Systems
One page for every system in our practice. What we configure and take live, what we only connect to, and what LevelOps itself runs on. The vendor links are plain homepages, and we say so rather than dress them up.
Most agency websites carry a wall of vendor logos with no explanation of what the relationship actually is. You cannot tell whether the firm has taken that system live, integrated it once, or simply likes the logo. This page removes that ambiguity for our own list.
We are tool agnostic, and that is a claim we can afford to make because no vendor on this page pays us. We recommend from what we have actually run in a shop or a warehouse, not from a comparison table. When a cheaper system does the job we say so, and when the honest answer is that no software fixes the problem we say that too. A badly sequenced order flow does not improve because it is now badly sequenced inside an ERP.
The page is split on purpose: the systems we configure and take live, the systems we connect to but do not implement, and the stack LevelOps itself runs on. Those are three different claims and they deserve three different headings.
This is the sequence, and it is the same whether the engagement ends in a full ERP or a two week process fix. Nobody gets a platform recommendation before the audit.
We map the current workflow, follow one real order from arrival to invoice, and find where it stalls. A platform named before that step is a guess with a logo on it. A good number of the useful findings in an audit turn out not to be software problems at all.
A four person shop that needs production planning is an MRPeasy or Katana MRP conversation. A business that genuinely needs connected finance, purchasing, manufacturing and sales modules is an Odoo or Business Central conversation. Buying the second when you need the first is the most common and most expensive mistake we see.
Feature lists converge. Projects break at the joins: Shopify to the MRP, the MRP to accounting, the purchase order that arrives as a PDF and never enters any of them. We scope the integrations before the modules, because that is where the risk lives.
Items, bills of materials and routings decide whether go live is calm or ugly. We provide import templates, validate the data in a test environment, and load it only once it survives that. Dirty data migrated cleanly is still dirty data.
We have finished audits by recommending a smaller system than the one the client asked us to install, and by recommending no purchase at all. Neither of those is good for our invoice in the short run. Both are the reason clients come back.
We name the internal owner during design, not at the end. Then role based training and hypercare after go live. If nobody in the building owns the system, it decays back to spreadsheets, and the implementation was theatre.
These are the platforms we configure, populate, integrate and take live. Every entry here is a system where we do the build, not just a connector we point at.
The core of our implementation practice. These systems answer what you have, what you are building and what you need to buy, which is where visibility goes first.
Cloud manufacturing and inventory software for small and mid sized makers. It handles live stock levels, bills of materials, work orders and visual production scheduling, and it connects natively to Shopify, which is why it shows up so often in the shops we work with.
We implement it end to end: items, bills of materials, routing, work centres, data migration and the Shopify sync. It carries more of our delivered manufacturing work than any other system on this page.
Cloud MRP built for small manufacturers and priced per user. Production planning, stock, purchasing, costing and a simple shop floor view. It is lighter than a full ERP and it does not pretend otherwise, which is its main virtue.
We implement it when the shop genuinely needs production planning but the team size and the budget do not justify a bigger system. Three of the engagements in our record run on it.
Inventory and order management, sold as Cin7 Core and Cin7 Omni, aimed at brands selling through several channels at once. Strong on allocation, multi warehouse stock and pick pack ship rather than on manufacturing depth.
One of the platforms our implementation practice covers, usually for distributors whose real problem is stock accuracy across channels rather than production scheduling.
For when the problem is no longer one department, and purchasing, production, inventory and finance have to share one record instead of reconciling four.
Open source, modular ERP. You start with the apps you need, inventory, manufacturing, purchasing, sales, accounting, and add more later. That modularity is also how an Odoo project gets out of hand when nobody is willing to say no to a module.
We implement it when a business truly needs several connected modules rather than one good tool. Our record includes a wholesale bakery rollout covering production and inventory.
Microsoft's mid market ERP, covering finance, inventory, purchasing and fulfilment, with an official Shopify connector. It fits companies already committed to Microsoft 365 across the business, where the licensing and the identity story are already settled.
On our implementation platform list, and the ERP we most often meet in companies whose IT is already Microsoft end to end. We also build order intake in front of it.
Where the order and the relationship live. Usually Shopify holding the catalogue and the accounts, with a CRM in front of it holding the pipeline.
