Provisioning fails at the handoffs. Before a circuit, a wave or an Ethernet service carries traffic, the order crosses order management, inventory, design, configuration, test and billing, usually in five or more systems bought in different decades that do not agree on what is in the network. Each handoff is a chance for the record and the network to diverge. A Tier 1 operator runs thousands of these a week, on a legacy stack and a new one at the same time, with fewer people than it had two years ago. The errors are not random. They concentrate in four places, and each one can be closed.

Most fallout starts before anyone types a command. The inventory says a port is free and it is not. The circuit ID on the order does not exist on the device, or exists under a name left over from a migration nobody documented. Design is done against the record, configuration is pushed against the network, and the difference surfaces at test, with the customer already holding a date.

On a large network the gap between record and reality is measured in percent, and every percent is a queue of orders that fall out of the automated flow and come back as manual work. The fix is unglamorous. Reconcile before you provision. Make the device audit the first step of the order and treat a mismatch as a stop.

A senior engineer configuring a PE router at two in the morning from a template is reliable. The same engineer configuring the fifth device of the night from memory, because the template did not cover that vendor’s syntax, is where the typo lives. Manual configuration fails on volume, and volume is what a Tier 1 operator has more of than anyone.

Automation takes the repeatable part. It does not take the exception: the legacy device the script does not know, the order that changed after design, the customer with a non-standard handoff. Those still need an engineer who knows the network, and that engineer is now the scarce resource.

Sales amends the bandwidth. The customer moves the demarcation point. A site survey finds no fiber in the building. A wholesale partner changes the NNI. The design approved on Monday is provisioned on Thursday against a different order, and nobody re-runs the design step because the ticket number is the same. In a mature process the order and the design carry a version, and the pre-check compares them. In most processes the pre-check is a person remembering.

An activation is a change. It deserves what a core migration gets: a method of procedure, a pre-check, a go or no-go, a named rollback owner and a post-check. On many provisioning floors it gets none of that, because the change is small and there are 300 more behind it. Small changes are where the outages come from. When one fails, the operator pays twice, once for the rework and once for the customer who was promised a date.

Four things are true at once for a US Tier 1 operator this year.

The headcount is smaller. T-Mobile cut 4,671 roles in the first half of 2026 (Fierce Network, 12 August 2026). AT&T’s $4 billion cost program to 2028 names “vendor rationalization” as one of its levers (Q1 2026 call, 22 April 2026). The order queue did not shrink with the org chart.

Two networks run in parallel. AT&T’s first FCC-granted copper discontinuance takes effect on 15 November 2026 (auto-granted 13 January 2026), while fiber and wave orders keep growing. The same desk carries a TDM disconnect and a 400G wave on the same day, in different systems, with different fallout rules.

The tools multiply. EMA’s Network Management Megatrends 2026 survey (22 April 2026, 352 respondents) found only 31% of organizations “completely successful” with their network operations strategy and put alert noise and tool sprawl among the top problems. Every extra tool is another place for the record to disagree with the network.

The senior engineers are being pulled toward design. Foote Partners reported on 21 August 2026, from 5,137 employers, that automation is devaluing “routine network operations work, including monitoring, basic troubleshooting, and configuration changes” while raising the market value of design and architecture skills. The people who used to catch the provisioning exception are being promoted away from it.

“Error-free” is a target. Five numbers make it a contract. Ask for each of them, defined, with the period, from whoever runs your provisioning, in-house or not.

Mikkena Global J.S.C. runs provisioning and service activation for Tier 1 operators from Pristina, Kosovo, inside the operator’s own process and change calendar. We publish a 99.99% provisioning SLA and price the work per order or per port with a rework guarantee, so the cost of a failed change lands on us. On one North American MSO’s network our engineers delivered 91,031 network designs with 85% fewer post-deployment fixes and cut restoration time from 8.5 hours to 3.4 (mikkena.com/case-study). On one US Tier 1 carrier’s migration program, rework went from 28% to 0%. Our day shift covers midnight to 11:00 Eastern, which is the change window most US operators use. If you want to see your fallout and rework rates as a unit price, send us one month of order volume for one product line. Pricing it takes us a day.

