Every "questions to ask your agency" article gives you the same bland list. Do you have experience? Can I see your portfolio? What is your timeline? These are fine questions. They are also completely useless for distinguishing a great agency from a mediocre one, because every agency on the planet gives polished answers to expected questions.
The questions that actually matter are the ones that force specificity. They are questions where a good answer looks fundamentally different from a bad one, and where the difference is obvious even if you are not a technical expert.
I have been on both sides of agency evaluations for twenty years — as a brand owner hiring agencies, and as an agency founder being evaluated by brands. These are the ten questions I would ask if I were hiring a Shopify development agency today.
Why the right questions change everything
The agency selection process is designed to favour the agency. They control the pitch. They choose which case studies to show. They put their most impressive person in the room for the sales meeting and their least impressive person on your project afterwards.
Your questions are the only tool you have to break through the performance. The right questions do three things: they reveal technical competence (or the lack of it), they expose how the agency actually works (versus how they say they work), and they test honesty (because agencies that are willing to give uncomfortable truths are the ones you can trust).
For a broader framework on agency evaluation, see our complete agency scorecard guide. This post focuses specifically on the conversation — what to ask, what to listen for, and what the answers really mean.
Question 1: Can I speak to the person who will actually build my store?
Why this question matters
Many agencies operate a sales-and-delivery split. The person who pitches your project is a salesperson or account director. The person who builds your store is a developer you have never met. This is not inherently wrong — it is how many professional services firms operate — but it creates risk.
The risk is that the salesperson makes commitments the development team cannot deliver on. "Of course we can build that custom feature." "Absolutely, we will hit a 95 PageSpeed score." "That timeline is very achievable." These commitments are easy to make in a pitch and hard to deliver in code.
What a good answer sounds like
"Absolutely. Let me schedule a call with our lead developer for your project. They can walk you through our technical approach and answer any questions about how we would build specific features." This answer tells you the agency has already thought about who will work on your project and is confident enough in their team to put them in front of clients.
What a bad answer sounds like
"Our account manager will be your point of contact throughout the project. They will relay any technical questions to the team." This tells you the agency does not want you talking to the technical team, either because the team is not impressive or because it has not been assembled yet. Either way, that is a problem.
Question 2: What is your target mobile PageSpeed score, and how do you achieve it?
Why this question matters
Page speed directly affects conversion rate. Google's research shows that as page load time increases from one second to three seconds, the probability of a bounce increases by 32%. For a Shopify store doing 20,000 sessions per month with a £50 average order value and a 2% conversion rate, a speed-related bounce increase of just 10% costs roughly £24,000 per year in lost revenue.
This is not theoretical. We see it in real data, and we wrote about the full commercial impact in our article on the real cost of a slow Shopify store.
What a good answer sounds like
"We target 90+ on mobile PageSpeed Insights. We achieve this by building lightweight themes with minimal JavaScript, lazy-loading images below the fold, using system fonts where possible, limiting app installations to under eight, and building custom Liquid solutions for functionality that most agencies solve with heavy apps. Our last five builds averaged a 92 mobile score at launch."
Notice the specificity. A number (90+), a methodology (minimal JS, lazy loading, few apps), and evidence (last five builds averaged 92). That is what competence sounds like.
What a bad answer sounds like
"We always aim for the best possible score." This tells you nothing. It is the agency equivalent of "I am a hard worker" on a CV. If they cannot give you a specific target, they do not have one.
Question 3: Walk me through a project that went wrong and what you learnt from it
Why this question matters
Every agency has had projects go wrong. Data migrations that corrupted product data. Launches that had to be rolled back. Features that were specced incorrectly. The question is not whether problems have occurred — it is how the agency handled them and what systems they put in place to prevent recurrence.
What a good answer sounds like
"We had a migration project last year where the client's legacy platform exported product data in a non-standard format. Our initial data mapping missed a set of custom fields, and we did not catch it until UAT. It delayed the launch by two weeks. Since then, we have added a data validation step after every import where the client reviews a sample of fifty products across five categories before we proceed. We also build in a one-week buffer specifically for data-related issues on migration projects."
