
Client testimonials & reviews
NovuLabs works with institutions whose software has to answer to a regulator as well as to a user, banks, hospital networks, government agencies, and the platforms that serve them. What clients consistently describe valuing is the engineering judgement applied before a line of code is written, and a system that can still explain its own decisions a year later.
Sample content. The testimonials on this page are illustrative placeholders written to show the format, not statements from real clients. They are here so the page can be reviewed before publication and must be replaced with verified client feedback. A role and a sector is sufficient attribution where an NDA prevents naming the institution. No star ratings, review counts or review structured data appear anywhere on this page, because no client has given any.
What clients say about working with NovuLabs
Clients are not named. Most of this work sits inside institutions bound by confidentiality, and the same convention applies here as on our case studies: the role and the sector are described, the institution is not.
The value showed up in the first two weeks, before any code: they went through our transaction data and told us our customer records were not resolvable to a single entity, which meant screening would generate noise no matter how good the rules were. Entity resolution became phase one. That reordering saved us a false start we had already budgeted for.
Our examiner asked why a specific alert closed the way it did, eleven months after the fact. The system produced the rule version, the data as it stood that day and the analyst’s narrative in one export. That single interaction justified the whole approach to our board.
We had been quoted for a payment switch by three firms. NovuLabs was the only one that asked what our settlement reconciliation looked like before quoting, and the only one whose estimate did not move afterwards.
Client experiences across industries
The constraints differ by regulator, so the feedback does too. Each group links to the service that engagement type maps to.
Banking & Fintech
The value showed up in the first two weeks, before any code: they went through our transaction data and told us our customer records were not resolvable to a single entity, which meant screening would generate noise no matter how good the rules were. Entity resolution became phase one. That reordering saved us a false start we had already budgeted for.
Our examiner asked why a specific alert closed the way it did, eleven months after the fact. The system produced the rule version, the data as it stood that day and the analyst’s narrative in one export. That single interaction justified the whole approach to our board.
We had been quoted for a payment switch by three firms. NovuLabs was the only one that asked what our settlement reconciliation looked like before quoting, and the only one whose estimate did not move afterwards.
Healthcare
The FHIR work was done properly rather than as a translation layer bolted on at the end. When we later connected a second hospital group, integration took days instead of the quarter we had planned for.
They were direct that HIPAA has no certification to buy and that our obligations sat with us, not with them. Every other vendor we spoke to let us keep believing otherwise. That was the point we decided.
Government
Public-sector procurement means the architecture has to survive review by people who will never use it. The documentation they produced alongside the build cleared our security review without a single follow-up round.
The NADRA integration was the part we expected to slip and the part that did not. They had already handled the failure modes we had not thought to ask about, like timeouts, partial matches, and what the citizen sees when verification fails.
Enterprise Software
We came in wanting a custom ERP. They mapped our processes, found that eighty per cent matched a product we could licence, and scoped a build for only the twenty per cent that was genuinely ours. They talked themselves out of most of the contract.
Their engineers joined our sprints and used our review process rather than importing their own. Six months in it is genuinely hard to tell from the commit history which half of the team is which.
AI & Automation
We asked for machine learning on the alert queue. They put a model on prioritisation and kept deterministic rules as the alerts of record, because an examiner needs a reason they can read. It was a better answer than the one we asked for.
SaaS
Multi-tenancy was designed for on day one instead of retrofitted at customer forty. We have onboarded institutional tenants with their own data residency requirements without touching the core.
Support after launch is the part most vendors treat as an afterthought. When the reporting schema changed, they had already tracked it and shipped the update before our compliance team raised the ticket.
Why clients choose NovuLabs
Stated as method rather than as outcome, and each one linked to the page where you can check it. An unlinked claim is only an assertion.
Engineering expertise
The person who takes your first call is an architect who would be accountable for the technical outcome. There is no pre-sales layer, so nothing is promised in a meeting that engineering has not agreed to.
Meet the teamSecurity-first development
Threat modelling, least-privilege data access and audit trails are designed in the first sprint rather than added before a security review. Retrofitting them is how projects lose a quarter.
How we workEnterprise scalability
Systems are designed around which record is authoritative and where it lives, so growth is an infrastructure question rather than a re-platforming one. Multi-tenancy and data residency are decided up front.
Platform architectureRegulatory depth
AML/CFT under the SBP and FMU frameworks, HIPAA and HL7 FHIR, PCI-DSS and scheme certification. Building for an examiner is a different discipline from building for a user, and it is the one we practise.
Industries we serveClear communication
We say what we do not know, and we say when a requirement would be better met by a product you can licence. A meaningful share of our discovery calls end in a scoping correction rather than a proposal.
Common questionsLong-term support
Tiered SLAs from business hours to 24/7 cover, with a named account manager for enterprise clients. On a regulated platform that also means tracking schema and reporting changes on your behalf.
Talk to usWork with NovuLabs
The most useful next step is usually a conversation about constraints rather than requirements. The discovery call runs 45 minutes, carries no commitment, and is taken by an architect who would be accountable for the outcome.
Or read how past engagements were structured in our case studies, see what we build, or check the frequently asked questions.