How to choose a Shopify agency

Choosing a Shopify agency is hard because the proposals you are comparing usually don't describe the same thing. One prices design and build only, another includes integration, a third sells post-launch support separately. Put the numbers side by side and the cheapest-looking one is often the one that ends up costing most.

This piece covers what to look at so you can actually compare them.

Ask this first: who will do the work?

At most agencies the person in the sales meeting is not the person who will build your project. That is normal. The problem is when nobody tells you who will. Ask directly: who does the development, how large is the team, is it in-house or subcontracted, and who is technically accountable for the project?

A clear answer is a good sign regardless of the work itself. An evasive one means either they don't know or they don't want to say, and both are your risk.

Four things to look for in a proposal

• A performance target: is there a measurable speed goal in the contract? "It will be fast" is not a target; "Lighthouse performance 90+ on mobile" is.
• Scope boundaries: does it state what is not included? A good proposal is as clear about exclusions as inclusions.
• A post-launch plan: what happens after go-live? How many days or months of support, and how is it priced after that?
• A handover list: what do you receive at the end? Code, account access, documentation, training.

Warning signs

These are the signals we see most often in the field that turn into problems later:

• Saying yes to everything in the first call. A yes given before the scope is understood comes back as a change request invoice.
• Fixed package pricing. No two e-commerce projects are alike; a fixed package either pushes you outside the scope or puts the agency underwater, and both end badly.
• Dependence on their own closed apps. When you leave, the site has to keep working.
• No documentation commitment. Without a handover you have to go back to them for every small change.
• Unable or unwilling to put you in touch with a reference. A happy client will take the call.

Ask what you take with you when you leave the contract, before you sign it.

Pricing models and when each fits

There are two main models. Fixed price is right where the scope is clear: what will be built is written down, the risk sits with the agency, and you know your budget. Demand a fixed price before the scope is clear and the agency prices in the risk, so you overpay.

Time and materials is right where the scope takes shape as you go: long-running development, continuous improvement, integrations that need discovery. Here the risk sits with you, which makes transparent progress reporting essential.

The healthy approach is usually a combination: a small paid discovery phase first, then a fixed price against the scope that comes out of it. Some agencies do discovery for free; in that case the depth is necessarily limited, so go in knowing that.

Five questions worth asking

• How will you measure performance on this project, and what is the target?
• What happens if something goes wrong on launch day?
• After the project ends, what can our own team manage, and what can't we?
• Can you walk us through the last three projects of a similar size and how long they took?
• Is there a situation where we would not be a good fit for you?

The last question tells you the most. An agency that says it suits every project is either very large or has not yet turned down enough work.