Sources: Fierce Network (12 August 2026); Broadband Breakfast on AT&T copper discontinuance (13 January 2026); EMA, Network Management Megatrends 2026 (22 April 2026); Foote Partners via Network World (21 August 2026).

Four kinds of vendor will run a legacy voice platform for you in 2026, and they do four different jobs. Platform vendors, Ribbon and Alianza (which owns Metaswitch) among them, support and replace the switches. Signaling service bureaus such as TNS, Syniverse and BICS carry SS7 for you as a hosted, managed service. Large managed network services firms such as Tech Mahindra, Prodapt, HCLTech, Cyient and Infinite run NOC operations across the whole estate, legacy voice included. Specialist engineering squads such as Mikkena run the TDM and SS7 NOC you already own, per device or per ticket, until the switch-off date. Which one fits depends on whether you are keeping the platform, replacing it, or running it down to a date.

The FCC’s order of 26 March 2026 cut the automatic grant period for discontinuance applications to 31 days and preempted state rules that delay it. AT&T’s first FCC-granted copper discontinuance takes effect on 15 November 2026 (auto-granted 13 January 2026). AT&T expects “a couple of hundred wire centers” to have zero customers by the end of this year (Q2 2026 call, 22 July 2026), and its legacy service revenue fell 26% year over year in the same quarter. In the UK, BT’s PSTN switch-off is fixed for 31 January 2027 (BT, 22 May 2026).

The platforms are aging on the same clock. Ribbon’s own TDM switch retirement calculator assumes a 25-year design life (ribboncommunications.com, accessed 21 September 2026), which puts most of the installed base past it. Microsoft laid off most of the Metaswitch team in June 2024 (Light Reading, 13 June 2024; Alianza, 18 June 2024) and sold the business to Alianza, which completed the purchase on 4 March 2025 and has committed to keep supporting the existing products.

The staffing picture is the hard part. The last two to four years of a TDM or SS7 platform are the hardest years to staff, because nobody builds a career on the last DMS in the region and the engineers who know it are retiring or being moved to fiber. A NOC contract for that period is not a modernization program. It is insurance against the one engineer who knows the switch leaving before the date.

Ribbon Communications holds the Nortel DMS lineage through Genband and sells the C15 and C20 call controllers and the G6 media gateway as the replacement path, with STP and Diameter signaling software that runs as an appliance, on COTS hardware or in a cloud. On 9 September 2026 Ribbon and TNS announced a managed, cloud-based voice service built on Ribbon’s application server, SBC and PSX, positioned as a path off legacy switches without a full replacement. Alianza completed its acquisition of Metaswitch from Microsoft on 4 March 2025, serves more than 1,000 service providers, and says it will continue to support the existing Metaswitch products while it integrates them. Nokia holds the 5ESS lineage through Lucent and Alcatel-Lucent and still hosts the documentation; get the support end date in writing.

None of them runs your NOC. Their services teams migrate you; until then, your people watch the platform.

TNS sells SS7 ISUP and TCAP signaling as a managed service on its own network, a managed conversion from SS7 to SIGTRAN that “reduces costly time-division multiplexing (TDM) circuits”, and routing, number portability data and access to every toll-free number in the United States and Canada (communications.tnsi.com, accessed 21 September 2026). Syniverse and BICS do the equivalent for mobile signaling and IPX; the two announced a 5G standalone roaming partnership on 3 March 2026, with Syniverse citing more than 830 connections worldwide.

Put your signaling on a bureau and you have removed the STPs from your estate. The switches, the trunks and the alarms behind them are still yours, so the NOC question remains.

Prodapt (6,000+ staff; AT&T, Windstream, Ziply, Brightspeed and Zayo are among the customers named on its site), Tech Mahindra (communications is about 32% of a $1.66 billion quarter, results of 17 July 2026), HCLTech (a six-year, $2.1 billion managed network services contract with Verizon Business, August 2023), Cyient (15,000+ staff) and Infinite (a telecom NOC covering mobile core, IMS and fixed voice, monitoring 45,000 devices and handling 220,000 tickets a year by its own count) will take the whole NOC, legacy voice included. These are multi-year managed services or time-and-materials contracts, and the legacy platform is one line in a wide scope.

