Andrew Redward — the person behind XBA.
Andrew Redward is an operations and IT systems manager turned independent automation consultant, based in New Zealand. XBA — Extend Business Automation Limited — is his consultancy: Accredo Saturn customisation, systems integration and process automation for small-to-medium businesses that want systems they understand and control.
Personal reply from Andrew within one business day.
What does twenty years inside an ERP actually teach you?
I've spent the last twenty plus years on the inside: running operations and managing the IT systems for a B2B wholesale distributor with branches across the country, with Accredo Saturn as the daily source of truth.
That sentence is on the homepage too, and on its own it sounds like every consultant's biography. What it means in practice is narrower and more useful. I was not brought in to advise on a system and then leave. I was the person the branch manager rang when a price resolved wrongly on a Friday afternoon, the person who wrote the MaxBasic that fixed it, and the person who still had to explain the variance at month-end. Every technical decision I made, I lived with afterwards.
The most valuable thing that teaches is what a customisation costs after it ships. A script that solves today's problem and is never documented is not an asset — it is a liability with a delay fuse, and the fuse burns down to the day the person who wrote it leaves. I have inherited those, and I have written a few early on that I would not write the same way now. It is why every XBA build ends with a written handover rather than a handshake, and why "no vendor lock-in" is a design requirement on this site instead of a slogan.
The second thing it teaches is where automation stops paying. When you own the reporting, you can see exactly what a manual step costs, and often the honest number is small. A ten-minute task done once a week by someone who is going to be at their desk anyway is not a business case. The same task done forty times a day at three branches, with an error rate nobody has measured, usually is. Twenty years of watching which was which is most of what I sell.
The specific ground covered, in the terms a prospect would want to check: MaxBasic event scripting and batch scripts, scripted forms and documents in Form Designer, custom tables and extension fields, special pricing rules, the OData web service, reporting and live dashboards, ERP migration planning, and integrations between the ledger and the systems around it — Microsoft 365, HubSpot, Xero, MYOB. Alongside that, and inseparable from it, the operational side: warehouse process, branch operations, finance process, and managing the people whose work the automation was supposed to make easier.
That last one matters more than the technology. An automation that is technically correct and operationally hostile does not get used; it gets worked around, and the workaround becomes the new manual process. I have watched that happen. It is why XBA's builds keep a person in the loop until the system has earned its confidence, and why the first question in an engagement is about the process rather than the software.
How do I actually work?
In stages small enough to walk away from. An engagement starts with a free process audit, moves to a prototype against a copy of your data in a sandbox, and only then becomes a production build. Each stage is quoted fixed-price before it starts and each one has to earn the next. The approach section on the homepage sets the four stages out in order.
The reason for the shape is not caution for its own sake. It is that the expensive mistakes in this work are made at the start, when everyone is confident about a process nobody has written down yet. A prototype against real data settles arguments that a specification cannot, and it settles them before there is a budget riding on the answer.
Remote-first, working with businesses nationwide. Scoping, prototyping, building and support all run perfectly well remotely; where being on the floor matters — mapping a warehouse process, standing next to someone at a go-live — I'll be on site.
Independent specialist, Accredo partner, or general IT developer — who should you call?
Three different kinds of supplier, and the answer genuinely depends on the job. XBA is the first column, and there are plenty of problems where one of the other two is the better call.
