Capability · Delivery & operations

Operations is measured by response-time attainment, not by how friendly the hotline sounds

After go-live the test changes to three checkable things: how fast a fault is located, how fast a plan is communicated, how often definitions get re-tested. We write them into terms — 24/7 emergency response, issues located within one hour with a plan attached, one year of free service after acceptance, supported by 15 offices for on-site implementation and enablement.

24/7 emergency responseIssue located within one hourOne year free service after acceptance15 offices across China60%+ engineering headcount

Position

Year one after go-live is the real test period

Platforms decay in the first two years: people move, policies change, filing definitions shift, and the configuration stays where it was. So operations is contracted on checkable lines — response-time attainment, whether the quarterly definition review actually ran, and whether training and the knowledge base reach posts that must operate the system alone.

The engineering claim is auditable too: CMMI Level 3 process management and ISO 9001 govern how we deliver, and the published scale behind them — 200+ staff, 60%+ engineers, 100+ software copyrights — is what sustains maintenance, not just a tender document.

200+

Staff (published basis)

60%+

Engineering share

15offices

Nationwide delivery offices

100+

Software copyrights

Six things happen after go-live, in order, each with a counterpart

Every step names a client-side post. A step without an owner is a step that did not happen.

  1. Cutover and parallel running

    Modules open per the cutover note; anything filing-related runs in parallel for one cycle and both sides are reconciled table by table. Output: cutover record and parallel comparison.

  2. Close migration and settle exceptions

    Exceptions raised during migration are decided by the business — complete them, or retire them with a record. Output: closed migration reconciliation.

  3. Train by role: administrator, business, risk

    Administrators learn workflow and field configuration and permission changes; business and risk learn their own steps and the filing outlet. Attendance and competence are recorded.

  4. Hand over the knowledge base

    Procedure per module and scenario, common exception handling, and definition notes — maintained with each release so new starters can self-serve.

  5. Respond around the clock, graded by impact

    24/7 emergency response with the issue located and a plan communicated within one hour. Filing peaks carry a higher priority class.

  6. Review definitions quarterly, then renew annually

    Each quarter the benefit formulas agreed in the blueprint are recalculated and compared with the prior period. One year of service is free from acceptance; beyond that, scope and response commitments continue under an annual agreement.

Artefacts, timing, acceptance

Implementation and migration close inside the platform range. Operations timing starts at acceptance; one year of free service is a standard contract term, priced annually thereafter.

ArtefactTypical timingAcceptance method
Cutover note and parallel comparisonWithin the 8–16 week buildBoth sides match per table, or differences explained
Migration reconciliation, exceptions closedBefore go-liveEvery exception carries a business decision
Role-based training with recordsAround go-live, in batchesTrained post operates the system unaided and signs off
Knowledge base and runbooksAt go-live, updated per releaseSample scenario resolved from documentation alone
Emergency response and diagnosis24/7, from acceptanceResponse-time attainment; issue located within one hour
Quarterly definition review reportEvery quarter after go-liveRecalculated to agreed formula, with remediation owners

Review content

The quarterly review runs four formulas, not a satisfaction survey

Filing cycle, manual-reconciliation hours, definition-consistency rate and alert confirmation rate are recalculated exactly as agreed at blueprint stage, compared with last quarter, and each deviation attributed to a field or a workflow node. The same report becomes next year’s budget case.

  • Formula unchanged from acceptance
  • Deviation traced to fields
  • Findings shared with three functions
How we define impact

Support base

Six mechanisms that keep year two running

Each has a record form, so it can be checked line by line at the annual agreement review.

24/7 emergency response

Intake at all hours, incidents graded by blast radius, diagnosis and resolution written back to the ticket.

  • Always open
  • Graded handling

One hour to locate and communicate

Within one hour the issue is located and a plan with scope of impact is sent — not an acknowledgement.

  • Timed diagnosis
  • Plan attached

15 offices, delivered locally

Implementation, cutover and training run from the nearest office, which cuts the definition loss of pure remote handover.

  • On-site delivery
  • Local training

One year free after acceptance

Service carries no separate charge for twelve months from acceptance; scope and response terms continue annually.

  • Twelve-month clock
  • Renewable scope

Training and knowledge base

Three post classes with their own procedures; handbooks and scenario steps updated with each release.

  • Role-based
  • Versioned docs

Migration and data duties

Legacy data maps against the dictionary and reconciles per table; unmappable fields go to the business as exceptions.

  • Per-table reconciliation
  • Exception closure

What sits inside the service scope

These are accepted under the same terms during the free year and afterwards. Anything else gets defined at the annual review.

Fault response and diagnosisVersion upgrades and regressionWorkflow and field configuration changesPermission matrix reviewFiling template amendmentsQuality-rule threshold tuningTraining and knowledge-base updatesQuarterly definition review

Duty split

Operations boundaries by hosting mode

Public cloud: we hold runtime and configuration baseline, you hold business definitions and permission approvals. Private cloud: you operate infrastructure, we handle application configuration, upgrades and remediation. Intranet: changes run inside your window and audit perimeter, with assistance only where your network policy allows. Response times and the quarterly review commitment are identical in all three.

  • Boundaries in writing
  • Same response terms
  • Changes reviewable
Security & localised IT

Who should not buy this

Anyone hoping operations removes the need for their own posts: the system administrator and field owners sit with you, and we supply tooling, enablement and review — not the person who answers a definitional challenge. If you only need a one-off data move and will not run the system, per-project data work suits you better than a service agreement.

A standard-edition budget with no definition maintenance should also defer a full implementation.

Questions before signing an operations term

Answers match our contract clauses and can be used in procurement paperwork.

What does 24/7 response actually commit to?
Intake at any hour, incidents graded by impact, the issue located and a plan communicated within one hour. Attainment is recorded as response-time attainment and read at renewal, so the test is arithmetic rather than impression.
What is included in the free year after acceptance?
Fault response, version and configuration baseline maintenance, routine workflow and permission adjustments, filing template changes, training and knowledge-base updates, and the quarterly definition review. New modules or a policy redesign fall under implementation scope and are priced separately.
What happens when a key user leaves?
That is what role-based training and the knowledge base defend against. Administrators, business and risk posts each have their own procedures and recorded competence, so a replacement picks the post up rather than relying on one person’s memory of the system.
How is an intranet supported — do you need to be on site?
Routine operations run inside your window and audit perimeter, with changes and actions logged in your records; the nearest office covers on-site support. Whether someone is resident follows your network policy and operating model, and we price to the actual arrangement.
Who restores data if something breaks?
Per the operations note: database and backup duties sit with you in private cloud and intranet modes, we supply the application-level repair plan and impact assessment, and we reconcile table by table after recovery.
What does year two cost?
The twelve months after acceptance are free. Beyond that an annual agreement covers at least response times, the quarterly definition review and configuration maintenance; extensions of function are scoped and priced against the work-breakdown structure.

Next step

Settle your operating boundary before discussing price

Tell us how IT duties are split today and when your filing peak falls. We return the hosting mode, response arrangement and quarterly review cadence that fits it.