Short version: a consultant tells you what to do and what it will cost you not to; an agency builds the thing. The reason the Shopify consultant vs Shopify agency question comes up so often in Canada is that the hardest problems here are diagnosis problems, not build problems. Bill 96 compliance, registering for PST in three provinces, keeping orders moving when Canada Post stops β none of those are hard to code. They're hard to scope. And a scoping mistake on any of them costs more than the engineering.
So the line sits roughly here. If you can already write the ticket, hire someone to build it. If you can't, pay for a diagnosis first. Paying an agency to figure out what you want while the clock runs at build rates is the most common way Canadian merchants waste money on Shopify.
What a Shopify consultant actually does
A consultant's output is a document and a decision, not a deployment. In practice, a two-to-three-week engagement produces things like:
- A read on whether your problem is the theme, the app stack, the catalogue structure or the offer. These get confused constantly. A store with a 3.1s LCP and a 0.9% conversion rate usually has two unrelated problems, and fixing the fast one first feels productive while changing nothing.
- A build spec an agency can quote against without a discovery phase. This is the part merchants undervalue. A spec that names the 14 templates, the metafield schema, the translation scope and the three apps being removed turns a CAD 60k range-quote into a CAD 38k fixed one.
- A ruling on platform tier. Whether you need Shopify Plus, or whether Checkout Extensibility on a standard plan plus Functions gets you 90% of it. We tell merchants under roughly CAD 4β5M in revenue to stay off Plus more often than we tell them to move.
- Risk flags. Which of your current setups will break on a Canada Post work stoppage, which product pages are non-compliant in Quebec, where your tax registrations don't match where you're actually collecting.
What a consultant does not do is merge a PR. That matters when you get to the Canadian specifics, because two of the three big ones need a lawyer or an accountant in the room, and the third needs an operations change, not a theme change.
Agency vs consultant vs freelancer: who carries the risk
The honest difference is liability and continuity, not skill. There are freelance Shopify developers in Toronto and Vancouver who are better at Liquid than most agency mid-levels. The trade is what happens on December 26th.
A freelancer is the right call for bounded work with a clear definition of done: a section build, a Klaviyo-to-Shopify data cleanup, a set of Shopify Functions for a tiered discount. Cheapest per hour, and no account management overhead. The risk is availability. Boxing Day is a real peak here, not a footnote to Black Friday, and a solo developer on holiday with family in another time zone is a single point of failure you've chosen to accept.
An agency is for multi-discipline work with a deadline: a migration off Magento or WooCommerce, a bilingual rebuild, an integration with an ERP and a 3PL at the same time. You're buying coverage, a QA pass you didn't do yourself, and someone who answers at 11pm when checkout throws an error. You pay a premium for it, and on a small scope that premium is genuinely not worth it.
A consultant sits before both. Some consultants are independents; some sit inside agencies as a paid diagnostic engagement. Ours does, and we'll say plainly that there's a conflict of interest baked into that β an agency-run audit tends to find agency-sized work. The defence is a fixed fee for the audit and no obligation to use us for the build. If a consultant won't quote the audit separately from the build, that tells you something.
Shopify consultant cost per hour in Canada
What we see quoted in the Canadian market, in CAD, for 2025β26:
- Independent Shopify consultants: roughly 120β250/hr. Toronto rates cluster at the top of that, and senior ex-agency people working solo in Toronto or Montreal will quote 250β350/hr for a short strategic engagement because they're pricing scarcity, not hours.
- Freelance developers: 50β110/hr for competent Liquid and app-config work. Below 40/hr you're usually buying someone who will follow instructions exactly, which is fine if your instructions are good.
- Agency blended rates: 110β220/hr for North American shops, materially lower for offshore delivery teams. A fixed-fee diagnostic is often the better buy: a 10β15 hour audit at a fixed CAD 1,800β3,500 rather than an open-ended retainer.
We publish our own delivery economics openly if you want a baseline to compare against β see what Shopify development actually costs and the Canada page for how we staff Canadian accounts across time zones.
Three Canadian problems a consultant can scope but not ship
Bill 96 and the bilingual storefront
If you have a place of business in Quebec or you're advertising to Quebec consumers, French on your storefront is a legal requirement with quality and prominence conditions attached, not a growth experiment. That reframes the whole project. The decision of what must be translated is a question for a Quebec-qualified lawyer. The decision of how β Shopify Markets with a /fr-ca subfolder, Translate & Adapt for theme and product strings, metafields for the long-form content, hreflang so Google doesn't treat the two as duplicates β is engineering, and it's not especially hard.
The part that breaks is app-generated text. Translate & Adapt won't reach strings hard-coded inside a third-party app's Liquid block or rendered client-side by its own JavaScript. Size charts, subscription widgets, shipping-protection upsells, review summaries. We find untranslated app strings on most first passes of a supposedly bilingual store, and on a meaningful share of those the fix is replacing the app, not configuring it. That's a diagnosis worth paying for before you sign a build contract that assumes the apps stay.
