Portfolio Results

An agency retainer gets tested by two questions. What changed in the measurement window after the team finished the work? What keeps being checked without someone remembering to do it?

The Results and Automations tabs in Agency → Actions answer those questions across every client brand at once. Results starts with the measured verdict and attaches the client and page. Automations shows which checks are active for each client, what they are allowed to do, and what they found in the last 30 days.

This is not a highlight reel. It is a full accounting. Good news appears first, but no movement, a drop, and an answer of Couldn't measure stay visible. That honesty is what makes the view useful in a renewal call.

Open the two lenses

Go to Agency → Actions. The page has three tabs:

• Work shows open and completed Actions across the portfolio.
• Results shows measured outcomes from finished Actions.
• Automations shows active and paused Automations by client.

Any brand filter applied on the page also scopes Results and Automations. Use that when you are preparing one client update. Clear it when you want the full portfolio view.

Results: verdict first, client attached

Results groups the table by outcome. Within each group, every row says what changed, names the client, links the page when one exists, and shows when Trakkr measured it. Select a row to return to the Action that produced the result.

The four headline counts are filters. Select Earned, No movement, Coincided with a drop, or Couldn't measure to isolate that group. Work-type filters use the same words as the product: Technical fix, Page improvement, Content refresh, New content, Outreach draft, Reply draft, Campaign, Setup, and Watch.

What the result words mean

Earned means the chosen measure improved between the before and after readings. The row states the numbers and the fixed window. It does not claim the Action caused the change.

No movement means the measure held steady over the window. It is a completed answer, not unfinished work. Use it to decide whether to change the approach, wait for a longer-term signal, or stop investing in the same idea.

Couldn't measure means Trakkr could not make an honest before-and-after comparison. The reason is part of the result. It does not mean the Action failed.

Coincided with a drop means the measure fell in the same window. Trakkr does not say the Action caused the fall. Search demand, site changes, competitors, model changes, and other work may have moved at the same time.

Reverted means a change that coincided with a drop was rolled back. Keep it in the client record when the client needs the full history. Keep the technical detail internal when it would distract from the decision and the current state.

Why a portfolio can be mostly Couldn't measure

A real portfolio can be mostly Couldn't measure. This is common early in an engagement and it is usually a tracking gap, not a wall of failed work.

The most common reasons are:

• Crawler tracking is not connected. Trakkr cannot see visits from AI crawlers to the affected page.
• Search Console is not connected. Trakkr cannot compare clicks and search activity for the page.
• The Action is not linked to a page. There is no target on which to take the before and after readings.
• The page had no activity in the window. A connected source was present, but there was no signal on that page before or after.
• Citation history was not available. The relevant citation snapshots did not cover the measurement window.
• There was no before reading. Work finished before Trakkr had a baseline to compare.

When unmeasured rows dominate, Trakkr groups them and explains the main reason once. It does not place a green mark beside each row. A long list of unmeasured work can look like success or failure depending on its styling, and neither reading is honest. The grouped explanation tells you what is missing and how many clients it affects.

Turn missing measurements into real answers

Work through this per client, not once for the agency:

1. 1.Connect crawler tracking. Add the Trakkr tracking setup supported by the client's site stack. Confirm that crawler visits reach Trakkr before relying on Reached.
2. 2.Connect Google Search Console. Use the property that covers the exact client domain and pages. This supports click and visit measures used by Visited.
3. 3.Check page links on Actions. Each page-based Action needs the page it changed. Add the correct page before the measurement window closes.
4. 4.Let the baseline form. New connections cannot recreate every missing before reading. The next Action will have a cleaner starting point.
5. 5.Keep report cadence active. Citation and visibility history comes from completed reports. Gaps in that history can leave a window without enough evidence.

The Pages journey helps diagnose the same gap. Available, Reached, Understood, Relevant, Selected, and Visited show which stage has evidence and which connection supplies it.

Automations: what keeps checking

The Automations tab groups Automations by client. The headline counts show Active, Paused, No checks yet, and findings from the last 30 days.

Each row gives you:

• The Automation name and goal.
• Allowed to, which summarizes its effective permission by work type.
• The number of checks in the last 30 days, how many found something, and how many failed.
• Whether it is Active or Paused.
• The time of the latest check, or No checks yet.

Permission words matter because two Automations with the same goal can have different authority. Tells you reports what it saw. Suggest only brings findings and changes nothing. Prepare drafts creates work for approval. Do it, then tell me can make a change and report it. Handle it can act within the weekly limit. The table shows the effective level after team and client controls apply.

Use this lens to find three problems before a client does: an important client with no Automation, an Automation that has never checked, or an Automation with repeated failed checks. Findings are not the same as changes. A finding can become a Suggested action that still needs your decision.

Build a client update from the view

Filter to one client. The Results tab can copy a plain-text update based on the rows on screen. Treat that as a draft, then add the context only you know.

Put these in front of the client:

• Earned results with the before and after figures.
• No movement when it changes the next decision.
• A drop when it affects risk, scope, or the next step.
• The full count of measured and unmeasured work.
• The connection needed when Couldn't measure is material.
• Active Automations that support the service they are paying for.

Keep these internal unless they help the decision:

• Failed checks that were retried and left no client impact.
• Drafts that never reached approval.
• Permission details that describe your internal operating model.
• Technical rollback notes once the client-facing state is clear.

Never hide a poor outcome. The distinction is relevance, not comfort. The client should see every result that affects the account, while your team keeps low-level operating noise in its own review.

Access and setup

Portfolio Results is part of the Agency workspace. The Agency entry appears only for a person who identified as an agency or a team with Agency Mode enabled, and only after the agency entitlement is confirmed. Scale and Enterprise accounts, plus an eligible active Scale trial, normally supply that entitlement.

The view needs at least one accessible client brand. It becomes valuable with several. For meaningful Results, keep reports active, attach pages to page-based Actions, and connect crawler tracking and Search Console for each client. For meaningful Automations, set the work-type permissions, activate the intended Automations, and review any that have not checked yet.

Common questions

Why does Earned not say the Action caused the change?

Because a before-and-after comparison is not a controlled test. Results state what moved in the same window and stop there.

Is Couldn't measure an error?

No. It is a first-class result. Open the reason, fix the missing connection or page link, and make sure the next Action starts with a baseline.

Why do my headline counts stay larger than the rows after filtering?

The headline shows the available distribution across the selected clients. The table shows the slice you chose.

Can I send the Automations table to a client?

You can, but most clients need a shorter summary: what is being watched, what was found, and what needs a decision. Keep permission and failure detail for the account team unless the contract calls for it.

How fresh is this page?

Both tabs read stored results and Automation history. Opening the page does not start a new measurement or check.