---
title: "The 15-minute referral system for agencies of 3 to 15 people"
url: https://theagencyarithmetic.com/pipeline-crm/referral-system/
dateModified: 2026-11-03
---

# The 15-minute referral system for agencies of 3 to 15 people

Two in three of the 612 agency owners and consultants in SparkToro's 2024 survey name client referrals as their top source of new business. In its 2025 survey, fewer than one in six call their pipeline healthy. Referrals can't be switched on, but two things about them can be managed: whether clients get asked, and what happens to a referral in the days after it arrives. Both fit in 15 minutes a week.

(Third-party data)

- By [Vinicius Diniz](https://theagencyarithmetic.com/about/)
- Published October 30, 2026
- Prices checked on October 30, 2026
- [Methodology](https://theagencyarithmetic.com/methodology/)

## How much do agencies depend on referrals?

Between 66% and 75% of agencies and service firms name referrals as their best source of new clients, depending on the survey. Only 14% of agencies describe their pipeline as healthy.

66% to 75% of agencies and service firms name referrals as their best source of new clients.

Source: SparkToro, State of Digital Agencies 2024, 612 agency owners and consultants, 2024 survey, period not stated; RSW/US, 2025 Survey Report, 2025 survey, period not stated (the 75% includes professional services firms). [See all benchmarks](https://theagencyarithmetic.com/benchmarks/)

14% of agencies describe their sales pipeline as healthy.

Source: SparkToro, State of Digital Agencies 2025, 376 agency owners and consultants, Sep to Oct 2025. [See all benchmarks](https://theagencyarithmetic.com/benchmarks/)

| Survey | Finding |
| --- | --- |
| SparkToro, State of Digital Agencies 2024 (612 respondents) | 66% named referrals from current and past clients as their top source of new business |
| RSW/US, 2025 survey report (published September 2025) | 75% cited referrals as their best lead source |
| SparkToro, State of Digital Agencies 2025 | 14% describe their pipeline as healthy; 32% say it is not very good |
| SparkToro, State of Digital Agencies 2025 | 70% have no full-time salesperson |

RSW/US sells outsourced new business development, which is one reason this page puts its number next to an independent one.

If referrals bring most of your clients and your pipeline feels thin, your agency is in the middle of the range, not at the bottom of it.

## Why referrals don't happen, and why some go nowhere

In earlier Hinge research on buyers, cited in its 2015 report, 72% of those who hadn't referred a provider said nobody had asked. About half of the professionals in a 2015 Hinge survey of 523 service firms had ruled out a referred firm before speaking with it; the top reason, cited by 43.6%, was not being able to tell what the firm does.

72% of buyers who hadn't referred a provider said nobody had asked, in earlier Hinge research.

Source: Hinge Research Institute, Inside the Buyer's Brain, cited in Referral Marketing for Professional Services Firms (2015) (buyers of professional services, not agency clients). [See all benchmarks](https://theagencyarithmetic.com/benchmarks/)

51.9% of professionals had ruled out a referred firm before speaking with it.

Source: Hinge Research Institute, Referral Marketing for Professional Services Firms, 523 professional services firms, 2015 (professional services firms, not only agencies). [See all benchmarks](https://theagencyarithmetic.com/benchmarks/)

Three questions about your last six months:

1. How many current clients did you ask, in so many words, whether they know someone you could help? If the answer is none, the first gap is the ask.
2. When the last referral arrived, how many business days passed before your first reply?
3. Did the person who referred them hear back from you, both when you replied and when it closed or didn't?

## How to ask: what field experiments found

How you ask changes how many referrals you get. Three studies outside agencies, each with a field experiment, point the same way: ask personally, frame the referral around what the other person gets, and start with clients who came to you by referral.

(Third-party data) Not agencies

| Finding | Study |
| --- | --- |
| Describing the referral as a benefit to the friend, with the same incentive, produced 86% more referrals than the control wording | Jung, Bapna, Golden and Sun, Information Systems Research, 2020. About 100,000 customers of an e-commerce platform, as summarized by co-author Ravi Bapna in Knowledge at Wharton |
| A personal request was more than twice as effective as a generic one | Benjamin, Kontra, Brucks and Simonson, field experiment at an athletic club, summarized by Stanford GSB, 2017 |
| Customers who joined by referral made 31% to 57% more referrals; reminding them how they joined raised referrals by about 21%, an effect the authors call very small in absolute number | Gershon and Jiang, Journal of Marketing Research, 2024. 41.2 million customers of a cashback company, 10 years of data, replicated in 19 of 20 companies |

None of these studies looked at agencies. Adapted to an agency, the ask looks like this:

Adapted from consumer research, not tested on agencies

> You came to us through \[the person who introduced you\] , and it's worked out well on our side. If someone you know is dealing with \[the problem you solved for this client\] , I'd be glad to give them a 30-minute look at it, no strings. Just reply with a name and I'll take it from there.

## What a steady ask is worth

At 11 clients asked twice a year, a steady ask can plausibly produce one or two new clients a year. The math depends on one number nobody has measured for agencies: how often an ask turns into an introduction.

Assumptions

- **Current clients Planning assumption**: 11
- **Asks per client per year Planning assumption**: 2
- **Asks that turn into a named introduction Planning assumption, the weakest number here**: 1 in 5
- **Introductions that become clients Planning assumption, borrowed from SparkToro's qualified-lead conversion (the most common range, reported by 39% of agencies); referral introductions were not measured separately**: 25% to 49%
- **Value of one client Planning example**: $6,800 a month for 24 months

Result

- **Total asks: 11 × 2**: 22
- **Introductions per year: 22 ÷ 5**: 4.4
- **New clients per year (2.16 before rounding at the high end)**: 1.1 to 2.2
- **Value of one client: $6,800 × 24**: $163,200
- **Revenue over the relationship, from the unrounded client counts**: about $180,000 to $352,000
- **Time: 22 asks × 15 minutes**: 5.5 hours a year
- **New clients per year at 1 in 10**: about 0.6 to 1.1

The value of one client is the same planning example as the [cold referral cost calculator](https://theagencyarithmetic.com/pipeline-crm/cold-referral-cost-calculator/).

