The engagements we take
Client names and site details are covered by confidentiality. These are the shapes of engagement we run, and what each typically involves.
Multi-site branch rollout
The situationA head office and a long tail of branches, each connected by whoever happened to be selling in that town. Different last miles, different routers, no common configuration, and an outage that took a day to even locate.
- Physical survey at every location before a single circuit was ordered — power, earthing, cable routes, rack space
- Carrier feasibility run per site, because the best carrier in one district is the worst in another
- One standard branch build: same hardware, same config template, same labelling, same spares list
- Diverse last miles where the site justified it, single-window escalation regardless of which carrier failed
- As-built documentation per site, handed over rather than retained
What changedA branch fault is now identified from the NOC rather than diagnosed on site, and a new branch is built to the same template instead of designed from scratch.
Security stack consolidation
The situationPoint products bought one incident at a time — a firewall from one vendor, endpoint from another, a web filter nobody could remember approving. Overlapping licences, gaps between them, and an auditor asking questions nobody could answer from the console.
- Current-state assessment: what is deployed, what is licensed, what is actually enforcing anything
- Shortlist drawn from competing OEMs we hold in the same category, ranked against the estate rather than a quota
- Phased cutover with the old control left running until the new one proved itself
- Policy written once and applied consistently, not per-device
- Auditor-facing documentation produced as part of delivery, not afterwards
What changedFewer consoles, fewer renewal dates, and a control set the security team can explain to an auditor without preparing for a week.
Exchange to Microsoft 365 migration
The situationAn ageing on-premise mail server, a mailbox estate nobody had audited in years, and a business that could not afford mail to stop for a working day.
- Pre-migration audit — mailbox sizes, shared mailboxes, distribution lists, the forgotten service accounts
- Licence model chosen against actual usage rather than a blanket assignment
- Staged coexistence so both platforms delivered mail during the transition
- Cutover over a weekend, with mobile and desktop profiles handled in advance
- Two weeks of hypercare on site, then handover to the internal team
What changedMail moved without a working day lost, and the estate that arrived on the new platform was the one that had been audited rather than the one that had accumulated.
Server room build
The situationA room that had grown into a server room rather than been built as one — dumped racks, domestic air conditioning, a UPS sized for a previous decade, and no access control worth the name.
- Civil and electrical coordination handled by us rather than left between three contractors
- Racks, containment and structured cabling designed for the load, with room to grow
- UPS and precision cooling sized against measured load, not assumed load
- Suppression, access control and environmental monitoring as part of the same project
- Commissioning tests witnessed, documented and handed over
What changedOne project, one contractor accountable for it, and a room that can be audited and expanded rather than worked around.
Taking over an estate someone else built
The situationAn incumbent leaving, documentation that did not match reality, and a client who could not afford a gap in coverage between the two.
- Discovery and documentation of what is actually deployed, against what the previous provider claimed
- Asset and licence reconciliation — what is owned, what is leased, what has quietly lapsed
- Parallel run with both parties covering, so nothing depended on a clean handover
- Monitoring and alerting rebuilt to our standard before we accepted the SLA
- Steady-state handover to a named service manager with a monthly review from day one
What changedCoverage continued through the transition, and the estate on paper now matches the estate in the rack.
Sensor pilot taken to production
The situationA pilot that worked on a bench and on three sites, with no answer for what happens at two hundred — how the SIMs are managed, who owns the dashboards, and what the security posture is when the devices are on the open internet.
- Pilot scope and success criteria agreed in writing before anything was bought
- Connectivity selected per deployment — SIM where the site moves or has no fixed line, fixed line where it does
- Device management, firmware and certificate lifecycle designed before the rollout, not after
- Platform and dashboards built around the decision being made, not the data available
- Production rollout with our own field team across every site
What changedThe pilot became a rollout with a device management story behind it, rather than a proof of concept nobody could scale.
Purpose-built hardware, with Perfect Electroexim
The situationA deployment where no catalogue device fitted — the environment was too hostile, the enclosure had to meet a standard, or the part in question was no longer manufactured anywhere.
- Requirement translated into a mechanical and electronic specification rather than a wish list
- Design and analysis in SolidWorks and Ansys, with structural and thermal validation before tooling
- Enclosures built to military and mining standards, including explosion-proof where the site required it
- Import substitution where a bought-in component had become unobtainable or unsupportable
- Prototype, field trial, then production — with MRO arranged for the life of the deployment
What changedHardware designed and built in India to the deployment it was going into, with the connectivity and platform delivered by the same integrator.
Collections and reconciliation integration
The situationMoney arriving through several channels and a finance team matching it to invoices by hand, at month end, from spreadsheets exported out of three systems.
- Gateway selection against the actual mix of collections — online, recurring, and offline
- Checkout and subscription integration, with refunds and chargebacks handled rather than ignored
- BBPS integration where recurring bills were being collected
- Automated matching between settlements, bank statements and the ledger
- Exception queue for the cases that genuinely need a person
What changedReconciliation became a daily exception review instead of a month-end reconstruction.
Client names, site counts and commercial terms are covered by confidentiality. These are the engagements we run and what each one actually involves — references are provided directly to evaluators in an active procurement.
Clients we deliver for, every month
Forty clients across payments, broadband, travel, healthcare, hospitality and manufacturing, on monthly engagements. See all 40 clients
Tell us what you are running today
We will tell you what we would change, and what it would cost. No obligation, and a named engineer on the call rather than an account manager.