This answer demonstrates self-awareness, learning, and process improvement. It also demonstrates honesty, which is the trait you most want in a long-term partner.
What a bad answer sounds like
"We have been fortunate — all of our projects have gone smoothly." Either they have not done enough projects to encounter problems, or they are not being honest. Both are disqualifying.
Question 4: How many apps will my store need, and which ones do you typically recommend?
Why this question matters
Shopify's app ecosystem is powerful but dangerous. Every app adds JavaScript to your storefront, increases page load time, and creates a dependency on a third-party vendor. I have audited stores running thirty or more apps where the combined impact added four seconds to page load time. That is not an edge case — it is common.
What a good answer sounds like
"For a store like yours, we would typically use six to eight apps. We build cart functionality, product filtering, and upsell features natively in the theme because those are performance-critical. For reviews, we recommend a lightweight option. For email, Klaviyo. For subscriptions, if needed, we would look at a dedicated solution. Here is why we chose each one." The agency should be able to justify every app and explain what they build custom instead.
What a bad answer sounds like
"We will install whatever apps you need to get the functionality right." This is the equivalent of a builder saying they will use whatever materials are cheapest. It signals an agency that solves problems by adding apps rather than writing code, and the result is always a slow, fragile store.
Question 5: What is included — and excluded — in your quote?
Why this question matters
The most common source of agency disputes is scope misunderstanding. The client assumes data migration is included. The agency assumed it was not. The client expects mobile-first responsive design. The agency built desktop-first and "made it work" on mobile. These are not bad faith issues — they are documentation failures.
As we discussed in our guide on choosing a Shopify agency, transparent pricing is one of the clearest signals of a professional operation.
What a good answer sounds like
"Here is our written scope document. It lists every deliverable, every page template, every integration, and every feature. It also has an explicit exclusions section. Data migration is included for up to 5,000 products. SEO redirect mapping is included. Post-launch training for your team is included — two one-hour sessions. What is not included: ongoing content creation, paid advertising setup, and custom app development beyond what is scoped."
What a bad answer sounds like
"We will build you a custom Shopify store — everything you need is included in the price." Without a written scope, "everything you need" means different things to you and to the agency. This ambiguity always resolves in the agency's favour, because they control the delivery.
Question 6: Who owns the code and the theme after the project is complete?
Why this question matters
This seems like an obvious question, but you would be surprised how many brands do not ask it. Some agencies retain ownership of custom theme code, meaning you are locked into their support services. Others use proprietary frameworks that cannot be maintained by a different developer. Both create dangerous dependencies.
What a good answer sounds like
"You own everything. The theme code, the custom features, the design assets — all of it. Once the project is complete, you can take it to any other developer or agency if you choose to. We use standard Shopify theme architecture and document our code so it is maintainable by any competent Shopify developer."
What a bad answer sounds like
"We use our proprietary framework, so you will need to continue working with us for any changes." This is a lock-in strategy disguised as a feature. Walk away. For more on why we believe in agency relationships without handcuffs, read our piece on why we do not do lock-in contracts.
Question 7: What does post-launch support look like, and what does it cost?
Why this question matters
The first three months after launch are when you discover everything that needs adjusting. Customer behaviour reveals UX issues your testing missed. Real traffic exposes performance bottlenecks. Seasonal events test functionality you could not simulate. You need an agency that will be responsive during this critical period.
What a good answer sounds like
"We include a 30-day post-launch warranty that covers any bugs or issues with the delivered scope at no additional cost. After that, we offer ongoing support packages starting at £1,500 per month for a bank of hours, or you can engage us on a pay-as-you-go basis at our standard hourly rate. Either way, the same team that built your store handles your support — not a separate team."
What a bad answer sounds like
"We can discuss support options after the project is complete." This tells you support is an afterthought, not a core part of their service model. It also gives them leverage — once your store is live and you are dependent on them, the support pricing conversation happens on their terms, not yours.
Question 8: How do you handle SEO during a build or migration?
Why this question matters
A beautiful new Shopify store that tanks your organic rankings is worse than keeping your old store. SEO needs to be considered from the very first day of the project, not bolted on at the end. This is especially critical for migrations, where incorrect redirect mapping can destroy years of organic authority overnight.
