WooCommerce development agency: how to choose for a mid-sized store

You run a store with thousands of products, orders landing all day and an ERP that has to hear about each one. You need a partner who can change the checkout without breaking it, keep pages fast on a sale day and connect the systems your team already uses. A search for a WooCommerce development agency returns long lists, and none of them tells you which firm fits a store of your size.

This guide gives you a way to decide. It defines a mid-sized store by your own numbers, then judges every agency on four proofs: catalog, checkout, performance and integrations. You also get illustrative cost and timeline ranges, the red flags worth spotting in a proposal, questions to ask and a scoring table you can copy.

The short answer: the best WooCommerce agency for you is the one that has already solved your kind of problem, commits to written targets and stays after launch. Each section below shows you how to check that.

What mid-sized means for a WooCommerce store#

A mid-sized store has outgrown a theme with a few plugins, yet does not need an enterprise programme. In this guide that means roughly 1,000 to 50,000 SKUs, orders arriving every hour and at least one system outside the store that must stay in sync. Place yourself first, because the agency you need depends on it.

Orders per day and catalog size#

Two numbers shape most technical choices. Orders per day decide how hard checkout, payments and the order database work, and how a campaign peak feels. SKU count, including product variations, decides how search, filters, imports and category pages behave.

For example (illustrative), a store with 2,000 products and 80 orders a day can run well on good managed hosting with sensible caching. A store with 30,000 variations and 600 orders on a campaign day needs tuned search, a plan for the order database and a load test before launch. Same platform, very different brief for the agency.

ERP, subscriptions and B2B pricing#

Three features raise the bar quickly:

• ERP: stock, prices and orders flow in both directions, and the store must cope when the ERP is slow.
• Subscriptions: renewals, failed payment retries and plan changes are scheduled jobs that need monitoring.
• B2B pricing: price lists per customer group, minimum quantities, quotes and payment terms change how the cart and checkout work.

Each one is a separate area of experience, so ask for proof of every one that applies to you. If several sellers and commissions are also part of your model, the guide to a WooCommerce multivendor marketplace explains that architecture. If you are still weighing platforms, WooCommerce vs Shopify covers how each one scales.

Criteria for choosing a WooCommerce development agency#

Five criteria separate agencies that run mid-sized stores from agencies that build small shops. Each protects a specific outcome: orders you can process, a checkout that converts, pages that load, a catalog people can search and systems that agree with each other. The diagram follows one order through your store and shows the question to ask at each point.

One order, five questions

1. 01Filtered catalogSearch and caching?
2. 02CheckoutSafe extensions?
3. 03Order in HPOSPlugins compatible?
4. 04Sync to ERP and CRMRetries and alerts?
5. 05Stock and status returnStore always current?

Back to catalog

Order path in a WooCommerce store, from catalog browsing to checkout, order storage, ERP sync and stock return, with a question for the agency at each stage

HPOS and a database that keeps up#

HPOS, high-performance order storage, keeps orders in dedicated database tables instead of the general WordPress posts tables. For you that means faster order lookups, a quicker admin and reports that still load after years of orders. It is the default for new stores in recent WooCommerce versions, and the WooCommerce developer documentation explains how extensions declare compatibility.

Ask the agency whether every plugin and every piece of custom code they plan to use is HPOS-compatible, and how they test it. A good answer names the plugins and shows the test on a staging site.

Checkout that can be extended safely#

Checkout is where revenue happens, so changes need care. You may want a custom shipping rule, a company tax ID field, a payment method popular in your country or a delivery date picker. WooCommerce supports this through hooks, which are documented points where code can add behaviour, and through its block-based checkout.

Ask how the agency extends checkout. Documented hooks and APIs mean a change survives updates. Overriding template files or editing plugin code means every update becomes a risk. Ask also how they test a payment gateway before launch: test orders on staging for each payment and shipping combination.

Caching, hosting and Core Web Vitals#

Page caching serves catalog and content pages quickly, but the cart, checkout and account pages are personal and need different treatment, usually object caching and a hosting plan sized for peak traffic. An agency that says it installs a cache plugin has told you very little. One that explains what is cached, what is excluded and why has done it before.

