24/7 service center, outcome-based commitment.

AMS (Application Maintenance Services, the English equivalent of TMA) refers to outsourcing the corrective and evolutionary maintenance of one or several applications. MSP (Managed Service Provider) is the broader English term, also covering infrastructure run, monitoring, end-user support. Managed Service Center is the umbrella term at Access: it is the outcome-based commitment with SLAs, whether on AMS, MSP, or application run. All three models rest on the same principle: outcome commitment (not effort), contractual SLAs, monthly reporting, penalties on miss.

Our service centers cover: (1) Application maintenance (AMS) on ERP, CRM, business applications (corrective and evolutionary); (2) MSP on ServiceNow, Dynamics 365, Power Platform with N2/N3 support; (3) Cloud run on Azure / AWS / GCP with continuous supervision; (4) 24/7 service center across multiple time zones (follow-the-sun); (5) End-user application support with contractual severity levels. Depending on your scope, we structure a dedicated team of 5 to 40 people with an adapted coverage plan.

An SLA (Service Level Agreement) defines measurable outcome commitments: incident acknowledgment time (e.g., 15 minutes on P1), resolution time (e.g., 4 hours on P1, 2 days on P3), application availability rate (e.g., 99.5%), reporting frequency. SLAs include financial penalties when commitments are not met and bonuses when they are exceeded. We propose standard SLAs per application type, but always adjust to your business context. Most of our service centers have 3 to 5 main SLAs + 10-15 secondary performance indicators.

For 24/7 continuous coverage without night-shift duty, we split the team across three zones: Tunis (Europe/Africa — 8am-5pm), Montreal (North America — 8am-5pm EST), and a secondary Paris team for critical European hours. Handovers between teams are structured (30-minute ritual at each zone change, documentation of ongoing incidents). This model ensures there is always a team in business hours working on your platform, with an on-call team for critical incidents outside business hours.

Onboarding a new AMS typically takes 8 to 12 weeks depending on complexity: weeks 1-2 SLA and scope framing, weeks 3-6 takeover of documentation and environments, weeks 7-10 operational transfer with shadowing of the outgoing provider or internal teams, weeks 11-12 SLA cutover with warranty period. For a simple AMS (1 app, <10 people), onboarding can be compressed to 6 weeks. For a multi-app portfolio, count 12 to 16 weeks.

Two main models: (1) Monthly fixed price on framed scope — flat price covering a baseline volume (e.g., 50 incidents/month, 20 evolutions/year), with revision rules if the volume explodes; (2) Fixed + variable — a fixed base for structure (coordinator, leads) + per-activity billing for evolutions. The amount depends on the application scope covered, the expected incident and evolution volumes, and the required SLA level (business hours or 24/7). We price it after the initial diagnostic. SLA penalties are charged against the affected month's fixed fee.

Three main advantages: (1) European time zone — Tunis teams work the same hours as your Paris/Brussels/Geneva teams, no shift like with India; (2) Native French and English bilingualism, simplifying multilingual user support; (3) Cost reduction of 40 to 60% on equivalent onshore AMS day rates, at equivalent quality and certifications. For a 15-20 person service center, the annual savings run in hundreds of thousands of euros vs. an equivalent team in Paris or Montreal.

Transitioning from an outgoing provider is a critical moment we frame in 3 phases: (1) Pre-transition — takeover of documentation, accesses, environments, shadowing the outgoing provider on live tickets; (2) Cutover — taking operational accountability with a warranty period (usually 4-6 weeks) during which SLAs are softened; (3) Steady state — full activation of contractual SLAs. We also stand up a bi-weekly transition committee with your teams to detect and address knowledge-loss risks.

Classic KPIs: application availability (target 99.5 to 99.9%), MTTR (Mean Time To Resolution — target 4h on P1, 2 days on P3), SLA-compliance rate by severity (target 95%), N2 resolution rate without N3 escalation (target 75%), user satisfaction measured quarterly (target NPS > 30), volume of evolutions delivered per month, production quality (zero critical regression). KPIs are reviewed in monthly steering committees and adjusted annually.

Yes, this is a frequent situation. We have taken over several MSPs from other providers with structured transitions. If the outgoing provider cooperates (often a contractual obligation), the transition runs in 8-12 weeks with progressive takeover. If cooperation is difficult (conflict, contentious exit), the transition can extend to 16-20 weeks with a reverse-engineering phase on the existing footprint. In that case, we adjust pricing to cover the extra knowledge-recovery effort.