 365, a formula can return many values that spill into nearby cells, as with UNIQUE, FILTER, SORT and SEQUENCE. Other formulas can refer to the whole result with the spill operator, for example =COUNTA(E2#). #SPILL! appears when the spill area is blocked by data, the formula sits inside a Table, the area overlaps merged cells, or the result would run past the sheet edge. Clear the blocking cells to fix it.

Create a common key, for example date plus amount plus cheque or UTR number. On each sheet add =IF(COUNTIFS(Other[Key],[@Key])=0,"Missing","Matched") and filter the Missing rows. Duplicated amounts need COUNTIFS on both sides so counts are compared, not just presence. For repeated monthly work, load both into Power Query and use Left Anti and Right Anti merges, which list unmatched rows directly.

Do not look up by name alone, because a duplicate name returns the first match and silently assigns sales to the wrong person. First flag duplicates in the master with COUNTIF(names,name)>1. Then build a composite key, such as name plus branch or name plus email, in both tables and look up on that. Rows still ambiguous should go back to the data owner for an ID rather than be guessed.

Situation: finance and sales reported different revenue for the same quarter. Task: find the gap before the review meeting. Action: I pulled both datasets, built a key on invoice number, and used lookups to list invoices present in only one file. Most of the gap came from credit notes that sales did not deduct and invoices booked on the last day. Result: both teams agreed a single definition and I added a reconciliation tab to the monthly pack.

First confirm everyone who opens the file has Microsoft 365 or Excel 2021, because older versions show #NAME? for XLOOKUP. Save a backup, list the VLOOKUPs using Find with Look in: Formulas, and replace them sheet by sheet, keeping an exact-match default. Compare key totals before and after, document the change in a notes sheet, and keep a version that still uses INDEX/MATCH if external clients need it.

Avoid repeating the same search: run one MATCH in a helper column and use several INDEX formulas on that position. Sort the lookup table and use XLOOKUP binary search (search_mode 2), which is much faster than a linear scan. Replace full-column references with Tables, set calculation to manual while editing, and for regular jobs move the join into Power Query or the Data Model with relationships.

Binary search halves the range repeatedly, assuming the data is sorted. If the lookup column is unsorted, has mixed text and numbers or contains blanks, it can stop at the wrong row and return a plausible but incorrect value with no error. Use it only on data you sort in a controlled step, such as a Power Query output, and test a few values against an exact-match lookup.