Two things sort this category quickly: the names of the people on the account who have worked a DMS or a 5ESS this year, and whether the monthly fee falls when the wire centers reach zero.

Mikkena Global J.S.C. lists legacy TDM and SS7 NOC operations as a service line and works the other way round from the three categories above. We run the platforms you already own; selling a replacement is somebody else’s job. L1 to L3 coverage around the clock from Pristina, an MTTR commitment in the contract, priced per device or per ticket, with a price that steps down as wire centers close. Our day shift covers 02:00 to 11:00 Eastern, where most US change windows sit. ISO 27001 and ISO 9001, with the access model for customer data agreed in writing before an engineer touches a platform. On one North American MSO’s network our NOC cut restoration time from 8.5 hours to 3.4 (mikkena.com/case-study).

Decide what you are doing with the platform: keeping it past 2030, replacing it, or running it down to a date. Then ask every vendor on the list for five things in writing.

Our answer to the question in the title is narrow on purpose. If you are replacing the platform, talk to Ribbon or Alianza and to a signaling bureau. If you are running it down to 15 November 2026, 31 January 2027 or a date of your own, we will price the NOC for one wire center cluster or one exchange group, per device, with the MTTR in writing. A one-page scope takes us a day.

Sources: FCC order 26-19 (26 March 2026); Broadband Breakfast on AT&T copper discontinuance (13 January 2026); Alianza on the Metaswitch acquisition (4 March 2025). Vendor figures are from each company’s own published material, as dated in the text.

Some things were simply never meant to work alone.

A handshake needs two hands. A conversation needs someone to speak and someone to listen. A bridge needs two sides to connect. A question needs an answer. A promise needs someone to make it and someone willing to trust it.

Even the word connection implies that there is something on the other side.

The more you think about it, the more you realize that much of what moves the world forward happens when two sides come together.

Telecom is no different.

In fact, for an industry built entirely around connecting people, places and technology, we sometimes spend surprisingly little time talking about one of the connections that matters most: the one between the people building it.

That is the idea behind It Takes Two to Telecom.

Look behind any successful network deployment and you will find far more than infrastructure.

There are engineers solving problems in the field. Project managers keeping hundreds of moving pieces aligned. Technicians working against deadlines. Operations teams responding when something changes. Customers expecting the network to work without ever having to think about what it took to make that possible.

And somewhere between all of those moving parts are partnerships.

The industry often calls them vendor relationships. We think that description falls short.

A vendor completes a scope of work.

A partner understands what the work means to the business around it.

There is a difference.

When timelines change, a partner adjusts. When something unexpected happens in the field, a partner helps solve it rather than pointing at the original scope. When pressure increases, a partner does not disappear behind another process or another layer of communication.

They show up.

Because successful telecom work is rarely about one company handing something off to another. It is about two teams moving toward the same outcome.

Telecom is filled with extraordinary technology. Fiber, wireless infrastructure, data centers, network equipment and the systems connecting them continue to become more sophisticated.

But technology does not eliminate the need for trust.

If anything, the complexity of modern networks makes trust more important.

You need to know that the people on the other side of the project understand the standard. That they will communicate when something changes. That they can operate at the pace the project requires. That when a problem appears, their first instinct is to help solve it.

That kind of confidence cannot be written into a specification.

It is built over time.

Project by project. Market by market. Conversation by conversation.

And when it exists, something changes. Two separate companies stop operating like two separate companies. The lines between “your team” and “our team” begin to disappear.

The work simply becomes the work we need to get done.

That is partnership.

The strongest partnerships are not built because both companies are identical.

They work because each side brings something the other needs.

One side may understand the network from the inside. The other brings the people, expertise and field experience needed to execute it.

One brings the challenge. The other brings another perspective.

One knows where the network needs to go. The other helps build the path to get there.

Different responsibilities. Same destination.

It is the same principle behind so many things that work in everyday life. Two hands complete a handshake. Two ends make a connection. Two sides allow a bridge to do what it was built to do.

