Network infrastructure consultancy · United Kingdom
Burrow surveys, designs and builds the networks underneath multi-site UK businesses — the switching, routing, wireless and connectivity everything else quietly assumes is fine.
00Position
We are not your IT department, and we are not trying to become it.
Most network problems are not solved by another helpdesk. They are solved by someone spending a week in the configs, writing down what is actually there, and designing the thing that should replace it. That is the job we do.
We work alongside in-house teams and alongside your existing MSP. You keep the relationship you have; we fix the layer underneath it and hand over documentation your own people can run from. If the honest answer is that you do not need us yet, we will tell you that in the survey report you have already paid for.
01Work
Scoped as projects with a start and an end. Priced before we begin.
Access and distribution switching, stack and uplink design, VLAN and addressing schemes, PoE budgeting for phones, APs and cameras, and structured cabling to spec. We survey what is actually in the comms room, design what should replace it, and cut it over out of hours. You get as-built documentation, not a verbal handover.
Multi-site WAN design, circuit procurement, and diverse failover that actually fails over when you test it. BGP and OSPF where they earn their place, static routing where they do not. SD-WAN when the site count justifies the licensing — and a straight answer when it doesn't.
Increasingly that means 5G fixed wireless where the fibre lead time is nine months, and low-orbit satellite where there is no fibre to wait for. Both are real options and both come with real caveats — CGNAT, no meaningful SLA, latency that moves. We scope them with those stated, not buried.
We will also provide the circuits and monitor them. Continuous polling across latency, jitter, loss and throughput — so you hear about a link that is degrading, not just one that has dropped — scheduled failover testing that proves the backup path works, and monthly availability reporting against what your carrier actually committed to. Rolling 30-day term. Monitoring a circuit should not need a three-year signature.
Predictive design, on-site validation and post-install survey. Warehouse racking, plant floors, listed buildings and full assembly halls — the environments where a desk-based design falls over on day one. You get heat maps, an AP schedule and a channel plan, not an assurance that it should be fine.
We survey DECT as well, which is unusual. It sits at 1880–1900 MHz, where no Wi-Fi survey tool can see it, so most estates get designed by guesswork — which is why the handsets drop in the same corner of the warehouse every week and nobody can say why. You get a base station plan and a coverage map alongside the AP schedule.
VLAN and VRF segmentation, 802.1X and network access control, firewall rule-base rebuilds, and getting production traffic off the same flat network as the finance team. The unglamorous work that decides whether an incident costs you an afternoon or a fortnight.
Refresh planning, switch and firewall vendor selection, and independent review of someone else's network proposal before you sign it. Charged by the day. We take no vendor commission, which is why we can tell you a quote is bad without it costing us anything.
02Method
Every engagement runs the same three stages, in order. You can stop after any of them and take the output elsewhere.
Two to five days on site and in the configuration, scaled to how many sites you have, then one to two days of analysis and write-up. Physical audit, traffic baselining, config extraction, a wireless walk where it matters. The report is the deliverable, and it is yours whether or not you use us for the work that follows.
2–5 days on site · report in 10 working daysA written low-level design with a bill of materials, a staged migration plan, rollback points and the risks stated plainly rather than buried. It is your document. Take it to any installer you like.
2–4 weeks · fixed feeOut-of-hours cutover against the agreed plan, as-built documentation, config backups in your repository, and two weeks of hypercare with a named engineer. Then we get out of the way.
Scoped per site · fixed price03Limits
No project is conditional on signing anything afterwards. Circuit monitoring is available on a rolling 30-day term, cancellable in a month, and it monitors your connections — it is not a helpdesk for your users and we will not pretend otherwise.
We pass through vendor pricing at cost and charge a stated procurement fee on top. You see both numbers on the same page of the quote.
We are a small delivery team, not a call centre. The engineer who surveyed your network is the one who designs it and the one on the bridge at 2am during cutover.
