Most Shopify Plus migration pitches sound the same. The case studies are similar. The methodology slides are interchangeable. The difference between a good partner and one you regret shows up in the answers to specific questions, not the deck. These 12 questions filter for capability, process discipline, and the post-launch commitment most agencies promise but few deliver.
How to use these questions
These aren't trick questions. A good partner will answer each one in detail without hesitation. A weak partner will pivot, generalise, or refer back to a deck. Take notes on the answers and compare across partners. The variance is bigger than most people expect.
Send the questions in writing if possible. Written answers are usually more honest than spoken ones because the partner has time to think rather than improvise.
Question 1: Walk us through your audit process before the build quote
What you're listening for: a structured 1 to 3 week discovery phase that produces a written output, not a verbal estimate. The output should include catalogue inventory, integration list, redirect surface estimate, and a risk register.
What's a yellow flag: "We can give you a quote based on this discovery call." Migration quotes that are written without audits are usually 30 to 50% off from reality.
What's a green flag: "Our audit takes 2 weeks, costs $X, and the deliverable is a build spec, an integration map, and a risk register that names everything undocumented. The audit fee is credited if you proceed."
Question 2: Describe your approach to the 301 redirect map
The single most diagnostic question. The 301 redirect map is the most under-resourced part of every migration and the cause of most ranking losses post-launch.
What you're listening for: a 3-pass process that combines a live site crawl, Google Search Console data for the last 12 months, backlink data, and existing redirect rules from the source platform. Manual validation of pattern rules. Sample crawls post-launch to verify redirects fire.
What's a yellow flag: "We use a redirect plugin and pattern matching." Pattern matching alone misses long-tail URLs that still rank.
What's a green flag: "We crawl the live site with Screaming Frog, merge GSC and backlink data, import existing redirects, then manually validate edge cases. The map is usually 5,000 to 50,000 URLs depending on platform. Every row gets validated against the new sitemap before launch, and we sample 100 to 500 URLs in the first 48 hours post-launch."
Question 3: How do you handle data migration scripts?
Mid-market migrations need custom scripts, not CSV imports. CSV imports lose data above 5,000 SKUs and don't handle complex variant or metafield structures.
What you're listening for: custom scripting against the source platform's database or API, written specifically for your data model, with weekly reconciliation passes against the source until counts match 1:1.
What's a yellow flag: "We use [migration tool]" without explaining when they switch to custom scripts. Off-the-shelf migration tools handle simple cases and break on complex ones.
What's a green flag: "We start with a target source platform script library that handles 70 to 80% of standard cases, then write custom logic for your edge cases. Reconciliation runs weekly with explicit count and field-level checks. We don't ship until 1:1 reconciliation passes."
Question 4: How do you reconcile data before launch?
Reconciliation is the discipline that separates clean migrations from ones that ship with broken filters and missing variants.
What you're listening for: weekly reconciliation passes during the data migration phase. Final reconciliation in the last week before launch. Explicit count checks for products, variants, customers, orders. Field-level checks for critical attributes. Sign-off on the reconciliation report before cutover.
What's a yellow flag: "We do a final QA pass before launch." A single QA pass at the end usually means counts haven't been checked along the way.
What's a green flag: "Reconciliation runs weekly and is treated as a release gate. The final pass before cutover is a written report, not a verbal sign-off. If reconciliation fails, we delay the cutover."
Question 5: How many cutover rehearsals do you run?
The cutover is the live launch window. Rehearsing it on the migration sandbox is what separates quiet launches from chaotic ones.
What you're listening for: at least two end-to-end rehearsals on staging before live launch. Timed runbook with specific people executing specific steps. Failure modes from earlier rehearsals get fixed before the next one.
What's a yellow flag: "We have a launch checklist." A checklist isn't a rehearsal.
What's a green flag: "We rehearse the cutover end-to-end at least twice on staging. The runbook has timed steps with named owners. The first rehearsal usually finds 10 to 20 issues. The second rehearsal should find none. We don't go live until the second rehearsal runs clean."
Question 6: What does your post-launch engagement look like?
The first 30 to 90 days post-launch are where things break and get fixed. A partner who hands off at launch hasn't done many migrations.
What you're listening for: at least 30 days of active engagement post-launch with a named team. A retainer model for months 2 to 6 covering bug fixes, schema validation, ranking monitoring, and edge case handling.
What's a yellow flag: "We provide 14 days of post-launch support." Two weeks is enough for the immediate post-launch issues but misses the second wave that surfaces in weeks 3 to 8.
