This guide is for a CTO or head of engineering whose team runs a live product and wants outside help without losing control. It gives an ownership matrix, drift warning signs and the cases where sharing is wrong. For the full support routine Netbase runs, see maintenance and support; for older systems that need upgrading while your team keeps them running, see legacy modernization.
In this guide
- Why do shared-operations arrangements turn into takeovers?
- How should operations be split between the in-house team and the partner?
- Which operating model fits: share, hand over or augment?
- How do you set up shared operations in six steps?
- How do you tell a partner is quietly taking over?
- Where AI fits
- What delivery record exists, and what does not?
- Alternatives: other routes and how to choose
- When not to share operations
- Limits of this guide
- Frequently asked questions
- Next step
Why do shared-operations arrangements turn into takeovers?
Most takeovers are not planned. A partner absorbs overflow while the in-house team works on the roadmap, and each month more knowledge, access and judgement settles on the partner's side. After a year, the in-house engineers can no longer fix production without asking, and the company has outsourced its product by accident.
The cause is usually an unwritten split. "The partner helps with support" does not say who approves a release at 18:00 on a Friday, who holds the database password, or who talks to customers during an outage. When nothing is written, the party that is awake and closest to the problem decides, and that becomes the rule.
The opposite failure is a partner held so far from production that it can only file tickets, which adds a queue instead of capacity. A good split gives the partner real work with written, checked limits.
How should operations be split between the in-house team and the partner?
Split by decision, not by task. The in-house team decides; the partner does a defined share of the work; both are told what the other did. The matrix below is a default to adapt, written for one live web or mobile product.
-
Product roadmap and priorities
- In-house team
- Decides and does
- Remote partner
- Is told; may propose
-
Alert watch during partner hours
- In-house team
- Is told of anything above an agreed severity
- Remote partner
- Does
-
First triage of alerts and tickets
- In-house team
- Decides the severity scale
- Remote partner
- Does, using that scale
-
Production access
- In-house team
- Decides who holds it and reviews it
- Remote partner
- Does work with named, scoped accounts
-
Hotfix for an agreed severity
- In-house team
- Decides if a fix ships now or waits
- Remote partner
- Does the fix, with code review
-
Scheduled patching and library updates
- In-house team
- Decides the window
- Remote partner
- Does, tests and reports
-
Release approval
- In-house team
- Decides
- Remote partner
- Prepares the release and the notes
-
Deployment to production
- In-house team
- Decides who may deploy
- Remote partner
- Does, when named in the runbook
-
Incident command during a major outage
- In-house team
- Decides and leads
- Remote partner
- Does the assigned operational work
-
Customer and stakeholder communication
- In-house team
- Decides and does
- Remote partner
- Supplies facts
-
Runbooks and system documentation
- In-house team
- Reviews and approves
- Remote partner
- Writes and keeps current
-
Third-party vendors, licences and cloud accounts
- In-house team
- Decides and owns contracts
- Remote partner
- Is told; raises renewals and risks
Three lines should never move to the partner without a decision at board or CTO level: who owns the product roadmap, who holds the root credentials and the cloud account, and who leads a major incident. Google's site reliability guidance separates the incident commander from the people doing operational work, so that one person holds the state of the incident while others fix; in a shared model that commander should sit in-house.
Production access follows the same logic. Give the partner named accounts scoped to the duties in the matrix, in line with the least-privilege principle, and review them when the scope changes. A shared admin login is the fastest route to a takeover nobody intended.
Which operating model fits: share, hand over or augment?
| Model | Who owns operations | Main risk | Choose it when |
|---|---|---|---|
| Shared operations (this guide) | In-house team decides; partner does written duties | Drift into partner ownership if the split is not reviewed | You have engineers who know the product and want more cover, not less control |
| Fully managed support | Partner runs the platform end to end | Knowledge leaves the company | You have no in-house engineers, or the product is stable and not core |
| Staff augmentation | Individuals join your team under your process | You carry all management load | You need hands for your own sprints, not an operations routine |
| Staged handover to a partner | Moves from you to the partner on a plan | Gaps during transition | You have decided to stop running it yourself |
If the decision is a handover rather than a share, follow a staged plan such as the transition and knowledge-transfer framework instead of letting it happen ticket by ticket. If the main worry is out-of-hours alerts for a small team, the guide to on-call and incident response for small teams covers rota design first.
How do you set up shared operations in six steps?
-
Inventory what runs
List the applications, environments, integrations, scheduled jobs and accounts the in-house team operates today, with the person who knows each one.
-
Write the split
Fill in the ownership matrix for every activity, including the severity scale, and put it in the support scope so it is part of the contract, not a slide.
-
Grant scoped access
Create named accounts for partner engineers with the rights their duties need, log their actions and set a date to review the list.
-
Run shadow weeks
For the first weeks the partner works alongside the in-house person on each duty before doing it alone, and writes the runbook as it goes.
-
Keep in-house engineers on the rota
Google's on-call guidance warns that knowledge gaps appear when engineers are out of touch with production; keep at least one in-house engineer in every rotation so the skill stays in the company.
-
Review the split on a cadence
In the weekly review, check what the partner changed, which decisions were escalated and whether any duty crept across the line without a decision.
How do you tell a partner is quietly taking over?
Watch for these signals and correct each one as soon as it appears:
- Only the partner can deploy or restore. Correct it by scheduling an in-house engineer to run the next release and the next restore test.
- Credentials sit only with the partner. Move root and cloud-account ownership back to a company account and give the partner scoped access.
