How to build a sales pipeline from scratch

If your deals live in your head, a notebook and three text threads, you don’t have a pipeline. You have a memory test. A real pipeline shows every open opportunity, where it stands, what it’s worth and what happens next, so you can spot problems before they cost you a month of revenue.

This guide walks you through building one from scratch in an afternoon: mapping how your customers buy, defining stages with exit criteria, setting probabilities, loading your deals and running a weekly review. At the end you’ll find example pipelines for contractors, realtors and lawn care businesses.

What a pipeline is for

A sales pipeline is a visual model of your sales process. Each column is a stage, each card is a deal, and the cards move left to right until they’re won or lost. It answers three questions at a glance:

• What’s open? Every deal you’re working, with its value.
• What’s stuck? Deals that haven’t moved in too long.
• What’s likely to close? A weighted view that feeds your sales forecast.

There’s evidence that the structure itself matters. In a 2015 survey of 62 B2B companies by Vantage Point Performance and the Sales Management Association, reported in Harvard Business Review, there was an 18% difference in revenue growth between companies that had defined a formal sales process and those that hadn’t. Growth was self-rated by executives on a 1–7 scale, so treat it as a strong signal rather than a precise number. Small businesses aren’t 62 B2B companies, but the lesson travels: define your stages clearly and everyone knows where a deal stands.

Step 1: Map how your customers buy

Don’t start by copying someone else’s stages. Start with your last ten won deals and write down what the customer did at each point:

1. They found you (form, call, referral, ad).
2. They talked to you and shared what they needed.
3. They let you see the job, the property or the problem.
4. They received a price.
5. They asked questions, negotiated or compared.
6. They said yes, signed or paid a deposit.

Now look at your lost deals. Where did they stall? Those sticking points are usually the stages that matter most, because they’re where follow-up makes the difference.

Step 2: Define stages with exit criteria

A stage name is only useful if everyone agrees when a deal belongs there. Give each stage an exit criterion: a concrete, observable buyer action that moves the deal forward. “Customer seems interested” isn’t an exit criterion. “Customer booked a site visit” is.

Here’s a generic service-business pipeline to adapt:

Stage	What it means	Exit criterion (move on when…)
New inquiry	Someone raised their hand	You’ve spoken and confirmed a real need
Qualified	Need, budget range and timing fit	A visit, call or consultation is booked
Appointment set	Meeting or site visit is scheduled	The meeting happened and you have enough to price
Quote sent	Customer has your price	Customer gives a verbal yes or asks for changes
Negotiation	Adjusting scope or terms	Customer signs or accepts
Won / Lost	Deal is closed	—

Keep qualification rules in one place so reps apply them the same way. If you haven’t settled on yours, read how to qualify leads before you finalize the second stage.

Step 3: Set stage probabilities

A stage probability is the chance that a deal in that stage eventually closes. Multiply each deal’s value by its stage probability and add them up, and you get a weighted pipeline, the most honest single number for what’s likely to come in.

With no history, start with guesses that climb as deals advance. Say you have:

• Four deals at “Quote sent” worth $8,000 each, at 40%: $12,800 weighted.
• Two deals at “Negotiation” worth $15,000 each, at 70%: $21,000 weighted.

Your weighted pipeline is $33,800, even though the raw total is $62,000. That gap is the point: it stops you from planning crews or hiring around money you probably won’t see.

After a quarter, replace the guesses with real numbers. For each stage, look at how many deals that reached it eventually closed. If only a quarter of your quoted deals close, 40% was optimistic, and your forecast should say so.

Step 4: Decide what every deal must include

A deal card without key fields is just a name. Require these before a deal enters the pipeline:

• Value. Your best estimate, updated when you send the quote.
• Owner. One person responsible for the next step.
• Expected close date. When the customer said they want to decide, not when you hope they will.
• Next step. A task with a due date, such as “call Tuesday to walk through the quote.”
• Source. Where the lead came from, so you can see which channels produce revenue.