What's a green flag: "Our default is 30 days of intensive engagement included, then a 6-month retainer at a reduced cadence. The same engineers who built the migration are on the account through day 90."
Question 7: Who specifically will be working on this project?
Pitches are often run by senior people. Builds often happen with junior people. The gap matters.
What you're listening for: named team members with roles and time allocations. Senior engineers in build positions, not just review positions. The pitch team should overlap meaningfully with the build team.
What's a yellow flag: "Our team is staffed based on availability." Translation: you don't know who's actually building this.
What's a green flag: "Here are the named engineers, their roles, and their time allocation per week. You'll meet them in week 1 and they'll be on every weekly review."
Question 8: Show us a similar recent migration
"We've done 50 migrations" is a weaker signal than a specific recent case study at a similar scale.
What you're listening for: a case study of a brand with similar revenue, similar source platform, similar B2B/B2C mix, and similar integration count. Reference customers willing to take a 30-minute call.
What's a yellow flag: "We have a case study book" without a specific match. Generic case studies don't tell you anything about whether they can do your specific migration.
What's a green flag: "We migrated [brand] last quarter from [source] to Plus. Similar revenue, similar B2B mix, similar integration count. Their head of engineering is happy to take a call. Here's the project summary."
Question 9: How do you handle integrations we haven't documented?
Most brands have integrations that nobody fully remembers. Discovery surfaces some. The rest show up during the build.
What you're listening for: an integration audit during discovery that includes interviews with engineering, ops, and customer service. Reasonable contingency in the build quote for 1 to 2 integrations that surface mid-project. A specific process for discovering what nobody documented.
What's a yellow flag: "We'll handle it as a change order." Translation: undocumented integrations will be expensive surprises.
What's a green flag: "Our audit includes interviews with engineering, ops, and customer service to surface integrations that aren't documented. We carry 10 to 15% contingency in the build quote for 1 to 2 integrations that surface mid-project. We'd rather budget for them than be surprised."
Question 10: How do you decide between native Shopify, an app, or custom development?
Most migration decisions land in a triangle: build it natively, install an app, or write custom code. Different partners default to different corners.
What you're listening for: a clear decision framework. Native first when capability exists. Apps for category-leading tools (Klaviyo, Recharge, Yotpo, Algolia, Gorgias). Custom development for differentiated workflows or things that compound recurring app costs above $10K per year.
What's a yellow flag: "We default to custom development for everything." Custom code is technical debt unless there's a clear business reason for it.
What's a green flag: "We default to native, then apps from the leading categories, then custom code only when there's a real business reason. We audit your current app stack during discovery and usually retire 30 to 50% of apps with the migration."
Question 11: How do you handle B2B specifically?
If you have any meaningful B2B revenue, this question matters. Shopify Plus B2B has matured but the workflow translation needs care.
What you're listening for: experience with Shopify B2B at scale, including company hierarchies, locations, payment terms, approval workflows, and catalogue pricing. Specific examples of B2B migrations from comparable platforms.
What's a yellow flag: "Shopify B2B is just like B2C with extra fields." It isn't.
What's a green flag: "We've migrated [N] B2B-heavy brands to Shopify Plus. Company hierarchies, locations, approved buyers, payment terms, catalogue pricing all map to native Plus features. Custom approval workflows usually need Shopify Functions or app-based extensions. Here's a recent B2B case study."
Question 12: What does the worst possible version of this project look like, and how do you prevent it?
The most useful question. A partner who has run a lot of migrations has seen them go badly. They should be able to describe what bad looks like and what they specifically do to prevent it.
What you're listening for: specific failure modes named. Specific mitigations attached. Risk register from past projects that gets updated.
What's a yellow flag: "We don't really have failure modes; our process prevents them." Every partner has had projects go badly. Honesty about it is a quality signal.
What's a green flag: "Three failure modes are most common. Redirect map gaps that surface 6 weeks post-launch and cost rankings (mitigated by GSC data inclusion and post-launch monitoring). Integration discovery that comes in late and pushes timeline (mitigated by a thorough audit and contingency budget). Stakeholder scope creep mid-project (mitigated by a written change order process). Here's how each gets handled if it happens."
What good partners decline
A note worth including: good migration partners decline some projects.
They decline projects where the brand wants a fixed-price quote without an audit. They decline projects where the timeline has been compressed below what's realistic. They decline projects where stakeholder alignment isn't there. They decline projects that combine migration with major content rewrites because diagnosing post-launch issues becomes impossible.