The commerce platform. For a wholesaler it holds the catalogue, the customer accounts, the pricing and the order record, and it is the platform our own product is built natively on.
We configure and migrate stores, then connect them to the ERP or MRP behind them. It is also the only commerce platform PDF to Order runs on, by design rather than by accident.
The enterprise tier of Shopify. It adds the B2B feature set, higher API limits, checkout control and scripting, which is what a merchant running wholesale and direct to consumer on one store actually needs.
LevelOps is a Shopify Plus Partner, listed in the Shopify Partner Directory. We implement Plus for manufacturers and distributors, including one full stack rollout alongside Katana MRP.
Not a separate product. It is the wholesale feature set inside Shopify Plus: companies, locations, buyer permissions, catalogues, price lists, payment terms and net terms on the same store as your direct business.
We configure it, and we are candid about where it stops. Buyers who send purchase orders by email will not log into a portal to place them, which is exactly the gap our own app fills.
CRM and marketing platform. Pipeline, deals, email, forms, meetings and reporting in one place, with a free tier that grows expensive quickly if nobody prunes the hubs you stopped using.
Our own CRM and pipeline system of record, and the CRM we most often connect on client projects. The deal closes in HubSpot, and then the purchase order still arrives as a PDF, which is the handoff we automate.
Vendor links on this page could become partner or referral links later, which would mean a commission if you sign up through them. Today they are plain vendor homepages and earn us nothing. Either way it costs you nothing, it does not change what we recommend, and you buy your licence directly from the vendor. Everything in this section is a system we have configured and taken live.
A list of logos is easy to write. Here is the shape of the work behind ours, so you can judge the recommendation rather than take it.
Katana MRP accounts for most of our delivered manufacturing implementations, across composites, natural cosmetics, artisan candles, pharmacy automation, luxury vehicles and waste sorting robotics. MRPeasy accounts for the next largest group, in robotics, consumer goods and food production. Odoo covers a wholesale bakery. One engagement was a full stack rollout, Shopify Plus and Katana MRP together for a healthcare supplier, with order capture from email and PDF on top.
Two of those records are published with the client named, Waste Robotics and LOKI Coach, because those clients agreed to it. The rest stay unnamed until consent is on file, so the public index is shorter than the practice. Several engagements were operations audits that never became a software project, and those count too, because a recommendation not to buy is still work.
An honest distinction. These are systems we build order intake and integrations in front of, without claiming to configure the system itself. If you are shopping for an implementer for one of these, we are not it, and we will tell you that on the first call.
In each case the system runs the back office and something else has to create the order in the first place. That is where we come in, upstream of the ERP rather than inside it.
Oracle's cloud ERP, and the heaviest system on this page. Finance, inventory, order management, manufacturing and reporting for companies big enough to staff the system properly, usually with a dedicated administrator.
We do not implement NetSuite. We build the order intake in front of it, so a validated Shopify draft order reaches NetSuite through the sync you already run.
SAP's ERP for small and mid sized companies, covering finance, inventory, production and the whole sales order to delivery to invoice document flow. It is normally deployed by a local SAP partner with middleware to the store.
We do not implement B1. We hand a validated order to it through your existing Shopify to B1 integration, so the document flow starts from clean data instead of a retyped PDF.
A Quebec built ERP for custom and made to order manufacturing. Strong on engineering data, quoting, and scheduling one off work, where a mistyped line does not ship the wrong box, it builds the wrong thing.
We do not implement Genius. We handle the emailed purchase orders for parts and repeat items that still get typed in by hand alongside it.
Intuit's small business accounting: invoices, bills, bank reconciliation, sales tax and reporting. It is accounting software rather than an order management system, and treating it as one is a common and expensive mistake in wholesale.
We connect it behind Shopify so an approved order reaches the books cleanly, and we scope the accounting integration on ERP and MRP projects. We do not sell QuickBooks bookkeeping services.
Same disclosure as above: plain vendor homepages, no commission today, and if that changes the disclosure stays. The difference in this section is scope, not quality. We integrate with these systems and we are not their implementation partner.
Not a client offer. This is what we pay for and work in every day, published because a recommendation is easier to weigh when you can see what the person making it lives with. Nothing in this section is a claim that we implement it for anyone.
Our own prospecting and email stack. Contact data stays in these tools rather than in our repositories, a rule we wrote down because it is tempting to break.
