How to Put Referrals in Your 2027 Pipeline Plan as a Line Item

If you have to double pipeline next year, referrals can't stay a nice-to-have. A four-step model to forecast them from a one-month pilot, with a worked example and a deliberate haircut.
Shankar Ganapathy
Co-Founder, Boomerang

It is planning season. Somewhere in your 2027 plan there is a pipeline number that is a lot bigger than this year's, and a list of channels that are supposed to produce it. Paid, events, outbound, partners. Referrals are usually in there too, as a sentence rather than a number: "leverage customer advocacy."

A sentence does not get budget. A line item does. Here is how to turn referrals into one, using a short pilot to replace guesses with your own data.

Why referrals rarely get a forecast

Referrals are hard to plan because they have always been hard to count. They arrive through the AE's inbox or the CEO's phone, they are rarely tagged, and nobody records the asks that went nowhere. So the plan either ignores them or puts in a hopeful number nobody believes.

The fix is to measure a small, deliberate version of the program for a month and build the forecast from that. Four steps.

Step 1: Measure your baseline

Before you change anything, count what you get today. Go back six months and find the introductions that turned into first meetings, by source:

  • Employees and executives
  • Customers
  • Investors, board members and advisors
  • Partners

Divide by six. That is your monthly baseline. If the count is shaky because referrals were never tagged, start tagging now (source type, who referred, asked or unprompted, date, outcome) and treat the first month of clean data as the baseline. We cover the fields in how to pilot a referral program before you buy a tool.

Step 2: Run a one-month pilot on a slice of your network

Pick roughly 10 percent of your possible connectors: your leadership team, your best-connected sellers, and the customers who have had a good moment recently, such as a strong QBR, a renewal or a go-live. Run a structured program with them for a month. Specific asks, to specific people at your target accounts, timed to good moments, tracked to the meeting.

Record three things from the pilot group: how many asks went out, how many became introductions, and how many introductions became meetings.

Step 3: Extrapolate, then cut it in half

Now scale the pilot result to the whole network and to a year, and then apply a deliberate haircut. The first 10 percent you pick will be your most willing connectors, so the rest will not perform as well. Halving the result is a simple, defensible way to account for that.

Step 4: Convert meetings into pipeline

Multiply by the meeting-to-opportunity rate and average opportunity size you already use for other channels. That gives you a pipeline number in the same units as the rest of your plan.

A worked example

These numbers are made up to show the arithmetic. Use your own.

  • Baseline: 6 referral meetings a month today. 72 a year without changing anything.
  • Pilot: 50 connectors (10% of 500) produce 15 meetings in the month. 0.3 meetings per connector per month.
  • Extrapolate: 0.3 x 500 connectors x 12 months. 1,800 meetings a year.
  • Haircut: Cut by half for less engaged connectors. 900 meetings a year.
  • Realistic cadence: Customers are asked two or three times a year, not monthly; assume the program sustains a third of the pilot pace. 300 meetings a year.
  • Pipeline: 300 meetings x 30% become opportunities x $40,000 average. $3.6M of pipeline.

Two things to notice. First, the cadence line matters as much as the haircut. People can be asked only so often, and a forecast that assumes the pilot pace forever will be wrong. Second, the answer is a range, not a promise. Put the halved number in the plan and keep the unhalved one as upside.

If your total pipeline target is, say, $15M, this example puts referrals at roughly a quarter of it. That is the kind of share worth discussing when you are trying to double, because it is pipeline from people who already trust you rather than more spend on the same cold channels.

What the line item needs to survive the budget meeting

  • An owner. Usually marketing for the program and reporting, with CS owning customer asks and the CEO's office owning executive and board asks.
  • A cadence per group. Roughly one ask a week for employees, two or three a year per customer champion, about one a month for investors and board members, and partners when a shared account shows a signal.
  • Its own reporting. Referral source on every opportunity, so the number in the plan can be checked against what happened each quarter.
  • A cost. People's time, any incentives, and any software. Without it, finance cannot compare the channel with the others.

Where software fits

The pilot runs fine by hand. The full-year number usually does not, because nobody can work out by hand which of 500 connectors knows which of your target accounts, draft every ask and chase every reply. That is the point at which a tool pays for itself, and the pilot gives you the number to judge it against.

That is what Boomerang does. Rudy, our agent, maps who across your employees, customers, investors and partners knows your target accounts, drafts each ask for the person who holds the relationship to send from their own account, and writes the outcome back to your CRM so the line item reports itself. Armis created 26,000 warm-intro paths with Boomerang and saw 10x ROI on revenue booked in a year. If you want to run the pilot with us, book a time.

Frequently asked questions

How do you forecast referral pipeline?

Measure your current monthly referral meetings, run a one-month pilot on about 10 percent of your connectors, extrapolate the result to your whole network and a full year, cut it by half to allow for less engaged connectors, adjust for a realistic ask cadence, then convert meetings to pipeline with your normal opportunity rate and deal size.

Why halve the pilot result?

The connectors you pick first are your most willing ones, so the rest of the network will not perform as well. Halving is a simple, conservative way to account for that. Keep the unhalved number as upside.

How often can you ask the same people for referrals?

Roughly one ask a week for employees, two or three a year per customer champion, about one a month for investors and board members, and partners when a shared account shows a signal.

Who should own referrals in the pipeline plan?

Usually marketing owns the program and reporting, customer success owns customer asks, and the CEO's office owns executive and board asks. One person should own the number.

What share of pipeline can referrals realistically be?

It depends on the size of your network and how often you can ask. Build the number from your own pilot rather than a benchmark, and plan on the conservative figure.