A partner who never declines is a partner who books revenue and lets quality slip. A partner who declines occasionally is a partner who has standards.
Where to start
If you're evaluating Shopify Plus migration partners, the first useful step is sending these 12 questions in writing and comparing answers. The variance between partners is bigger than the variance in case studies or pricing.
For broader context on Shopify Plus migrations, see the complete migration guide. For the cost picture across the full year-one investment, Shopify Plus migration cost breakdown covers the line items in detail.
For deeper coverage of what to expect from a Shopify Plus agency, what a $5M-$100M+ ecommerce brand should expect from a Shopify Plus agency goes into the operational standards. For the operational architecture that makes a migration pay back, why your Shopify Plus store needs an operations strategy covers the layer beyond the storefront.
Or browse the rest of our Platform Migration insights for more on platform-specific guidance, cost analysis, and the decisions that come with a re-platform.
Frequently asked questions
What's the most important question to ask a Shopify Plus migration partner?
Ask how they handle the 301 redirect map. The redirect map is the most under-resourced part of every migration and the cause of most ranking losses post-launch. A good partner can describe a 3-pass process combining live site crawl, GSC data, backlink data, and existing redirect rules in detail. Partners who can't describe this process haven't done enough migrations.
How can I tell if a migration partner has done my source platform before?
Ask for a recent case study with similar revenue, similar source platform, similar B2B/B2C mix, and similar integration count. Generic case studies don't tell you anything specific. A reference customer willing to take a 30-minute call is the strongest signal. "We've done 50 migrations" without a specific match is a weaker signal than one matched recent build.
Should I pay for a migration audit if it's not free?
Yes. Audits that are credited against the build cost if you proceed are a fair model. Audits that are "free" usually mean the partner is racing to a build quote without doing the underlying work. The audit takes 1 to 3 weeks and outputs a build spec, integration map, redirect surface estimate, and a risk register. The cheapest insurance available against scope slippage.
What does good post-launch engagement look like?
At least 30 days of intensive engagement post-launch with a named team, plus a 6-month retainer at reduced cadence. The same engineers who built the migration should be on the account through day 90. Partners who provide only 14 days of post-launch support catch the immediate issues but miss the second wave that surfaces in weeks 3 to 8.
How many cutover rehearsals should a migration partner run?
At least two end-to-end rehearsals on staging before live launch. The runbook should have timed steps with named owners. The first rehearsal usually finds 10 to 20 issues. The second rehearsal should find none. If a partner only has "a launch checklist" rather than rehearsed runbooks, that's the warning sign.
What should I look for in a partner's reconciliation process?
Weekly reconciliation passes during data migration with explicit count and field-level checks for products, variants, customers, and orders. Final reconciliation in the last week before launch as a written report, not a verbal sign-off. If reconciliation fails, the cutover delays. Partners who do "a final QA pass before launch" without weekly checks usually ship with broken filters or missing variants.
Should I work with a Shopify Plus partner for the migration or hire freelancers?
For mid-market migrations ($5M to $100M revenue), a Shopify Plus Partner agency is almost always the right call. Migrations need coordinated workstreams across data, design, engineering, and SEO, plus account management for stakeholder alignment. Freelancers can handle individual workstreams but rarely handle full-project orchestration. The exception is brands with strong in-house engineering who hire freelancers for specific gaps.
How do I evaluate a Shopify Plus partner without any technical background?
Send the 12 questions in writing and compare answers. Written answers are usually more honest than spoken ones because the partner has time to think rather than improvise. Weight the answers heaviest on questions about redirect maps, reconciliation, cutover rehearsals, and failure modes. If a partner answers those four with confidence and specificity, the rest tend to follow.
A Shopify Plus Agency for Strategic Design & Advanced Engineering
Building something ambitious?
- A capable migration partner does five things consistently: real audits, custom data scripts, weekly reconciliation, twice-rehearsed cutover, and 30+ days of post-launch engagement. Each of these maps to a question.
- The single best filter is asking how they handle the 301 redirect map. Partners who can't describe their process in detail haven't done enough migrations.
- Partners who quote without doing an audit are guessing. Partners who quote without seeing your data are guessing twice. The audit shouldn't be optional.
- "We've done 50 Shopify migrations" is a weaker signal than "show us a recent migration of similar size and shape." Ask for the case study, not the count.
- Post-launch retention is where the difference compounds. Ask explicitly: who is on this account in week 12? Most projects need at least one engineer engaged through day 90.