Prospecting and contact data platform. Company and person search, enrichment, sequences and dialling, priced on credits that disappear faster than anyone expects.
We build our own outbound lists here. Search first, enrich only the rows we will genuinely call, because enrichment spends credits. Lists stay in Apollo and never enter a repository.
Cold email and sequence tool. Multichannel sequences, deliverability warmup, and per industry variants of the same message so a sequence does not read as a mail merge.
Our own outbound email sequences run here, split by industry. Click tracking from it is deliberately excluded from our referral analytics, because otherwise it flatters itself.
Email marketing and transactional email in one account: lists, templates, automation and an API for form and lifecycle sends, at a price that suits a small team.
Where our own marketing and lifecycle email sends run, and one of the endpoints our website forms can post to.
The systems of record for how we talk to each other and to clients. Each owns a domain, which means our repositories do not hold a second copy of it.
Team messaging: channels, threads, huddles, and an integration surface wide enough that alerts, approvals and automation output all end up there rather than in email.
Our communication system of record, and where our automations post their weekly summaries so a human actually reads them.
Async video and meeting recording. It records calls, transcribes them, and makes the transcripts searchable, with a light CRM layer over companies, contacts and deals on top.
Our meeting recording system of record. Discovery calls are recorded and transcribed there, and the customer language on our pages is pulled from those transcripts rather than invented in a marketing meeting.
Business phone and text messaging for a small team: numbers, calling, SMS, and CRM logging so a call becomes a record instead of a memory.
Our VoIP and SMS system of record. It is how we take and place calls, including the outbound ones.
Collaborative interface design in the browser. Multiplayer files, components, prototypes and developer handoff, with the version history that design files always needed.
Our design system of record. Site and product screens are designed there before anything gets built, and the file is the source of truth over any screenshot pasted into a document.
File storage plus the Docs, Sheets and Slides suite around it. Unremarkable, universal, and the reason most companies never think about where a finished document lives.
Where our finished documents and binaries live, deliberately not in git. Our repositories hold what we authored and revise, not the PDF we sent a client last March.
Our public surface. Worth naming because it explains a limit: the live site is not generated from a repository, so what is published there is the published version.
Visual website builder with a CMS and hosting, aimed at marketing sites that a non developer can edit without filing a ticket. Strong on layout control, weaker the moment you need real application logic.
The system of record for our live website. If a page differs between the site and a repository, the site wins, because that is what a visitor sees.
Website builder that also runs a full domain registrar and DNS service, which is the part most people forget it does.
Where our domains and DNS records are managed. Nothing else, and that is fine.
The part of the stack that makes the rest measurable or automatable. Deliberately boring: a warehouse, a model provider, an editor, and a workflow tool for the gaps.
Google Cloud's serverless data warehouse. SQL over large datasets with no cluster to manage, and native exports from Google Analytics and Google Ads, which is why marketing teams end up in it.
Our analytics system of record. The queries we wrote live in a repository, the data lives here, and raw event tables are never the source for a channel number.
The company behind the Claude models, offering an API alongside coding and agent tooling. The models are strong at reading messy documents and long context, which is the part relevant to our work.
Our AI system of record, and the models behind the internal automations that draft, research and check our own work before a person signs it off.
An AI code editor, plus cloud agents that run tasks against a repository on their own branch. It is a code editor first, which is what keeps the agent output reviewable.
Where we write code, and where the agents that maintain our own repositories run. Every change they make still arrives as a pull request a person reads.
Open source workflow automation. Nodes, triggers, credentials and self hosting, for the integrations no vendor ships and no connector covers.
The integration layer we reach for when two systems have no native connector. We also say plainly that it adds cost and complexity most smaller operations do not need, so it is a last resort rather than a default.
Listed for completeness and for the boundary it draws. Financial data does not enter our repositories, which is a rule rather than an oversight.
Cloud accounting: bank feeds, reconciliation, invoicing, payroll in some regions, and reporting. Popular outside the United States and comfortable for a team with no in house accountant.
Our finance system of record. It owns the books, which is also why no financial figure lives in our repositories.
Receipt and bill capture. It reads invoices and receipts, extracts the fields, and pushes coded transactions into the accounting ledger so nobody keys them in from a shoebox.