When a deal is lost, record a lost reason from a short list (price, timing, went with a competitor, no response, not a fit). After a few months, the pattern tells you whether you have a pricing problem, a follow-up problem or a targeting problem.

Clean data isn’t a chore that only big companies care about. In Salesforce’s 2026 State of Sales research (a vendor survey of 4,050 sales professionals in 22 countries, conducted August–September 2025), 79% of high performers said they prioritize data hygiene, compared with 54% of underperformers.

Step 5: Load your current deals

Pull every open opportunity out of your inbox, your phone and your spreadsheet. For each one, ask: “Is there a real next step with this customer?” If not, it’s either a lead to nurture or a deal to mark lost. Be ruthless. A pipeline padded with dead deals feels good for a week and then stops being trusted.

If you’re importing from a spreadsheet, a CSV import with column mapping saves time. In Exa, you can map columns (including custom fields), choose whether duplicates are skipped, updated or created, and then drag each deal into the right stage on a Kanban board.

Step 6: Review it every week

A pipeline that isn’t reviewed decays. The same HBR-reported survey found that companies spending at least three hours a month managing each rep’s pipeline saw 11% greater revenue growth than those spending less, and the authors stressed that the time should go to moving deals forward, not just to forecasting.

A 30-minute weekly review for a small team:

1. Stalled deals first. Anything that hasn’t moved in longer than usual for its stage.
2. Missing next steps. Every open deal should have a dated task.
3. Close dates in the past. Update them or mark the deal lost.
4. This month’s likely wins. What would it take to get each one signed?
5. One coaching question per rep. “What’s blocking your biggest deal?”

Exa can flag “rotting” deals that sit in a stage too long, which makes the first item quick. For a deeper look at running reviews and metrics, see our guide to sales pipeline management.

Example pipelines for three industries

These are starting points. Adjust stage names to the words your team already uses.

Industry	Stages (in order)	Example probabilities	Watch for
Contractors and remodelers	Inquiry → Site visit booked → Estimate sent → Negotiation → Won / Lost	10% → 30% → 45% → 70%	Estimates that go quiet after three days; follow up before the homeowner books another contractor
Realtors (buyers)	New inquiry → Showing booked → Offer submitted → Under contract → Closed / Lost	5% → 20% → 50% → 85%	Long searches; buyers may need weeks of nurturing before an offer
Lawn care and home services	Quote requested → Estimate booked → Quote sent → Scheduled → Active customer / Lost	15% → 40% → 55% → 90%	Seasonal timing; quotes requested in early spring close fastest when you schedule the first visit on the call

A few notes on these examples:

• Realtors often run two pipelines: one for buyers and one for listings, because a seller’s journey (valuation, listing agreement, on market, offer accepted) is different.
• Contractors with both small repairs and big remodels may want separate pipelines, since a $400 repair and a $60,000 kitchen don’t move at the same speed.
• Lawn care businesses care about recurring revenue, so “Active customer” is the real finish line, not the first mowing.

You’ll find more detail on each workflow on our contractors, realtors and home services pages.

Common pipeline mistakes

• Too many stages. If you need a legend to explain the board, merge stages.
• Stages based on your activity, not the buyer’s. “Called twice” isn’t a stage. “Agreed to a site visit” is.
• No lost stage. Without one, dead deals sit in “Quote sent” forever and inflate your forecast.
• Probabilities nobody updates. Revisit them every quarter using actual win rates.
• Quotes living outside the pipeline. When quotes are tied to deals, you can see which quotes are waiting and follow up. Our guide on writing a winning quote covers that step.

Final thoughts

Building a pipeline isn’t about software first. It’s about agreeing on how your customers buy, writing down when a deal really moves forward and keeping every deal honest with a value, an owner and a next step. Do that, review it weekly and adjust your probabilities with real data, and your pipeline becomes the most useful page in your business. If you want to set one up with custom stages, weighted totals and stalled-deal alerts, see Exa’s pipeline features.