Updated: Oct 10, 2025
The lease is signed on Friday. Fit-out starts Monday. The store in Pune must open the following Saturday. Your team does not have a week for staging or a spare engineer to camp on site. You need a design that travels in a box, boots in minutes, and behaves the same way in Noida, Chennai, and Chandigarh. This is what a one-week rollout looks like when you treat the network as software and policy as code.
Here is the promise. In seven days, you open the site on time, keep POS and ERP steady, and publish a runbook the business can trust. You make four hard choices on day zero, you pre-flight once, and you test with thresholds the day before opening. Everything else is detail.
For stores with heavy SaaS usage, use a local internet breakout; for kiosks and pop-ups, use a centralised breakout. You pick one and write it down. The reason is latency and audit. Local gives faster app performance. Central keeps control simple.
Where two last-mile types exist, take both. Else choose the best wireline you can get and add LTE on day one. The second circuit removes risk. LTE buys you time when fibre slips.
Use 802.1X with certificates for staff devices so policy follows users. For guest and shared devices, use PPSK or a captive portal with rate limits. This keeps credentials off sticky notes and gives you clean device histories.
Use hub and spoke for Auto VPN. Add a second hub for resilience or analytics later. You can get fancy next month. This week, you want predictable paths and clean logs.
An Engineer in Bengaluru builds a single template in the Meraki dashboard. It defines VLANs for POS, staff, and guests, WiFi SSIDs, SD-WAN health thresholds, content filtering, and a small set of security rules. Variables for each city live in a simple file: site code, WAN addressing, SSID name, and hub pair.
Day 1. The template and variables are finalised. SSO is tested, admin roles are assigned, and the change window is agreed. Floor maps are gathered, the pre-flight list is confirmed with the landlord, and ISP hand-off details are captured.
Day 2. Serials are claimed and bound to the template, site tags are set, and floor maps are uploaded. Auto VPN hubs, Mumbai and Delhi, are selected. LTE units are staged as fallback, pre-authorised admin accounts are in place, and shipping to the site is confirmed with access details.
Day 3. Devices arrive at the Pune address on Day 3. The coordinator unboxes the MX security appliance, MS switch, and MR access points. Power is on, a single cable goes from the ISP handoff to the MX, the rest are switch ports and PoE. The MX checks in, claims its template, builds Auto VPN to hubs in Mumbai and Delhi, and starts enforcing traffic rules. The APs appear, pull the wireless profile, and broadcast the right SSIDs. A field engineer checks the signal on a phone, sees a clean band plan, and lowers power on one AP above the tills.
Day 4. Identity goes live on Day 4. Staff laptops authenticate with 802.1X and certificates from ISE. Printers and scanners use MAB and live in their own role. Guests get PPSK with a speed cap. Role tags are present at the edge, not only at the core. You can now deny east-west traffic by role without ACL explosions.
Day 5. On this day, the team steers critical traffic with SDWAN. Voice prefers the cleaner path, SaaS does the same, and both fail over when loss or jitter crosses a line for more than a few seconds. Cellular sits as a reserve. A small policy sends software updates at night so tills do not stutter at 6 pm.
Day 6. This day is for tests with thresholds. Pull the primary link, failover should happen in under ten seconds, and packet loss should stay under one per cent. Place a five-minute voice call, MOS should stay above 4.0 and jitter under 30 ms. Run 802.1X and captive portal tests, aim for 99 per cent success and a portal load under five seconds. Walk the floor, check RSSI at tills is better than minus 65 dBm and that roaming stays under 150 ms. Run three POS and ERP workflows and confirm response time stays inside your budget.
Day 7. You did it. The store opens today. Take a bow! The dashboard shows normal. You publish a short runbook, a contact list, and the SLOs you will hold yourself to. No one notices the network, which is the point.
A retailer stood up 20 kiosks across NCR in ten days. They chose local breakout for stores and central for kiosks. Five malls received dual ISP, and three sites ran LTE first. One fibre missed its date. The kiosk opened on LTE and cut to fibre a week later. Onboarding time per site fell from three days to one, change failure rate dropped to three per cent, and tickets per site fell by 28 per cent after week two. This is the outcome you want to see on your own numbers.
Templates and variables live in Git. Engineers propose changes as pull requests. Site operations track ISP tickets. Project managers own cutover windows and communications. Service delivery signs the handoff and starts the SLO clock. A small JSON or YAML file holds per-site variables so the dashboard template can stamp out a new branch in minutes.
Publish targets tied to experience, not to ports. MTTR under 30 minutes for access incidents, change failure rate under five per cent, 95 per cent 802.1X coverage by day 14 for a new site, and detect to quarantine under five minutes for a non-compliant device. Alert on WAN brownouts, authfailure spikes, Auto VPN down, AP overutilisation, DHCP pool at 80 per cent, and DNS block surges. Keep thresholds simple so the NOC acts fast.
Independent research points to material downtime cost and to gains from centralised operations when teams adopt cloud-managed designs. Use these as external signals in your board note, then trend your own numbers after go-live. (Gartner on network operations and downtime cost), (Cisco Meraki documentation on zero-touch provisioning).
Carriers miss dates, so ship with cellular and run day one on LTE. Contractors loop switches, so enable port security and BPDU guard on every access port. Power flickers, so set a UPS for 30 minutes and a clean shutdown SOP. Address plans arrive late, so use DHCP and reserved ranges on day one and move to the final plan in the weektwo change window. These controls keep schedules safe.
Day 1, template and variables are final, and everyone knows the four choices.
Day 2, serials are claimed, templates are bound, and tags are set.
Day 3, devices arrive, cellular is tested, and the site appears in the mesh.
Day 4, identity goes live, staff on 802.1X, guests isolated.
Day 5, SDWAN health policies protect voice and SaaS.
Day 6, failover and quality tests meet thresholds, gaps are fixed.
Day 7, the store opens on schedule, and the runbook and SLOs are published.
As a Cisco Gold Partner, Proactive designs and runs one-week site rollouts across metros and tiertwo towns. We bring templates you can audit, a pre-flight that landlords understand, and a test plan your NOC can run. We leave you with working alerts, clean dashboards, and a runbook your operations team will follow.
Book a one-hour rollout session. You leave with a template pack, a city variable sheet, a pre-flight checklist, and a day six test script tailored to your environment. Write to [email protected] to schedule.
