Our Services
These are data-driven, in-house applications. When you run a business, the business runs on data. Sometimes a commercial package fits perfectly — and when it doesn’t, you build your own, and you get significant benefits: the software can grow with your business, and your business grows with the software.
We’ve taken companies from startup through their tenth anniversary. The initial product was rarely what they ended up with. It enabled them to grow into a profitable company — we can’t name names on that one — but every decision made to grow the company was backed by the software. As the company grew, the automation grew with it: productivity improved, more work got done, and it didn’t just dump more work on the founder.
That same idea shows up everywhere — cargo delivery times, payment rates for car parking or rentals, keeping track of orders, delivering content by email and watching open rates. Membership, trade, shipping, ops: all of it needs data-driven analytics.
All systems need to talk to each other. Over the years we’ve dealt with a multitude of them — POS systems that are difficult to work with because there’s little or no documentation, and systems where we’ve had real support from the manufacturers.
Solar panel work is a good example: data integration where we worked with the manufacturer into their client’s stack. Sometimes it’s as simple as someone who’s built a platform on Salesforce and needs it to talk to their accounting system.
What they don’t want is a recurrent, unpredictable cost for something that should just run. They’d rather have a system that works quietly in the background and does the job.
Once the data is together from the various systems, this is where it gets interesting. We’ve always built and designed systems with the idea that the system is a data source itself.
As AI has come along, the user interface and the data structures are data themselves. Giving AI access to that information becomes trivial — and the results have been exponential. Questions about the data, and answers that are relevant to operational needs, become straightforward.
We’re also looking at data inferred by speech — input via speech, particularly for improved performance — and at importing unstructured documents, where AI can turn something messy into something we can use. Where we have a corpus of data, intelligent translation becomes possible too.
One of our early projects was converting off a dBase vendor stack — and legacy live systems more broadly. We don’t have a huge catalogue of flashy migration case studies, but the pattern is real.
Where competitors are locked into a system and that system becomes end-of-life, they often have no choice but to stay stuck on an antiquated stack. There’s nowhere to go that doesn’t mean starting again under someone else’s lock.
Our way forward is open source software that is live every day and keeps growing. We’ve been lucky over the years to pick long-lived projects — systems that stay useful because they can still change.