What if the client never agreed on a specific metric at the start?

It is harder but not impossible. Look for the closest proxy the client already tracked before the engagement: inbound leads, response times, deal volume. Be open that it is a reconstructed baseline, not one formally agreed at the outset.

Should ROI reporting include vanity metrics alongside the core number?

Sparingly, and clearly labeled as context rather than the headline. Leading with a vanity metric when the core agreed number is weaker reads as deflection once the client notices. That damages trust more than reporting the honest core number would.

How often should ROI actually be reported, not only tracked?

Monthly is a common cadence for ongoing engagements, with a more detailed review quarterly. Reporting too rarely lets a client's sense of value drift with no chance to correct it. Reporting too often can bury a strong result in noise.

How to prove ROI to a client

Agree on the metric and the starting baseline before the engagement begins, not after the client asks for proof. An ROI argument built after the fact rarely convinces anyone, including the person making it.

app.salescrew.io/inbox

The unified reply inbox with classified threads

The short answer

Agree the metric, and the baseline it is measured against, with the client before the engagement starts. Building an ROI argument after the fact, once results are in, is far less convincing and often contested.
A single, clearly defined metric the client already cares about is more persuasive than a dashboard of many numbers. It is harder to dispute and easier to remember than an internally invented composite score.
Vanity metrics can add useful context but should not replace the core agreed number. Leading with a vanity metric when the core number is weak reads as deflection once noticed.
Regular reporting, commonly monthly with a deeper quarterly review, stops a client's sense of value drifting unchecked between the start and end of an engagement.

Why ROI arguments built after the fact rarely land

The most common mistake in proving ROI is building the case only once a client asks for it, usually near a renewal decision or after doubt has already set in. At that point, the baseline has to be reconstructed after the fact, often from imperfect records. The client has every reason to question whether that baseline is accurate or merely convenient.

Agreeing on the metric and the starting point before the engagement begins removes that ambiguity. If both sides agreed at the outset that success looks like a specific, measurable change from a specific, measurable starting point, the ROI conversation later becomes a matter of reading the agreed number, not defending a retroactively chosen one.

The metric itself matters as much as the timing. A metric the client already tracks and cares about, response time to inbound leads, qualified meetings booked, deals closed, is more persuasive than an internally invented composite score. That kind of score needs explaining before it can even be understood, let alone trusted.

What to set up before, not after, an engagement

Element	Set before the engagement starts
The core metric	One number the client already understands and cares about
The baseline	The metric's actual value before the engagement begins, recorded and agreed
The reporting cadence	How often the metric will be reported, and in what format
What counts as attributable	Which changes in the metric will be credited to the engagement versus other factors

What to actually do once reporting is underway

Report the number consistently, even in periods where it is not moving the right way. A client who only sees reporting during good months notices the gaps, and trusts the good-month numbers less as a result. If outside factors clearly affected the metric during a period, note them honestly rather than omitting the period. An obviously selective report undermines trust in every other number presented.

It also helps to separate, in the reporting itself, what the engagement directly caused from what it merely coincided with. A metric that improved during the same period as an unrelated change on the client's side, a new product launch, a seasonal shift in demand, deserves an honest caveat rather than an implied claim of full credit. Clients tend to trust a report more, not less, when it is careful about attribution. It signals the numbers are not being stretched.

Disclosure: SalesCrew is our product. Deals, meetings and pipeline stage history are logged with timestamps. That gives a team the raw record to build a consistent ROI report against an agreed baseline. The marketing module can connect cost-per-lead by channel to pipeline outcomes in the same database. It does not decide which metric matters to a specific client or set the baseline for you. That agreement happens directly with the client.

A metric chosen after the results are in looks exactly like it sounds

Choosing or redefining the success metric once results are known, even if the new metric is defensible, reads as moving the goalposts to a client who was not part of that decision. Lock the metric in writing before the engagement starts.

Questions

What if the client never agreed on a specific metric at the start?: It is harder but not impossible. Look for the closest proxy the client already tracked before the engagement: inbound leads, response times, deal volume. Be open that it is a reconstructed baseline, not one formally agreed at the outset.
Should ROI reporting include vanity metrics alongside the core number?: Sparingly, and clearly labeled as context rather than the headline. Leading with a vanity metric when the core agreed number is weaker reads as deflection once the client notices. That damages trust more than reporting the honest core number would.
How often should ROI actually be reported, not only tracked?: Monthly is a common cadence for ongoing engagements, with a more detailed review quarterly. Reporting too rarely lets a client's sense of value drift with no chance to correct it. Reporting too often can bury a strong result in noise.