Neither side becomes less important because the other exists.

They become more capable because they work together.

Anyone who has spent enough time in telecom knows that perfect plans rarely stay perfect.

Conditions change. Timelines move. Requirements evolve. A deployment that looked straightforward on paper becomes something entirely different once work begins in the field.

That is where the difference between a transactional vendor and a true partner becomes obvious.

A transactional relationship works best when everything goes according to plan.

A partnership proves itself when it doesn’t.

The question stops being, “Whose problem is this?”

It becomes, “How do we solve it?”

That shift may sound small, but across a large network, it changes everything.

It means fewer walls between teams. Faster decisions. Better communication. More accountability. And ultimately, a stronger operation.

You stop spending energy managing the relationship and start spending it moving the work forward.

There is another reason we believe so strongly in this philosophy.

Networks are too important to depend on relationships that only work when everything is easy.

Scale requires resilience, not only in infrastructure but in the ecosystem supporting it.

Having the right partners creates options. It gives teams additional expertise, capacity and perspective when they need it. It reduces dependency on a single way of doing things and creates an organization that can respond when circumstances change.

The objective is not to add complexity.

It is the opposite.

The right partnership should make a complicated business easier to operate.

It should give you someone else who understands the problem before you have to explain every detail. Someone who knows the standards. Someone who can step in when capacity is stretched. Someone who treats the outcome as seriously as you do.

Because the best partnerships do not create more work to manage.

They create more capability.

For us, It Takes Two to Telecom is more than a campaign line.

It describes the kind of company we have always wanted Mikkena to be.

We believe our job is not simply to execute work and move on to the next project. Our job is to understand what our partners are trying to accomplish and become part of the team responsible for making it happen.

That means listening before acting.

It means communicating clearly.

It means understanding that what looks like one project on our side may be one piece of a much larger commitment on our partner’s side.

It means taking ownership.

And above all, it means showing up.

Today, Mikkena’s team of more than 150 engineers operates across all 50 states, supporting some of the largest organizations in telecom. The scale has changed over the years. The principle behind how we work has not.

Networks are built through connections.

So are companies.

So are partnerships.

So is trust.

And after everything we have learned building alongside our partners, we keep coming back to the same simple truth:

It takes two hands to shake.

Two sides to build trust.

Two teams to create a true partnership.

And sometimes, it really does Take Two to Telecom.

For four years, BEAD was a paperwork program. Allocation formulas, challenge processes, subgrantee selection, final proposals. The organizations that did well were the ones with strong grant writers and patient lawyers.

That phase is over. All 56 states and territories submitted Final Proposals, 53 have NTIA approval, and the first BEAD-funded projects are entering construction this summer. Most build activity runs 2026 through 2028, with builds in the largest states, Texas among them, stretching to 2030.

The skill that wins a subgrant and the skill that delivers one are not the same skill. A lot of the awards went to regional ISPs, electric cooperatives, and municipal utilities who have built networks before, but not at this scale, not on this timeline, and never under this compliance regime.

Here is what changes now, and what to check before your first shovel.

BEAD is not a normal build. Funded projects carry obligations that most regional operators have never had to document:

Build America, Buy America sourcing. Domestic content requirements on materials, with documentation to prove it, in a market where fiber and optical components are already supply-constrained by AI data center demand. Sourcing decisions that used to be a procurement conversation are now a compliance conversation, and a waiver request has a timeline attached.

The letter of credit. Before signing a subgrant agreement, awardees post an irrevocable standby letter of credit or a surety bond against the subaward. It can step down as take rates hit thresholds, but for a small operator it ties up real capital at exactly the moment construction costs peak, and the paperwork to reduce it is yours to drive.

Labor compliance, self-certified. The 2025 restructuring removed the workforce development scoring from the application phase, but subgrantees now self-certify compliance with federal labor and employment law. Self-certification is not less exposure, it is exposure with your signature on it and no state reviewer catching problems before they become findings.

Detailed and ongoing reporting. Location-level progress against the specific broadband serviceable locations in your award. This is not a quarterly narrative. It is address-level data, and it has to reconcile.

