Most new door-to-door companies start the same way. A top rep or manager leaves a big org, brings a few loyal reps, buys a canvassing app and a CRM, and starts knocking. Sales come fast. Then the problems show up somewhere other than the door: jobs that stall after the signature, commission disputes, cancellations nobody saw coming, and an owner who knows how many doors were knocked but not how much cash the company actually has.
One of us built and ran a D2D solar sales company for four years. These are the systems we'd put in place in the first 90 days, in this order, and the ones we'd make wait.
Before the first door: one record per household
Pick the system of record before the first knock. Your canvassing app is built for reps at the door: pins, knock results, appointments. It isn't built to run the account after the signature. Decide on day one that the CRM holds the customer, the deal and the job, and that the canvassing app feeds it.
Then connect them, so that a set appointment creates or updates one CRM record per household with the rep attached. Every report, commission and cancellation you track later depends on that one record being right. We wrote more about this in Your canvassing app is not your CRM.
Weeks 1 to 2: the comp plan, written down
Write the comp plan before anyone sells under it: what a rep earns, when it's earned (signed, installed, first service, or past the cancellation window), what triggers a chargeback, and how overrides work. Then put every deal's commission status in the CRM next to the deal.
New companies skip this because the founding reps trust the founder. That trust lasts until the first chargeback. A written plan and a commission status on every deal turn an argument into a lookup.
If you sell solar in Utah, your reps and setters now have to be W-2 employees, which moves all of this into payroll. See Utah's solar retailer law.
Weeks 2 to 4: contracts and the handoff
Put contracts and e-sign inside the flow, not beside it. A signature should move the deal to the next stage on its own, with the documents attached.
Then build the handoff from sale to delivery as a stage with an owner, not a text to the install manager or service tech. In pest control, the gap between the sale and the first service is where accounts die. PestPac reports pre-service cancellations as high as 30%, falling to about 5% when the first service is booked on the spot. That's a vendor figure without a stated method, but it matches what operators see: the faster the first service, the fewer cancellations.
For solar, the same gap runs from signature to site survey, design, permit and install. Each stage needs an owner and a time limit, or accounts go quiet.
Month 2: the weekly scorecard
You can't manage what lives in five tools. By month two, the owner should get one weekly view:
| Number | Why it matters |
|---|---|
| Doors knocked and appointments set | Activity, by rep |
| Sits and signed contracts | Close rate, by rep |
| Installed or first service completed | What actually becomes revenue |
| Cancellations before install or first service | Where the money leaks |
| Chargebacks and commissions owed | What you owe, and when |
| Cash in and cash committed | Whether you can make the fall payout |
If those numbers come from the CRM, the scorecard takes minutes. If they come from asking the office, it takes a day and it's wrong.
Month 3: cash and the back end
Summer programs pay back end commission in the fall, on sales from months earlier. PestPac suggests reserving 20 to 25% of summer revenue for fall payouts. Whatever your number is, put it in the scorecard as its own line, so cash in the bank never looks like cash you can spend.
The lesson from running our own company: a blitz needs a stable recurring base under it, and a capital-heavy model needs a cash buffer before it needs another office. Systems are how you see both coming.
What can wait
- Custom software. Use the tools your reps already know and connect them first.
- Dashboards beyond the scorecard. One weekly view that's trusted beats ten that aren't.
- AI anything, until the data underneath is clean. Automating a broken handoff just breaks it faster.
- A second office, until the first one's numbers come from the system without anyone touching them.
The order matters more than the tools
Every tool on the market can do most of this. What fails is the order: companies build marketing before the handoff, dashboards before the data, and offices before the comp plan. Build it in the order the money moves, door to signature to delivery to cash, and the company you started this summer is still standing next summer.