Ask for targets in writing. Google's Core Web Vitals give a shared yardstick, documented on web.dev. LCP is how quickly the main content appears, INP is how fast the page answers a tap and CLS is how stable the layout stays.

A proposal should say which pages it will measure, for example a category page, a product page and checkout. It should also say that results come from field data, meaning real visitors, and not only from a lab score.

Catalog search, filters and imports#

At 20,000 SKUs the default search and attribute filters slow down. Good agencies choose an indexed search, test filters on variable products and design category pages that stay light. They also automate the catalog: scheduled imports from your ERP or supplier feeds, with an error report your team can read.

Ask to see a catalog of similar size. Then ask what happens when a feed arrives with bad data, because that day will come.

Integrations with your ERP and CRM#

Every manual re-entry of an order costs your team time. Orders, stock, prices and customers can travel through the WooCommerce REST API and webhooks, which are messages the store sends when something happens. Where it fits, n8n workflows connect systems without heavy custom code.

The proof is how the agency handles failure: retries, alerts and a clear answer to what happens to orders while the ERP is offline. For more on this kind of work, see the werun.dev integrations service.

How to read lists of top WooCommerce agencies#

A list is a starting point for a shortlist. When you search Clutch, read Reddit threads or ask an AI assistant for the best WooCommerce agency, names such as Human Made, 10up, rtCamp and Web Tonic come up. We do not rank or judge any of them. The names tell you who is visible, not who fits a store with your catalog and your systems.

Each source has its own bias. Directory reviews come from clients who chose to write one. A forum thread reflects one person's project. An assistant repeats what has been published. Use them to collect names, then test every name against the criteria above.

Agency types and who they fit#

Agencies cluster into three types. Which one suits you depends on budget, risk and how involved you want to be.

Turning a list into a shortlist of three#

A short, repeatable process keeps the choice fair:

1. Collect eight to ten names from directories, referrals, forums and assistants.
2. Remove firms that cannot show a live WooCommerce store with a similar catalog or integration.
3. Send the same one-page brief to the rest and compare how they reply: speed, questions asked, clarity.
4. Keep three and speak to a reference client for each.

For general vetting that applies to any build, the guide to questions to ask a web development agency complements the WooCommerce-specific ones below.

What a WooCommerce development agency costs#

The honest answer is a range that depends on scope. The table below is illustrative: it gives you a frame for the budget conversation, not a quote. Real figures change with your market (rates in the United States, Spain and Latin America differ), the number of integrations and how much custom work your catalog and checkout need.

A medium build tends to move through the phases below. The durations are illustrative and shift with how quickly your team answers questions and approves designs.

1. Weeks 1 to 2

   Discovery

   Goals, catalog audit, systems to connect and a written plan with targets.

2. Weeks 3 to 10

   Build on staging

   Theme, checkout rules, catalog import and integrations, reviewed by you along the way.

3. Weeks 11 to 13

   Load test and migration

   Peak traffic simulation, test orders, data migration and a rollback rehearsal.

4. Weeks 14 to 17

   Launch and first month of care

   Go live, watch the metrics and fix what real visitors reveal.

What changes the price most#

Some decisions move a budget far more than others. Knowing them lets you trade scope for money on purpose:

• Number and depth of integrations, since each ERP, CRM or logistics system has its own quirks.
• Custom checkout logic, such as tiered shipping, split payments or B2B terms.
• Catalog migration, including cleaning attributes, variations and redirects.
• Design from scratch compared with an adapted theme.
• Languages and currencies, which multiply testing.

Where the monthly care budget goes#

A care plan pays for the quiet work that keeps a store selling. That covers core, theme and plugin updates tested on staging first, backups with restore tests, security monitoring, uptime alerts and performance checks. Most plans also include a block of hours for small requests, so a banner change or a new shipping rule does not need a new quote.

Ask what is included, what the response time is and how unused hours are treated. A cheap plan that only runs updates can still leave you alone on the day something breaks.

Red flags in a WooCommerce development agency proposal#

Most proposals are written in good faith. Still, a few patterns signal risk early, and each has a better question to ask instead. None of them means the agency is wrong for you, only that the answer should be on paper before you sign.