04Fees
| Engagement | What you get | From |
|---|---|---|
| Network review | LAN and WAN survey — 2–5 days on site, 1–2 days write-up, report within 10 working days. Three sites in the baseline. Wi-Fi and DECT surveying quoted separately, below. | £2,400 |
| Wireless & DECT survey | Predictive design or on-site validation of one building, covering both Wi-Fi and the 1880–1900 MHz DECT band. Heat maps, channel plan, AP schedule and base station plan. | £1,400 / site |
| Low-level design | Full LLD, switch and circuit bill of materials, migration plan and risk register. Yours to take anywhere. | £4,800 |
| Project delivery | Build, configuration and out-of-hours cutover, scoped and fixed before we start. | £950 / day |
| Connectivity & monitoring | Circuit provision with continuous link monitoring, degradation alerting, scheduled failover testing and monthly availability reporting against carrier SLA. Response next business day; out-of-hours engineering by arrangement. Rolling 30-day term. | £45 / site / mo |
| Advisory retainer | Two days a month of design authority, tender review and escalation cover. | £1,600 / month |
Excludes VAT and hardware. Travel charged at cost outside Yorkshire and the North West.
05Sectors
Estates of 20–200 branches where every site is a single point of failure and nobody can say which firmware is where. Standardised builds, zero-touch provisioning, 5G backup that keeps the tills up, and a refresh plan that fits a trading calendar.
Plant networks that grew around the machines rather than the other way round. Segmentation between production and corporate, resilient warehouse wireless, DECT coverage that reaches the far aisle, and cutovers planned around a shift pattern.
Networks that grew by accretion — a switch added per hire, cabling nobody has documented since 2014. Re-racked comms rooms, right-sized circuits, and resilience that reflects how full the building actually is.
Schools and colleges running high device density on capital funding cycles. Surveys that survive a full hall, and designs costed to a governors' approval date.
06Questions
Usually the opposite. Most managed service agreements cover support and operations, not infrastructure design and delivery. We scope the work with your incumbent in the room, share the design with them, and hand the finished network back into their support. We are not pitching to replace them.
Yes, and a reasonable number of clients do exactly that. The findings report is written to be useful on its own — it names what is wrong, what it will cost to fix, and what can safely wait another year. There is no obligation to continue and no discount for doing so.
For the right site, yes — and we design with both. Low-orbit satellite is often the only sensible answer for a remote depot, a construction compound or a site where the fibre quote came back at £40,000 and nine months. 5G fixed wireless is frequently a better diverse backup than a second fibre that shares the same duct as the first.
The caveats are real and we will put them in writing. Both are CGNAT by default, so anything needing inbound access requires a public IP option. Neither carries an SLA worth the name. Latency on satellite moves in a way that voice and some line-of-business applications notice. Used as a remote-site primary or a genuinely diverse backup, they are excellent. Used as the link into your core, they are not.
Detection and notification are automated, so you know as soon as a link degrades or drops — overnight included. That is usually the valuable part on its own: most alerts turn out to be a failover doing exactly what it was designed to do, and knowing that at 3am is considerably better than finding out at 8am.
Our own engineering response is next business day as standard, with a plain account of what happened and what we are doing about it. Out-of-hours response is available by arrangement and priced separately. We would rather agree that with you up front than imply cover we have not committed to.
Cisco and Meraki, HPE Aruba, Juniper Mist, Extreme, Fortinet, Ubiquiti and MikroTik in practice. We hold no reseller targets, so the recommendation follows the requirement and the budget rather than a quarterly quota. Where you already have an estate we will normally design to extend it, not replace it.
Surveys typically start within three weeks. Delivery work is booked against cutover windows, so a multi-site build usually lands six to ten weeks out. Emergency remediation is a different conversation — put URGENT in the subject line and we will come back the same day.
Start here
A site list, a topology diagram, or one sentence about the outage that keeps happening. We will tell you within two working days whether it is a survey, a design, or something you can fix yourself on a Tuesday.