How paperwork reaches our own books. We use it internally and we do not hold ourselves out as a Dext implementer.
Plain vendor homepages, no commission today, and if a tracked link is ever added the disclosure stays. Listing a tool here means we use it ourselves. It does not mean we implement it, support it or resell it.
A directory that only lists what a firm sells is marketing. This part is the boundary, and it is more useful to you than another logo would be.
SPS Commerce, TrueCommerce and the retailer mandated networks around them. We do not implement or resell EDI. Our position is that document driven order intake complements EDI rather than replacing it: EDI handles the big trading partners who mandate it, and everything arriving as a PDF or an email is what we handle.
SAP S/4HANA, Oracle Fusion and the Infor family. Our practice is deliberately in cloud systems for small and mid sized manufacturers and distributors. At tier one the right answer is a specialist integrator with a bench, and pretending otherwise would waste your budget.
WooCommerce, BigCommerce, Magento and the rest. We have no implementation record on them and our product is Shopify native by design, so we have nothing useful to tell you about them. That is a limit, not a judgement of the platforms.
We do not publish an implementation record for any shipping platform or warehouse management system, so treat us as neutral there. If a project needs one we will scope the integration around it and let a specialist own the system itself, rather than claim a competence we cannot point at.
Outside our scope entirely. We have no opinion worth paying for on payroll, applicant tracking or field service management, and a recommendation from us would just be a summary of the same review sites you can read yourself.
If a system is not on this page, the default answer is that we have not run it, and we would rather say that than let a discovery call imply otherwise. Ask directly and you will get a direct answer.
Affiliate links are normal on a page like this. Undisclosed ones are not, and they are also a search penalty, so here is exactly how ours work.
How an implementation actually runs, from audit to hypercare.
The system behind most of our delivered manufacturing work.
The delivery record, client by client, including what we cannot name yet.
Tell us what is breaking and which system you are on.
Book the audit. We map your current order flow, follow one real order end to end, and tell you which system fits, including when the answer is the cheaper one or no new software at all.
We will tell you if we are not the right team for your system. That answer costs you one call rather than one quarter.
FAQ
Today, no. Every vendor link here points at the vendor's own homepage and earns us nothing. The page is built so it could carry a tracked referral link later, which is why each outbound link is marked as sponsored in the page code and why the disclosure prints under every directory automatically instead of depending on us remembering it.
No, and the way to check that is the record rather than the promise. Most of our delivered manufacturing work sits on two systems, Katana MRP and MRPeasy, which is a much narrower list than any referral programme would want us to publish. We also tell people when the cheaper system is the right answer, and when the answer is a process fix rather than software.
Sometimes. This list is what we have run, not the outer limit of what we will look at. If your system is not here, tell us what it is and what is breaking. If we are not the right team for it we will say so on the first call, which is cheaper for both of us than finding out in month three.
That is the common case and usually the right instinct. Pulling out a working ERP is expensive and rarely fixes the thing that actually hurts. More often the fix sits in the handoffs around the system: the order that arrives as a PDF and gets retyped, the stock that is right in one place and wrong in another.
Yes, and that is a normal project shape for us. The honest version starts with why, because a migration inherits every bad habit in the data unless items, bills of materials and routings are cleaned and validated in a test environment first. That is slower at the start and much faster at go live.
No. You buy your licences directly from the vendor, in your own name, on your own billing, and you can cancel without asking us. We are paid for the implementation and integration work. That separation means the size of your subscription is not our revenue, so we have no reason to talk you into more seats than your team will use.
Katana MRP carries the most delivered work, followed by MRPeasy, then Odoo, and Shopify Plus alongside Katana on one full stack rollout. The published records live on our projects page, including Waste Robotics and LOKI Coach, which are named because those clients agreed to it. Other engagements stay unnamed until consent is on file, which is why the index is shorter than the practice behind it.
Because a recommendation is worth more when you can see what the person making it lives with every day. It also keeps the page honest. Implementing Katana MRP for a manufacturer and paying for Slack are two very different claims, and listing them together without saying which is which would blur a distinction that matters to a buyer.
Yes. We deliver role based training for purchasing, production, inventory and finance, and we stay on through hypercare after go live. Support runs in English and French. A system nobody inside the company owns drifts back to spreadsheets, so the handover is part of the work rather than an upsell after it.