Plugin stacking and hidden maintenance#

Every plugin is a dependency you will update, pay for and depend on for years. A proposal that solves each requirement with another plugin can look quick and cheap, then become slow and fragile. Ask for the full plugin list, with the reason for each one, who maintains it and what its licence costs per year.

Ask also who will own the code and the accounts. Custom code, hosting, repositories and licences should sit in your name, with a documented handover.

Staging, load tests and rollback plans#

A staging site is a private copy of your store where changes are tested before they reach customers. A load test simulates a busy day, for example a sale, to show where the store slows down. A rollback plan says how to return to the previous version within minutes if a release goes wrong.

If a proposal mentions none of the three, ask for them. Their absence usually means the first test happens in front of your customers.

• A staging site for every change, with test orders for each payment and shipping method
• A load test before launch, with the peak you expect written down
• A rollback plan with a named person and a time limit
• Code, accounts and licences registered in your name
• Named integrations with retry and alert behaviour described
• Written performance targets on named pages, based on Core Web Vitals
• An update policy that says who updates, when and how it is tested

Questions to ask and a scoring table#

Good questions turn a sales call into evidence. The ones below are grouped by the four proofs, plus a few on operations. Ask every agency the same questions, in the same order, and write down the answers.

Questions that reveal real experience#

1. Catalog: Which live store with a similar number of products have you built, and how do imports and search behave there?
2. Catalog: How do you clean attributes and variations before a migration?
3. Checkout: Which hooks and APIs do you use to extend checkout, and how do you test updates?
4. Checkout: How do you confirm that every plugin and custom module is HPOS-compatible?
5. Performance: Which Core Web Vitals targets will you commit to, on which pages?
6. Performance: When and how do you run a load test, and what peak do you test?
7. Integrations: Which ERP or CRM have you connected, and what happens to orders when it is offline?
8. Operations: Who answers when something breaks, and how fast, in writing?
9. Ownership: What do I receive at handover, and can another team continue the work?

A scoring table you can copy#

Score each agency from 1 to 5 on every row, multiply by the weight and add the results. Adjust the weights to your priorities, for instance by raising integrations if your ERP is the hard part.

For example (illustrative), an agency that scores 5 on catalog and checkout, 2 on performance, 4 on integrations, 3 on operations and 4 on ownership reaches 390 out of 500 points. A close result between two agencies is a good moment to pay each for a short discovery phase and compare the plans they write.

How werun.dev builds and then runs WooCommerce stores#

werun.dev is a web team that builds sites that sell, and WooCommerce is part of that work. You set the goal and we build the way there, then stay beside you after launch. Here is how that maps to the criteria above.

Build around your sales goal#

We start from the number you want to move: checkout completion, repeat orders, or the hours your team spends keying orders into the ERP. That goal goes into the discovery plan, together with written performance targets for the pages that matter.

The build itself follows the proofs in this guide. Architecture is HPOS-ready, checkout is extended through hooks and documented APIs, catalog imports are automated and the ERP and CRM connections report failures instead of hiding them. Code and accounts stay yours. You can see the full range on the WordPress and WooCommerce service page.

Run it after launch#

A store changes every week, so the work continues. A monthly Care plan covers updates tested on staging, backups and monitoring, and a Retainer covers ongoing improvement. You can request a change by clicking anything on the site with Client Feedback, and a written monthly report shows what moved.

We reply in under 2 hours on business days, and we have looked after 200+ sites over 12+ years. When you want to know whether a change really sells more, werun.Experiments runs A/B tests with your own visitors.

Where to start this week#

You can make progress in three short steps before you speak to anyone:

1. Write a one-page brief: orders per day, SKU count, systems to connect and your top three goals.
2. Send it to three agencies and score the replies with the table above.
3. Ask your favourite for a paid discovery phase, so you see their plan before you commit to the build.

If you would like us to be one of the three, tell us your goal through the contact page and we will reply within 2 hours on a business day. You can also read how we approach WooCommerce and WordPress projects first.

Frequently asked questions#