# Agency Reporting and KPIs

> What an agency should report at months one, three, six and twelve, how to read attribution it controls, and the dashboard spec that ends vanity metric theatre.

Source: https://saas-marketing.net/guides/saas-agency-reporting-and-kpis/
Topic: SaaS Marketing Agencies
Type: guide
Published: 2026-09-11
Last updated: 2026-09-11
Publisher: SaaS Marketing (saas-marketing.net)
License: CC BY 4.0. Quote or republish with attribution and a link to https://saas-marketing.net/guides/saas-agency-reporting-and-kpis/

## Short answer

A SaaS marketing agency should report leading indicators in months one to three, mixed indicators by month six, and pipeline created and revenue influenced by month twelve. The reporting contract belongs in the statement of work: named metrics per service line, a dashboard built on your data rather than theirs, a self reported attribution question at signup as a cross check, and a banned list covering impressions, average position and MQLs without downstream conversion rates.

## Key takeaways

- Pay for tracking and reporting setup in the first 30 days, before any campaign spend, or month six becomes an argument about numbers.
- Months one to three are leading indicators only. Demanding pipeline at month two produces fabricated pipeline.
- Never accept a report built exclusively in a platform the agency controls. The dashboard should read from your CRM.
- A self reported attribution question at signup routinely disagrees with platform attribution by 20 to 40 points, and both are useful.
- Sessions falling while demo requests hold is now the normal pattern in AI search, and a reporting contract must anticipate it.
- Tie renewal to pipeline created rather than traffic delivered, and write the threshold into the contract before month one.

---

The reporting argument always happens at month six. The agency shows a slide with rankings up and traffic up, the CEO asks where the revenue is, and nobody can answer because the tracking was never built to answer it.

Fix this in the statement of work, before the first campaign runs. What follows is the reporting contract I would write, month by month.

## Set the reporting contract before the work starts

Reporting requirements belong in the SOW alongside deliverables, with the same specificity. Name the metrics, name the source system, name the cadence, and name who owns the accounts.

The single most valuable clause is this one: the first 30 days of the engagement are spent on tracking and reporting infrastructure, paid for at the normal rate, before any campaign or content spend. Agencies resist it because it delays visible output. Accept the delay. An engagement that starts producing content in week two and gets measurement right in month five has five months of unattributable work.

Every analytics property, tag manager container, ad account, dashboard and CRM integration is created under your organisation's accounts, with the agency added as a user. Write it into the SOW. The day you change agencies, this clause is worth more than the rest of the contract.

Put the specifics in the [SaaS agency brief template](/templates/saas-agency-brief/) at the pitch stage, so you find out during procurement which agencies flinch.

## What to expect at month one

Nothing that looks like a result, and everything that makes results measurable later. Month one is an evidence report, not a performance report.

Expect: tracking implementation verified with test conversions, baseline metrics recorded and frozen in a document both sides sign, CRM fields agreed for opportunity source, the target keyword or account list approved, a content or campaign calendar for months two to four, and the first assets in production.

What you should refuse: any pipeline number, any claim about improvement, and any report that arrives as a PDF export from an ad platform.

## Month three: leading indicators, honestly labelled

By month three you can read direction, not outcome. Different service lines mature at different speeds and the report should say so explicitly.

Hold the agency to the middle column at month six. The right hand column is the renewal conversation.

## Month six and month twelve: the accountability report

Month six is where a report has to show conversions, not just traffic. Ask three questions and expect specific answers: which pages or campaigns produced demo requests, what the cost per qualified opportunity is, and what the agency intends to change based on that.

Month twelve is judgement. The figure that matters is pipeline created, sourced from your CRM, against the total cost of the retainer plus media spend. If the ratio is below 3:1 in pipeline terms for a mid market SaaS business, the engagement is not working, and the [agency retainer ROI calculator](/calculators/agency-retainer-roi/) will give you the number without a spreadsheet argument.

**3:1** Minimum pipeline created to total agency cost ratio worth renewing on for mid market SaaS

Two fairness points. Long sales cycles mean a twelve month review on a 150,000 dollar ACV product is reading pipeline, not closed revenue, and that is legitimate. And an agency inheriting broken tracking should be judged from the date tracking was fixed, not from the contract start date.

## Attribution the agency controls versus what the CRM says

Every ad platform over-reports its own contribution. This is not fraud, it is the consequence of each platform seeing only its own touchpoints and claiming credit within its own attribution window. A report assembled from platform dashboards will exceed reality, sometimes by a factor of two.

The correction is a single source of truth in your CRM, with an opportunity source field that sales is actually required to complete. Then add a cross check the platforms cannot influence: a free text how did you hear about us question on the signup or demo form.

**Setting up the cross check**

Self reported and platform attribution routinely disagree by 20 to 40 points. Anyone presenting one of them as truth is either selling a tool or has not looked. [Measuring SaaS content performance](/guides/saas-content-performance-measurement/) covers the content specific version of this problem.

## The banned metrics list

Write this into the SOW. Every metric here can improve while the business gets nothing, which is what makes them attractive in a report.

**Metrics that should not appear alone in an agency report**

MQL count is the worst offender because it looks like a business metric. An agency can double MQLs by loosening the definition, and the only defence is demanding the downstream conversion rate in the same table. The definitional work sits in [demand generation metrics for SaaS](/guides/demand-generation-metrics/).

