Memo · Concept · 2-minute read
Signing up the long tail of small ISPs
About 1,500 small ISPs are too small for a reseller program and too many to sign by hand. The fix: an offer built for them, and agents that handle research, outreach and onboarding, piloted in Virginia.
The opportunity
Distribution through people rural customers already trust
1,458
small fixed-broadband providers in the US, each covering 25,000 homes or fewer
9.3M
homes those small providers cover combined
32
small providers in Virginia alone, the pilot state
Source: FCC National Broadband Map, data as of June 30, 2026. Homes covered, not subscribers.
Why it's been hard
Four likely reasons, all fixable
01
Built for bigger partners
Reported reseller terms expected millions in annual revenue. An 800-customer ISP can't meet that.
02
Sounds like "help your competitor"
Many small ISPs lose customers to Starlink. A reseller pitch gets ignored.
03
The math doesn't work by hand
One sales call per tiny ISP costs more than the few dozen subscribers it brings.
04
Generic outreach
Local owners answer emails that show you know their county, not templates.
What 17 Virginia ISPs show
The need is already visible on their own websites
Piedmont Broadband (Sperryville) already installs Starlink at homes its towers can't see, but gets paid only once.
The offer
Three tiers, so a 300-customer WISP and a 10,000-member co-op can both say yes
Refer
- They do
- Send out-of-range addresses to a tracked link
- They earn
- A bounty per activation
Install
- They do
- Sell, install, first-line support
- They earn
- Their own install fee plus a monthly share
Resell
- They do
- Bill the customer under their own brand
- They earn
- A larger monthly share
No feesNo minimumsThey keep the customerCan move customers to their own network later
Starlink would set the actual numbers. The prototype uses placeholders.
The motion
Agents do the volume. A person approves and takes the calls.
Agent
Find
Every small ISP in a county or state
Agent
Research
Sourced profile in about a minute
Agent
Rank
Fit score with the reasons shown
Agent + person
Reach out
Personal email, approved before sending
Agent + person
Reply
Sort replies, draft honest answers
Self-serve
Onboard
Sign, train, demo kit, first customer
Proposed pilot
Virginia, six weeks
Weeks 1–2
Full Virginia list, profile and score every ISP, set offer terms
Weeks 3–5
Approved outreach to the top 50, calls handled by a person
Week 6
Results and a go / no-go
Reply rateCalls bookedPartners signedFirst activationsCost per ISP researched
Only Starlink can answer
Open questions
- What tripped up the earlier attempt: the offer, the targeting or the outreach?
- What can partners earn today, and who signs the agreement?
- Is there a list of ISPs already contacted, or already reselling?
- How should channel conflict work where Starlink sells direct?