What a good answer sounds like
"SEO is built into our process from day one. We start with a crawl of your existing site to map every URL. We create a full redirect map before we write a line of code. We structure the new theme with proper heading hierarchy, schema markup, and semantic HTML. We optimise collection pages with unique content and proper internal linking. We run a pre-launch SEO checklist that covers canonical tags, meta descriptions, image alt text, robots.txt, and sitemap configuration. After launch, we monitor Google Search Console for crawl errors and ranking changes for 30 days."
For a deeper look at why collection pages matter, read our guide on collection page SEO as a growth lever.
What a bad answer sounds like
"We can add SEO to the project as an additional service." If SEO is not integrated into the build process, it will not be done properly. Period.
Question 9: What is your approach to conversion rate?
Why this question matters
This question separates web design agencies from ecommerce agencies. A web design agency will talk about visual design, brand expression, and user experience. An ecommerce agency will talk about add-to-cart rate, cart abandonment, checkout friction, and how specific design decisions affect revenue.
What a good answer sounds like
"We design every page with a conversion objective. Product pages are structured to maximise add-to-cart rate — we use a sticky add-to-cart on mobile, place social proof near the buy button, and keep the critical information above the fold. Cart drawers are designed to encourage upsells without adding friction. We use real data to inform layout decisions, not just design trends. On our last three builds, the average conversion rate improvement was 22% compared to the client's previous store."
What a bad answer sounds like
"We create beautiful, user-friendly designs that reflect your brand." Beautiful and user-friendly are not conversion strategies. They are descriptions of what every design agency claims to deliver. If the agency cannot talk about conversion rate with the same fluency they talk about design, they are not an ecommerce agency.
Question 10: Can I see your contract before the proposal?
Why this question matters
The contract reveals the agency's true priorities. Specifically, look at three things: the payment terms (do they want 100% upfront?), the termination clause (can you leave if the project goes badly?), and the IP ownership clause (who owns the code?).
What a good answer sounds like
"Of course. Here is our standard agreement. Payment is structured in three stages: 30% on signing, 40% at design approval, and 30% on launch. The contract includes a break clause if either party is unsatisfied after the design phase. All intellectual property transfers to you on final payment. We also include a 30-day post-launch support period at no additional charge."
What a bad answer sounds like
"We will send you the contract once you have agreed to proceed." This is a negotiation tactic, not a transparency practice. You should see the contract before you decide, not after. Any agency that makes you agree in principle before showing you the legal terms is not operating in good faith.
Bonus: The question most brands forget to ask
"What will you say no to?"
This is the most revealing question you can ask, and almost nobody asks it. A good agency will tell you what they are not good at, what types of projects they turn down, and what requests they push back on. An agency that says yes to everything is an agency that over-promises and under-delivers.
The best agencies we have worked with — both as clients and as partners — were the ones who told us what they would not do. That willingness to say no is what gave us confidence in their yes.
When we are asked what we say no to, the answer is straightforward. We say no to projects where the brand wants the cheapest possible build rather than the best possible outcome. We say no to scope that cannot be delivered in the timeline the brand needs. We say no to technology choices that would hurt the store's long-term performance, even if the client specifically requests them. And we say no to lock-in contracts, because we believe we should earn the relationship every month.
How to use these questions
Do not ask all ten questions in a single meeting — you will come across as interrogative rather than thorough. Instead, spread them across two conversations. Use the first meeting (questions 1-5) to assess technical competence and process. Use the second meeting (questions 6-10) to assess partnership model and commercial fit.
Take notes on the specificity of each answer. Agencies that give vague answers to specific questions are telling you something important — they either do not have the expertise to answer properly or they do not want to commit to specific standards.
For a structured scoring framework to use alongside these questions, see our agency evaluation scorecard. And for the broader context on what makes a good Shopify agency, start with our guide on how to choose a Shopify agency in the UK.
If you are evaluating agencies and want honest answers to every question on this list, start a conversation with us. We will tell you exactly where we excel, where we are not the best fit, and what we would do differently on your project. No pitch deck. No pressure. Just a straight conversation.