## The dashboard spec to require in the SOW

Specify the dashboard, not just the report. A monthly slide deck is a narrative; a dashboard is a fact.

The spec: built in a tool your organisation owns, reading from your CRM as the primary source and platforms as secondary, refreshed at least daily, accessible to your team without going through the agency, and containing a fixed set of views that do not change month to month.

| Dashboard view | Primary source | Refresh | Who reads it |
| --- | --- | --- | --- |
| Pipeline created by source | CRM | Daily | CEO, CRO |
| Cost per opportunity by channel | CRM plus ad platforms | Weekly | Marketing lead |
| Conversion rate by landing page | Analytics plus CRM | Daily | Agency and CRO lead |
| Content performance by page | Search Console plus CRM | Weekly | Content lead |
| Self reported attribution tally | Form data | Weekly | Marketing lead |
| Spend pacing against budget | Ad platforms | Daily | Finance |

Fixed views matter because changing dashboards is how underperformance gets hidden. If the month six report contains three charts that did not exist in month three, ask what happened to the original three.

## When sessions fall and demo requests hold

This is now the normal pattern, not an anomaly. AI Overviews appear on roughly half of queries in published tracking studies and absorb the informational clicks, while commercial intent traffic largely survives. A report that leads with a traffic decline is telling you about Google's interface, not about your programme.

Restructure the reporting baseline. Lead with conversions and pipeline, show organic sessions split into branded and non branded, and add a citation tracking view: which AI answers name your brand for your priority queries. That last one is imperfect and worth doing anyway, because it is the only visibility you have into a surface that increasingly precedes the click.

A team I know cut their content retainer in half after a 30 percent organic session drop in 2025. Demo requests from organic were flat throughout. They cut the programme that was still working because the metric on the front page of the report was the wrong one.

## Tying renewal to the right number

Write the renewal criteria at signing, not at month eleven. The version I would use: renewal is contingent on pipeline created from agency owned channels reaching an agreed multiple of total cost by month twelve, measured in the client's CRM, with the multiple set according to sales cycle length and ACV.

That single clause changes agency behaviour from month one, because it makes the CRM the scoreboard. It also protects the agency from the opposite failure, a client who moves the target every quarter.

If the numbers keep failing to justify the retainer, the alternative is not always a different agency. Run the [agency versus in house cost comparison](/calculators/agency-vs-in-house-cost/) before assuming the next retainer will be different, and read [SaaS growth agency](/guides/saas-growth-agency/) on what the model is genuinely good at. Where an agency holds the [agency of record](/glossary/agency-of-record/) position across several service lines, the reporting contract matters more, not less, because there is no second vendor whose numbers you can triangulate against, and the [marketing retainer](/glossary/marketing-retainer/) structure itself should be reviewed at the same time.

## What to do next

Open your current SOW and check three things: who owns the analytics accounts, which metrics are contractually required, and what the renewal is tied to. Most contracts are silent on all three.

Then add the self reported attribution field to your signup form this week. It takes an hour, costs nothing, and within a month you will have a cross check that no platform dashboard can influence. Broader guidance on selecting and managing partners sits in [SaaS marketing agencies](/saas-marketing-agencies/).

## Frequently asked questions

### What should a marketing agency report in the first month?

Setup evidence and leading indicators only: tracking implementation completed, baseline metrics recorded, keyword and account targeting agreed, assets in production, and the first campaigns live with spend pacing. Any agency reporting pipeline results in month one is either counting pre-existing demand or inventing numbers.

### What KPIs should a SaaS company hold an agency to?

Pipeline created, qualified opportunities, cost per opportunity and content-sourced signups are the accountability metrics. Rankings, sessions, impressions and engagement rate are diagnostic metrics used to explain those results. Confusing the two is the most common failure in agency relationships.

### How do you check agency reported attribution?

Compare the agency dashboard against your CRM opportunity source field, and add a free text how did you hear about us question at signup. All three will disagree. The pattern of disagreement tells you more than any single number: consistent under-reporting in self reported data usually means brand and word of mouth are doing work no platform can see.

### What metrics should be banned from agency reports?

Impressions without click data, average keyword position as a standalone figure, social engagement rate with no downstream action, MQL counts without SQL conversion rates, and any metric shown as a percentage change with no absolute number beside it. Each can rise while the business gets nothing.

### What if sessions drop but demo requests hold steady?

That is now common. AI Overviews appear on a large share of B2B queries and absorb informational clicks while leaving commercial intent traffic intact. The right response is to shift the reporting baseline to conversions and citations rather than sessions, and to check whether the lost traffic was ever converting in the first place.

### Who should own the reporting dashboard, the agency or the client?

The client. The data sources, the dashboard tool account and the tracking implementation should all be in accounts you own, with the agency given access. Agencies that insist on owning the tracking setup are creating a switching cost, and you will discover it on the day you try to leave.

### When can you fairly judge an SEO or content agency?

Month six for early signal, month twelve for judgement. Rankings and impressions should move by month three or something is wrong, first meaningful conversions by month six, and pipeline contribution that justifies the retainer by month twelve. Judging at month three cuts good programmes and judging at month eighteen funds bad ones.