Ariox Company

Snapshot

What they do

Ariox connects the systems small and midsize manufacturers run on — ERP, manufacturing execution, cloud and shop-floor platforms — so a manufacturer's data moves between them instead of being re-keyed. Grand Rapids, Michigan.

Engagement

Partner program infrastructure built inside their existing HubSpot: a partner data model, referral capture and an automated commission system.

Focus

Giving a fast-growing partner channel something to run on, without outgrowing the CRM they already had.

Delivery

Co-delivered by Value-First Team and USS.

The relationship

Jun 2025

Discovery and scoping. A new distributor relationship had opened a partner channel, and the shape of it turned out to be more layered than a partner list — several partner types, partners who are more than one type at once, and commission terms that differ per contract.

Jun 2025

Two-month foundational engagement agreed: design the model, prove it in a sandbox, rebuild in production after sign-off. Data organisation first, payout automation deliberately deferred.

Aug 2025

Implementation begins. The decision to create a deal at the moment a partner hands over an opportunity — rather than once it qualifies — makes the channel visible from its earliest stage.

Sep 2025

Commission system designed. A sandbox is stood up in a distinct colour so it can never be mistaken for production, and the multi-tier distributor structure is resolved through association labels rather than hard-coded rates.

Oct 2025

The completed commission system is demonstrated end to end, from an invoice being marked paid through to a commission record with the correct rate applied.

Feb 2026

A full readiness audit of the live portal, run against a request for three partner dashboards — object inventory, partner and deal analysis, association mapping, and a metric-by-metric account of what the data can and cannot yet support.

Where they started

A partner program that had outgrown the place it was kept. A new distributor relationship had opened a channel expected to generate thirty-five to forty viable conversations a month, with several new partner applications arriving on top of that — and nobody whose job it was to manage any of it.

Underneath, the partner program had been bolted onto a CRM built for direct sales. The referral form on their website created deals, but nothing connected a deal back to the partner who had sourced it. Partner classification did not exist as data at all — no type, no tier, no structure. A single pipeline carried partner and direct business together, which meant the partner channel could not be measured as a channel: there was no way to answer what it had contributed, or which partners were actually producing.

Commissions were the sharpest edge. Rates varied by contract rather than following a schedule, they differed depending on which kind of relationship the deal came through, they tapered over the first few years, and they were payable only once the end customer had actually paid. Every one of those calculations was done by hand, then reconciled between an invoicing system and a separate finance system that did not talk to each other. It worked at the volume they had. It would not survive the volume they were expecting.

And it all had to be built inside the HubSpot tier they were already on. The instruction at kickoff was explicit: don't add platform cost before there is something working to justify it.

What we delivered

A finalised partner data model — Partner, Company, Contact, Deal and Commission, with the associations between them, the referral intake sequence, the automation set and the reporting design, each decision recorded alongside the reasoning that produced it.
A Partner object built for how their program actually works — multi-select type, so a partner can be a reseller and a service partner at once, plus code, status, region, industry, program start date and owner, and partner-to-partner associations so a distributor's downstream tier is representable rather than flattened.
Referral capture wired to the model, with a partner-introduction deal stage and a company name required before an opportunity can advance into it — so partner-sourced deals enter early without creating orphans nothing can be reported on.
An automated commission engine. When an invoice is marked paid, it reads the deal, the subscription and the partner relationship, resolves the rate that applies to that specific kind of relationship, and writes a commission record — replacing a calculation that had been reconstructed by hand each quarter.
Renewal logic that reflects the actual agreement — first-year and renewal commissions handled differently, with renewals flagged for confirmation that the partner is genuinely still engaged before a payout is warranted.
A commission pipeline with an audit trail — open, paid or void, with supporting proof attachable to the record, and rate fields restricted to admin access so the numbers driving payouts cannot be edited from the sales floor.
A sandbox-to-production path — built and tested in an isolated environment, visually distinct so it could never be confused with live data, then rebuilt in production after sign-off.
A production readiness audit ahead of partner reporting: a full inventory of the partner and deal estate, the association map, and an honest metric-by-metric account of which dashboard measures the current data can support and which need groundwork first — with the specific properties required and the sequence to get there.

What changed

The partner program became something the business can see. Partners went from being an arrangement held in contracts, spreadsheets and individual memory to structured records in the CRM — classified, associated to the companies and people they work with, and countable. Fifty-nine of them now sit in the system where none did before.

The commission rules stopped living in people's heads. The logic that determines what a partner is owed — which relationship the deal came through, which year of the contract it falls in, whether the customer has actually paid — was written down, encoded and made checkable. That mattered for a specific reason: the people who have to approve these payments were sceptical of automating them, so the system was designed to be audited rather than trusted. Every commission it produces can be traced back to the rule that produced it, with proof attached.

And the unusual shape of their distributor relationship stopped being an obstacle to modelling. Rather than forcing a layered, multi-party structure into a flat partner list, the design carries it — which is what makes it possible to ask, later, which partners and which routes are actually producing.

In a sentence

Ariox's partner program was growing faster than anything tracking it, so we built the partner and commission infrastructure into the HubSpot they already had — turning what a partner is owed from something a person reconstructed by hand into something the system works out the moment the customer pays.