What questions should we ask before hiring an agency to lead a Lightspeed-to-Shopify migration for a multi-location retail brand?

Why is a generic replatform reference not enough to hire on?

A generic replatform reference is not enough, because a retail point of sale (POS) cutover fails in places a website migration never reaches. An agency that moved thirty stores off Magento moved thirty catalogs. A Lightspeed cutover differs in three ways: the inventory record counts stock on real shelves, some data objects are liabilities, and failures happen at a till in front of customers. Shopify calls its free Store Migration app "best suited for smaller brands" and points multistore retailers to another path (Shopify blog, July 2026). So the screening questions test what only someone who has stood in a store on cutover night can answer. Each one maps to a step in the Lightspeed cutover order and rollback plan.

Which questions test an agency on Lightspeed data and balances?

Three questions test whether an agency knows Lightspeed's data and the balances inside it; ask each before you volunteer the answer:

1. "Which Lightspeed series are we on, and how does that change your plan?" A competent answer names your series and one consequence. Lightspeed X-Series loyalty is a dollar balance (Lightspeed, October 2026). For R-Series, Lightspeed's Loyalty articles count "Loyalty points" (Lightspeed, October 2026), though one comparison article calls Lightspeed Loyalty cash back "in dollars" (Lightspeed, October 2026). Red flag: treating Lightspeed as one product, or saying R-Series cannot export gift cards. It can, from Customers > Gift cards (Lightspeed, October 2026).
2. "Where does our gift card liability land, and who reconciles it?" A competent answer checks every code before promising to keep it, because Shopify accepts a supplied code only if it is 8 to 20 letters and numbers (Shopify developer docs, October 2026). Lightspeed R-Series codes can run to 32 characters (Lightspeed, October 2026). Moving gift card balances intact covers the load and the reconciliation. Red flag: reissuing every code without checking which ones Shopify can keep.
3. "What are our loyalty balances denominated in, and what receives them in Shopify?" A competent answer asks before it names a destination, because Shopify store credit holds dollars (Shopify Help Center, October 2026) and a points app holds points. Where Lightspeed loyalty balances should land weighs both. Red flag: naming a loyalty app first.

Which questions test tax, rollback and month-end sign-off?

Three questions test whether an agency has carried a cutover through tax, register day and the first close:

1. "Who owns the sales tax registrations, and what do you assume about them?" A competent answer says you do, because Shopify's US setup guide tells merchants "you need to contact each of the relevant agencies and register with them" (Shopify Help Center, October 2026). The agency should plan to rebuild tax settings, since the Store Migration app does not import them (Shopify blog, July 2026). Red flag: any answer implying tax settings transfer.
2. "What is the rollback on register day, and what is the abort condition in numbers?" Sales rung up after cutover exist only in Shopify, so rollback means keeping Lightspeed live and read-only, not reversing. A competent answer adds a final export on cutover night and a numeric abort rule. Red flag: "We do a full backup first." A backup is not a rollback when the new transactions are what you would lose.
3. "Who signs off the first month-end close, and are you in the room?" A migration ends when finance closes a month on Shopify with no adjustment nobody can explain. Red flag: a statement of work that ends at go-live.

Which free questions show an agency has not scoped a Lightspeed cutover?

The pilot store, the exclusions list and the reasons to postpone are three questions that cost nothing and expose an agency that has not scoped the work:

• "How many stores do you want in the pilot, and which one?" The right answer is one store, and a busy one that takes many payment types, because a quiet store hides the failures.
• "What are you explicitly not doing?" Every honest scope has an exclusions list, so an agency that cannot produce one has not scoped the work.
• "What would make you tell us to postpone?" The answer should include peak season, an unreconciled gift card book and a catalog restructure in flight, the same conditions under which we would not start a Lightspeed cutover.

Why should no agency price a Lightspeed cutover before seeing your row counts?

No agency can price a Lightspeed cutover honestly before seeing your row counts, because Lightspeed's export ceilings make the work scale with rows, not stores. A standard Lightspeed X-Series customer export stops at 1,000 rows (Lightspeed, October 2026), though a documented link edit has exported 15,000 (Lightspeed, October 2026). Sales history exports stop at 1,000 (Lightspeed) or 10,000 per file (Lightspeed, October 2026), depending on which Lightspeed article you read. So we do not estimate a Lightspeed cutover before we see the series and the counts for gift cards, loyalty balances and customers. An agency that quotes a fixed price first is either carrying a large contingency or about to discover one. Ask how its quote breaks into the line items a migration quote should carry, and whether a paid discovery phase leaves you a data map you own.

When should you not hire an agency at all?

Do not hire an agency when the work is small, when the blocker is a decision, or when peak is less than 10 weeks away: