---
title: "How to Prove Social Media ROI When the Client Doesn’t Track Revenue"
description: "Most small clients can’t tell you which customers came from social or what they were worth. You can still put a defensible number on it: count the conversations, get the client to confirm a short list each month, and build the value from their own answers. A worked example, the formula, and how to present estimates honestly."
author: "Kaitlyn Bricker"
author_url: "https://www.inexra.com/about"
date: "2026-10-14"
updated: "2026-10-14"
canonical: "https://www.inexra.com/resources/prove-social-roi-without-revenue-tracking"
---

# How to Prove Social Media ROI When the Client Doesn’t Track Revenue

When a client doesn't track revenue, build the number from the client's own answers. Count the conversations social started each month, ask the client which of a short list became customers, and ask what a typical customer is worth to them. Those two answers give you a close rate and an average sale, which turn conversations into an estimated value the client has effectively signed off on. Show the formula and label which numbers are confirmed and which are estimates.

That's the method. The rest of this guide is a worked example and the mistakes that make clients stop trusting the number.

## Why the usual advice doesn't fit

Most guides to social ROI assume one of two things: that the client has revenue data you can attribute, or that leads happen on a website where a form fill or a purchase can be tracked. Neither is true for a lot of small clients. A plumber, a florist or a physio takes bookings by phone and DM, keeps the books in an app the agency never sees, and couldn't tell you their close rate if you asked.

The fallback most agencies use is engagement. Hootsuite's [2023 survey](https://www.hootsuite.com/newsroom/press-releases/hootsuite-unveils-the-trends-that-keep-on-trending-for-social-marketers) found 69% of social marketers named engagement as their top metric for showing ROI, and Sprout Social's [2025 research](https://sproutsocial.com/insights/press/social-media-wins-the-budget-war-with-8-in-10-marketing-leaders-reallocating-funds-from-other-channels-to-social-signaling-major-shift-in-business-strategy/) found only 44% of marketing leaders rate their social team as expert at measuring business impact. Engagement is easy to report and hard for a business owner to care about. "Likes were up 12%" doesn't answer whether the retainer paid for itself.

## The one hard number you can always own

Even when you can't see revenue, you can see conversations. Every DM or comment thread that asked about buying, booking or pricing is a fact: it happened, it's in the inbox, and you can usually say which post started it. That count is the foundation. Everything else is built on top of it.

Define a lead with the client before you start, so the count means the same thing to both of you. A good default: someone who asked about buying, booking or pricing, with enough detail that a reply could move them forward. Then track each one with the post that started it. There's a template for this in [how agencies prove Instagram brings in leads](https://www.inexra.com/resources/prove-instagram-brings-in-leads).

## Get the client to confirm a short list

Once a month, send the client the list of that month's leads (names or handles, what they asked, the date) and ask: "Which of these became customers, and roughly what were they worth?"

This takes the client five minutes and does more for your credibility than any dashboard. You aren't asking them to export data or install anything. You're asking them to recognise people they already dealt with.

A few practical notes:

- **Keep the list short.** Twenty names is fine. If there are two hundred, send the twenty most promising and say so.
- **Ask for ranges, not exact figures.** "Somewhere between $300 and $500" is an answer most owners will give. "Exactly what did they spend?" often gets no reply.
- **Accept "I don't know" for some rows.** Mark them unknown, don't guess.

## A worked example

Here's a quarter for a catering client who doesn't track where customers come from.

|                                       | Month 1 | Month 2 | Month 3 | Quarter |
| ------------------------------------- | ------- | ------- | ------- | ------- |
| Conversations asking to book or price | 18      | 24      | 21      | 63      |
| Client confirmed as customers         | 4       | 6       | 5       | 15      |
| Client said "don't know"              | 3       | 2       | 4       | 9       |

From the client's answers, the typical job is $400 to $600, so call it $500.

**Confirmed value:** 15 customers × $500 = **$7,500**. This is the number the client has effectively confirmed: they named the customers and gave the job size.

**Close rate:** 15 confirmed out of the 54 leads the client could answer for = about 28%.

**Estimated value of the rest:** the 9 unknowns × 28% × $500 = about **$1,250**. This is an estimate, built from the client's own close rate.

The formula is the one most ROI guides use (leads × close rate × average sale, as in [AgencyAnalytics' example](https://agencyanalytics.com/blog/social-media-roi)). The difference is that every input came from the client, not from an industry benchmark they've never heard of.

## Present confirmed and estimated value separately

This is where most agencies lose the client's trust, and [lost trust in the numbers](https://www.inexra.com/resources/why-social-media-agency-clients-leave) is how renewals slip: they blend everything into one big number. Keep the layers visible.

> **This quarter, 63 people asked about booking through Instagram.**
> You confirmed 15 became customers, worth about **$7,500** at your typical job size.
> If the 9 you weren't sure about converted at the same rate, that's roughly another **$1,250**.
> Your retainer for the quarter was $3,600.

The client can see exactly what's confirmed and what's estimated, and they supplied the numbers behind both. If they push back on the estimate, you lose $1,250 of the story and keep $7,500 of it. If you'd blended them, a challenge to any part would undermine the whole thing.

Notice the last line. Putting the retainer next to the value is uncomfortable the first time and very effective after that, provided the value is real.

## Proxies, ranked by how close they are to money

If the client won't confirm customers at all, you're left with proxies. Use the ones nearest to a sale and say plainly that they're proxies.

1. **Conversations asking to buy or book**, with the post that started each. Closest to money and fully in your control to count.
2. **Booking requests or quote requests**, if the client uses a booking tool or form you can see.
3. **Clicks on tracked links to a sales or booking page.** Real intent, but only from the people who click.
4. **Calls**, if the client will note "found you on Instagram" or you can use a tracked number.
5. **Branded search growth.** Useful supporting evidence, but correlational. Lots of things raise branded search.
6. **Reach, followers and engagement.** Context, not proof.

## When to set this up

At kickoff, if you can. Agree the definition of a lead, a monthly target ("30 booking enquiries a month"), and the monthly confirmation question before the first post goes out. Then every report has the same structure, the renewal conversation is just reading the result, and the number belongs to both of you. For what to bring to that conversation, see [what to show a client at the renewal call](https://www.inexra.com/resources/what-to-show-clients-at-the-renewal-call).

## FAQ

### How do I prove social media ROI if my client doesn't track sales?

Count the conversations social started that asked about buying or booking, ask the client each month which of them became customers and roughly what they were worth, and calculate value from those answers. Report the confirmed value and any estimate separately.

### How do I calculate the value of a lead from Instagram?

Multiply the client's close rate by their average sale. If the client doesn't know their close rate, work it out from a few months of asking which leads became customers. For example, if 15 of 54 leads became customers at about $500 each, each lead is worth roughly $140.

### What metrics should I use instead of revenue for a service business?

Conversations asking to buy or book, booking or quote requests, clicks on tracked links to a booking page, and calls, in that order. Reach and engagement are context, not proof of return.

### Is engagement a good enough metric to show clients?

Not on its own. It's easy to report but doesn't tell a business owner whether the work paid for itself. Lead with conversations and customers, and use engagement to explain why those numbers moved.