At 1 in 10, the result halves: about 0.6 to 1.1 new clients a year. Even at the low end, one new client pays for the hours many times over.

## The system: five stages, three rules, 15 minutes

The system is a list of every open conversation, three rules for what happens to it, and one 15-minute review each week. It can live in a spreadsheet or a free CRM.

**Five stages, no more:**New conversation · First call done · Proposal sent · Waiting · Won or lost.

**Three rules:**

1. Every new conversation goes in the list the day it arrives, even if the reply has to wait.
2. Every proposal gets a reminder 3 business days later, and a second one a week after that.
3. Everyone who refers a client hears from you twice: when you reply to the referral, and when it closes or doesn't.

**The weekly 15 minutes (pick one day):**

- Move every conversation to its real stage.
- Send the follow-ups that are due.
- Under each current client, write one line: who might they know, and when will you ask?

**Where to keep it.** A spreadsheet works for a few conversations a month. HubSpot's free CRM covers 2 users and 1,000 contacts with no time limit (Vendor spec, checked October 30, 2026); the limits that end it are a third user, more than 1,000 contacts, or more automated email follow-up than the free plan's single automated action. Pipedrive has no free plan; its entry plan is $14 per seat per month billed annually, with a 14-day trial (Vendor spec, checked October 30, 2026). Agency Arithmetic has not tested either hands-on.

## When this system won't help

Skip it, or expect little from it, in these cases:

- **Fewer than 5 current clients.** At 4 clients, two asks a year each and the planning rate above, the math gives 1.6 introductions and fewer than one new client a year.
- **Your clients don't like the work.** Asking unhappy clients for referrals costs more than it brings. Fix delivery first; [how long clients stay at small agencies](https://theagencyarithmetic.com/client-reporting/how-long-clients-stay/) covers why clients leave.
- **You sell mostly by outbound.** The system protects referrals. Cold outreach needs a different process.
- **You can't say in one sentence what you do and for whom.** About half of the professionals Hinge surveyed had ruled out a referred firm before talking to it, most often because they couldn't tell what it did. Write that sentence before asking anyone.

## The one-afternoon start

Put every open conversation in one list and send one ask, this afternoon.

1. Search your inbox for the last 90 days and list every conversation with a potential client. Give each one of the five stages.
2. Send the follow-ups that are already overdue.
3. Write your one-sentence description: what you do, for whom.
4. Pick one current client who came to you by referral and send them the adapted ask.

More on [pipeline and referrals for agencies of 3 to 15 people](https://theagencyarithmetic.com/pipeline-crm/), with every sourced number in [the benchmarks behind this site](https://theagencyarithmetic.com/benchmarks/).

## Sources

- [SparkToro, “The State of Digital Agencies in 2024: Client Referrals Drive Agency New Business…”, January 16, 2025](https://sparktoro.com/blog/the-state-of-digital-agencies-in-2024-client-referrals-drive-agency-new-business-but-is-it-enough-and-are-sales-pipelines-a-cause-for-concern/). Accessed October 30, 2026.
- [SparkToro, “Digital Agency New Business Still a Concern, Referrals Still Rule for Lead Gen…”, February 17, 2026](https://sparktoro.com/blog/digital-agency-new-business-still-a-concern-referrals-still-rule-for-lead-gen-and-are-sales-pipelines-finally-recovering/). Accessed October 30, 2026.
- [RSW/US, “2025 Survey Report”, press release, September 4, 2025](https://www.rswus.com/survey/2025-rsw-us-survey-report-rolling-into-2026/). Accessed October 30, 2026.
- [Hinge Research Institute, “Referral Marketing for Professional Services Firms”, 2015](https://hingemarketing.com/wp-content/uploads/2016/03/hinge-research-referral-marketing.pdf). Accessed October 30, 2026.
- [Jung, J., Bapna, R., Golden, J. M., and Sun, T. “Words Matter! Toward a Prosocial Call-to-Action for Online Referral”, Information Systems Research, 2020; summarized by Ravi Bapna in Knowledge at Wharton, June 27, 2019](https://knowledge.wharton.upenn.edu/article/how-to-increase-word-of-mouth-referrals/). Accessed October 30, 2026.
- [Stanford GSB, “The Personal Touch, Quantified”, May 31, 2017](https://www.gsb.stanford.edu/insights/personal-touch-quantified). Accessed October 30, 2026.
- [Gershon, R., and Jiang, Z. “Referral Contagion: Downstream Benefits of Customer Referrals”, Journal of Marketing Research, 2024 (author accepted manuscript)](https://faculty.wharton.upenn.edu/wp-content/uploads/2020/06/gershon-jiang-2024-express-referral-contagion-downstream-benefits-of-customer-referrals.pdf). Accessed October 30, 2026.
- [HubSpot, “CRM pricing”, checked October 30, 2026](https://www.hubspot.com/pricing/crm). Accessed October 30, 2026.
- [Pipedrive, “CRM pricing plans”, checked October 30, 2026](https://www.pipedrive.com/en/pricing). Accessed October 30, 2026.

Correction, November 3, 2026: The opening attributed SparkToro's 2024 referral figure to small agencies. The survey had 612 agency owners and consultants, with no cut by size.