| Independent specialist (XBA) | Your Accredo partner (QSP) | General IT / dev supplier | |
|---|---|---|---|
| Best at | Customisation and integration depth on a system already in place — scripting, forms, pricing rules, API work, the automation around the ledger. | Licensing, upgrades, product support and implementation. The relationship you keep regardless of who does the bespoke work. | Everything outside the ERP: websites, networks, general application development, infrastructure. |
| Where it falls short | One person. Capacity is finite and there is no bench to escalate to, so large parallel programmes are not the right fit. | Bespoke development competes with the support queue, so capacity for one-off customisation work varies. | Rarely has ERP-native depth. The risk is an integration built from the outside that a ledger cannot safely absorb. |
| Cost shape | Fixed price per stage, quoted before the stage starts. Free audit up front. | Typically licence and support fees plus billed project work. | Usually time and materials, with the ERP learning curve on your invoice. |
| Call them when | A process specific to your business is eating the week, and the fix has to live inside or beside Accredo without disturbing the ledger. | Anything touching licensing, version upgrades, or the supported product itself. | The problem is genuinely outside the ERP, or you need capacity across many systems at once. |
| Do they conflict? | No. XBA does not sell licences and does not replace your partner — see the FAQ on partners. | No. Customisation built on top is documented so your partner can read exactly what is installed. | No, provided whoever writes to the ledger understands how it posts. |
// The three are complementary. Most businesses running Saturn end up using at least two.
Where does XBA say no?
Any consultancy that never declines work is selling capacity rather than judgement, so it is worth writing down where this one stops.
When a standard product would do the job. If what you need is ordinary scheduling, ordinary job sheets, ordinary CRM, and you can live with the way a product ships, buying one is cheaper and faster than anything bespoke. Custom is better only when the standard product cannot express the thing that makes your business money. The field service case study works through exactly that decision, including the conditions under which the honest answer was to buy instead.
When the process should change, not the software. Automating a bad process makes it a fast bad process. Some engagements should end at the audit with a recommendation to fix the sequence of work and revisit the software question in six months. That is a real outcome, and the written map you keep is worth having either way.
When the manual step is already bounded. Ten minutes a week is not a business case. If the number does not justify the build, I will say so rather than find a way to make the spreadsheet look expensive.
Where the work belongs to someone else. Licensing, version upgrades and support of the shipped product sit with your Accredo partner, and they should stay there. Nothing XBA builds requires you to change that relationship.
There is a commercial argument for saying all of this out loud, and it is not modesty. Every technical claim on this site is one a prospect will test in the first meeting. A supplier who tells you when not to spend money is easier to believe when they tell you when to.
Want that read on your own process? The audit is free — and yours to keep.
Book a free process auditQuestions about who you'd be dealing with.
Who is Andrew Redward?
An operations and IT systems manager turned independent automation consultant, based in New Zealand, and the founder and sole director of Extend Business Automation Limited. Twenty-plus years running operations and the IT systems for a B2B wholesale distributor with branches across the country, with Accredo Saturn as the daily source of truth: MaxBasic scripting, scripted forms, custom tables, special pricing, the OData API, reporting, migration planning and integration.
Is XBA an Accredo partner or reseller?
No, and that is deliberate. XBA is independent: it sells no Accredo licences, resells no software, and takes no margin on any product it might recommend. Your QSP relationship, licensing and upgrades stay exactly where they are. XBA builds the customisation, integration and automation layer on top and documents it so your partner can read precisely what is installed.
How big is XBA?
One person. Andrew scopes the work, writes the code and answers the phone afterwards, which is the point — but it also means capacity is finite and a large parallel programme is not the right fit. If a job needs a team, the honest answer is to say so at the audit rather than halfway through the build.
Where is XBA based, and does it work outside its own region?
New Zealand-based and remote-first, working with businesses nationwide. Scoping, prototyping, building and support all run perfectly well remotely. Where being on the floor genuinely matters — mapping a warehouse process, standing alongside someone at a go-live — being on site is part of the job.
What's worth reading before you get in touch?
If you are running Accredo Saturn and want to know what can actually be extended, the Accredo Saturn capability page is the technical one — scripting, scripted forms, custom tables, special pricing, the OData API and reporting, with the integration section covering how data gets in and out.
If you would rather see the work than read about the capability, the field service management case study is the longest write-up on the site: the problem, the two-path integration decision and why it was made that way, and an honest account of when a build like it is not worth doing.
And if you already know which process is costing you, skip both and book the free process audit. It is a written opportunity map, it costs nothing, and you keep it whether or not anything gets built.