The 25 percent non-federal match. Most awards require it, and it has to be documented and drawn correctly.

None of these are technically hard. All of them consume engineering and program hours that were not in the original staffing model, because in a normal build they do not exist.

The bigger risk is more mundane.

You won an award to serve several thousand locations across a service territory that was chosen precisely because nobody wanted to build there. That means low density, difficult terrain, long drops, and permitting across multiple jurisdictions with no shared process. Every one of those factors multiplies the engineering hours per passing relative to a suburban overbuild.

Meanwhile, the same team doing that design work is still running your existing network, which did not get smaller when the award came through.

And you are hiring into the tightest network engineering market on record. Enterprise Management Associates put the share of organizations struggling to hire and retain network engineers at 52 percent this year, up from 26 percent in 2022. Every BEAD subgrantee in your state is recruiting from the same pool at the same time, along with every hyperscaler building in the region.

The slip is not dramatic. Design falls two months behind, permitting starts late because permits follow design, construction crews are scheduled against dates that design cannot support, and you pay crew standby while your reporting shows locations passed against a milestone you are going to miss.

Run this before construction, not after the first slip.

1. Hours, not headcount, split by where the work happens. Model the total engineering hours the award requires, then divide them into work that requires someone on site (OSP design walkouts, splice supervision, field validation, construction oversight) and work that does not (transport and IP design, configuration and turn-up, service validation, routing policy, monitoring design, documentation, reporting). A single blended estimate hides the shortage and, worse, hides the fact that these two categories have completely different fill options.

2. Your actual available capacity. Take current team hours and subtract everything committed to running the existing network. In most organizations that leaves 5 to 20 percent, not the number the org chart implies.

3. Permitting lead times by jurisdiction, mapped to design completion dates. Permits cannot start before design finishes. If you have twelve municipalities and one designer, sequence it now.

4. Make-ready and pole attachment exposure. This is where schedules die. Pole attachment regulation is itself contested right now, with NARUC’s Telecommunications Committee passing a resolution in July arguing the FCC lacks authority to preempt state pole attachment rules over policy disagreements. Assume the process is slower than your plan.

5. Material sourcing against BABA, with lead times. Fiber and high-density optical supply is tight because AI data center demand overtook telecom as the primary growth driver for optical cable this year. Order windows that were comfortable in 2023 are not comfortable now.

6. Documentation ownership, named. As-builts are the first deliverable to slip and the most expensive to reconstruct. Someone’s name goes on this or it will not happen.

7. Reporting reconciliation. Confirm your location-level progress data can actually reconcile to the BSL list in your award. If your field systems and your award data do not share a location identifier, fix that before you have 4,000 records.

8. Surge plan for the off-site work, defined in advance. Field crews you will source locally, and that market is what it is. The off-site engineering load is the part you have options on, and it is the part that quietly consumes your senior local people if you do not plan for it. Decide now, not in month four at spot rates with no time to onboard anyone.

Item eight is the one that separates programs that hold their dates from programs that do not, for a reason that is easy to miss.

The instinct during a build is to point every available engineer at the field, because the field is visibly behind. What actually happens is the opposite. The logical work does not go away, so your senior local engineers end up splitting their week between a construction site and a config backlog, doing neither well, and the design queue becomes the thing that holds up permitting, which holds up crews.

Spot contract engineering does not fix it either. A contractor who arrives in month four does not know your network, your standards, or your territory. They spend six weeks learning on your clock and then leave with the context. Repeat that three times over a two-year build and you have paid a premium for discontinuity.

The alternative is to establish the off-site capacity before you need it, with a team that learns your standards during the planning phase and is already productive when the backlog appears.

That is what Mikkena provides. Network engineering from our team in Kosovo, covering the logical layer: design and OIC engineering, provisioning and service activation, access and aggregation, Metro Ethernet and MPLS transport, NOC and ENOC operations, and the migration engineering that moves services onto new infrastructure without outages. We do not send crews. What we do is make sure the people who can be on site are actually on site, instead of clearing a config queue at 9 p.m.

Source: NTIA, BEAD program.