Provincial tax that doesn't match where you're registered
Shopify will calculate GST, HST, QST and PST correctly once you've told it where you're registered. The failure mode is upstream of the platform: merchants collecting BC PST without being registered, or crossing a provincial registration threshold mid-year and not noticing, or treating Quebec's QST as if it stacks the same way HST does. A consultant can map your order data against your registrations and hand you a list of gaps. Then your accountant decides. Any consultant who gives you a firm answer on whether you must register in Saskatchewan is overstepping, and you should discount everything else they told you accordingly.
A Canada Post fallback you can actually switch to
The 2024 work stoppage taught most Canadian D2C brands the same lesson twice: a single carrier rate in checkout is a business continuity risk. The technical work is small β add Purolator or UPS or a regional courier as a second calculated-rate source, set up rate-shopping rules in your shipping app or 3PL, decide whether you eat the margin difference or show the customer a higher price. The hard part is the commercial decision. At what point do you switch? Who signs off? Do you keep a lite-rate account warm year-round for a carrier you use for 4% of volume, and pay the base fees for insurance you hope not to use?
That's a scoping conversation with your ops lead and your CFO. An agency will happily wire up whatever you decide. It cannot make the decision for you, and it shouldn't try.
The arithmetic that makes advice-first worth it
Here's a real shape of problem, with numbers you can check. A Canadian apparel brand with 1,200 published SKUs needs a compliant French storefront. Average product description, 150 words. Translation quoted at CAD 0.18/word for human French-Canadian copy, not machine output.
1,200 Γ 150 = 180,000 words. At 0.18, that's CAD 32,400 in translation alone, before any development.
Now restructure first. Most of those descriptions repeat: fabric, fit, care instructions, sizing, origin. Move the repeating blocks into metafields and shared snippets, leave a genuinely unique 40-word paragraph per product, and you're translating 1,200 Γ 40 = 48,000 words of product copy plus maybe 12,000 words of shared blocks and theme strings. 60,000 words at 0.18 = CAD 10,800.
Saving: CAD 21,600. Cost of the engagement that produced the restructure: a 16-hour scoping and content-architecture pass at CAD 175/hr = CAD 2,800. The metafield migration itself is build work and costs whatever it costs either way, so it drops out of the comparison.
Roughly 7.7x on the diagnosis. Not every engagement returns that, and I'd be lying if I said most do. But this is the class of decision a consultant exists to catch, and an agency that starts building on day one never catches it, because the brief said "translate the site."
Should you run a Shopify audit before hiring an agency?
Yes, if any of these are true: you don't know why your conversion rate is what it is, you've had two agencies fail on the same store, you're about to spend more than about CAD 25k, or you're being quoted wildly different numbers for what you think is the same work. That last one is the clearest signal. A 3x spread in quotes almost always means the brief is ambiguous, and the ambiguity is yours to fix.
Skip the audit if the work is genuinely bounded. "We need Shop Pay Installments messaging on PDPs and a French checkout" doesn't need discovery. It needs a developer for four days.
A useful audit is boring and specific. Core Web Vitals per template, not a sitewide average. Every app scored on the milliseconds of main-thread work it adds against the revenue it produces. Checkout drop-off by province, because Quebec behaves differently. The apps-versus-revenue table is usually the one that changes the conversation; stores accumulate 24 apps and use nine. If speed is the whole of your problem, you may not need a person at all β SwiftStore will scan, fix what's automatable and keep watching the score, and that costs less than an hour of consulting. We'll also do a first-pass review free before anyone quotes anything.
Strategy consultant or implementation partner?
Different jobs, and the titles get used loosely. A strategy consultant works on the commercial model: catalogue and merchandising logic, pricing and bundling, whether cross-border into the US still makes sense now that the low-value de minimis treatment Canadian shippers built their US pricing around has been tightened, and whether you should be quoting duties at checkout instead of surprising customers at the door. Output: a plan and a P&L view.
An implementation partner owns the build, the QA and the post-launch stability. Output: working software and someone accountable for it at 2am on Boxing Day.
Some firms do both. When they do, ask who specifically is doing the strategy work and whether that person stays on through delivery. The failure pattern is a senior strategist in the pitch and a junior on the build, with a spec neither of them fully owns. If you want to test this, ask to talk to the developer who'll actually be writing the Liquid. We staff it the other way around and are happy to put the build lead on the first call β that's part of why we let clients hire the developer directly when a full agency engagement is more than they need.
When to hire a consultant, and when not to
Hire one when the decision is expensive and reversible only at cost: platform migration, a replatform to Plus, a bilingual rebuild, an ERP integration, a catalogue restructure past a few thousand SKUs. Also hire one when you're the bottleneck β a founder doing merchandising, paid media and ops has no capacity to write a build spec, and that's exactly when agencies get handed vague briefs.
Don't hire one to validate a decision you've already made. Don't hire one for a store doing under CAD 500k where the real constraint is traffic or product, because no amount of Shopify architecture fixes a demand problem. And don't hire one if your last three agencies all told you the same thing and you didn't like the answer.
If you're sitting between the two right now, the cheapest next move is to write down the three questions you'd want answered before you'd feel safe signing a build contract. Send them to us. If a 30-minute call answers all three, you didn't need a consultant, and we'll tell you that.