- Fixes arrive without documentation. Make an updated runbook or change note part of the definition of done for every fix.
- The roadmap follows the ticket queue. Put priorities back in the in-house product review and let the partner propose, not decide.
- In-house engineers stop joining incidents. Re-add them to the rota and to post-incident reviews.
- Nobody can say what the partner does. Re-read the matrix in the next review and update it.
If a full handover is the right call after all, follow the audit, stabilise and upgrade sequence.
Where AI fits
AI helps both teams read the same production picture faster. A model can group related alerts, summarise logs around an error, draft a triage note in the agreed severity scale and draft the runbook entry after a fix. Engineers still decide severity, review every change before it ships and confirm that a fix worked, and AI tools get read access to logs and monitoring data only. Netbase works with the major commercial and open-source AI tools and models, chosen per project.
- Available capability
AI-assisted triage under human review
AI-assisted notes and changes pass the same code review and approval as any other change, and the in-house team keeps the final decision; not yet tied to a published shared-operations case.
What delivery record exists, and what does not?
Netbase runs, maintains and upgrades what it builds, remotely and alongside the client's in-house team, under an agreed support scope, and response targets are set per contract. Governance runs on weekly reviews, KPI dashboards, a dedicated account manager and a project manager, with Slack or Zoom and client dashboards for day-to-day status, as set out in how Netbase works. Ongoing support and co-building is the sixth and last step of its delivery lifecycle.
Netbase also operates its own products. Cloodo, one of the Netbase Business Divisions, is an AI-powered digital workplace that connects internal staff and outsourced specialists in one hybrid workspace; see the Cloodo record. Support work follows Netbase's security practices, including secure code review and version control, role-based access control and MFA for admin dashboards, with contributors under NDA.
What does not exist: no published client case of a shared-operations split, no incident count, response time, uptime figure or retention figure. Most Netbase projects are delivered on fixed-price contracts; milestone-based, monthly team retainer and KPI-linked models are also offered, with terms set per scope.
Alternatives: other routes and how to choose
- Hire more in-house engineers. Best when operations is core to the product and you can recruit and keep the skills; slowest to start.
- Fully managed support from a partner. Best when no one in-house knows the system and the product is not where you compete; see application maintenance and support.
- Staff augmentation. Best when you need extra people inside your own process and have the management time to direct them.
- Shared operations. Best when you have a capable core team, need more cover or specialist depth, and want the product knowledge to stay in-house.
Selection criteria: who must hold the product knowledge in three years, how many hours of cover you need beyond your team's day, whether you can review a partner's work each week, and how much of your engineers' time operations may take before it hurts the roadmap.
When not to share operations
- No in-house owner. If nobody in the company can review the partner's work, the share is a takeover in name only; choose managed support with clear reporting instead.
- A system you plan to retire within months. The setup cost of a split rarely pays back; keep it as it is or hand it over.
Limits of this guide
- The matrix is a default for one live product; regulated systems may need a stricter split set by your compliance owner.
- No figure in this guide comes from a Netbase client case.
Plan the next step with a Netbase consultant
Frequently asked questions
Yes. Give the partner named accounts scoped to the duties in the written split, log their actions and review the list when the scope changes; root credentials and the cloud account stay with the company.
The in-house team should lead a major incident and own customer communication. The partner does the operational work it is assigned and reports what it changed, so one person holds the full picture.
Keep at least one in-house engineer in every on-call rotation, rotate releases and restore tests between both teams, and include in-house staff in every post-incident review.
No. Netbase works alongside your in-house team under an agreed support scope; your team keeps the decisions, and response targets are set in the contract.
Next step
Tell us what your team runs today and where it needs cover, and we will propose a written split: book a solution review.
Related services and solutions
AI-assisted legacy modernization that keeps your orders and data safe Netbase provides legacy application modernization services for commerce and operations teams to move ageing platforms, such as Magento 1 stores, onto supported technology without losing orders, customers or data. AI speeds up code analysis and data mapping; every migration runs on a written, human-approved plan with rehearsal, staged cut-over and rollback, proven by Netztech's move to Magento 2.
Learn More
Application maintenance and support for live web, mobile and business software Netbase maintains and supports software that is already live: web platforms, online stores, mobile apps, ERP and CRM systems and internal tools, including software another vendor built. We take it over after a code and environment review, then fix defects, keep frameworks current, monitor it, make small changes and report monthly, under service levels agreed in your contract.
Learn More
SaaS product accelerator: launch an AI-ready SaaS on proven Netbase modules A SaaS product accelerator is a set of reusable Netbase modules for accounts, billing, roles and integrations that helps founders and product teams launch subscription software faster, with room for in-product AI from the first release. Reusing these modules can shorten development time, and the approach is proven on Printcart, Netbase's own web-to-print SaaS.
Learn More
Work with Netbase remotely
Every Netbase engagement is delivered remotely, so clients in any country and any industry can work with us; Netbase has worked with clients in 80 countries. Contracts and invoices are issued by Netbase JSC, and payment is made by bank transfer to Netbase JSC's company account or PayPal Business. Talk to the CEO, Huy Nguyen (David), on WhatsApp or by email.
Discuss a project
Netbase JSC helps organizations design, build, modernize, and operate digital products and AI-enabled business systems.project@netbasejsc.com
+84 937 869 689
91 Nguyen Chi Thanh, Dong Da, Hanoi, Vietnam
Get in touch
Tell us what you want to build, modernize, or operate.