How to run a tool

Summary#

Every tool in Metric Vault runs the same way: pick the tool, type one value, press Run, and read the report that appears below the button. Learning this once means you can use any of the tools without learning a new interface. This page covers the whole flow, including the prompt that offers you a previous run for free, what the loading and error states mean, and what happens to your credits at each step.

Purpose#

Metric Vault has more than seventy screens. If each one had its own way of taking input and showing progress, the product would be seventy separate things to learn. Instead, one runner drives them all: the same input row, the same Run button behavior, the same loading line, the same result container, the same export row. The differences between tools live in the data they return, not in how you operate them.

Requirements#

• A Metric Vault account, signed in. Anonymous visitors can use the public tools at metricvaultai.com/free-tools, but the dashboard tools require an account. See Free tools.
• For anything above the free technical tools, a paid plan. The Free plan can run the ten technical SEO tools and nothing else.
• Enough of your monthly credit allowance left to cover the tool, if it costs 3 credits or more. See How credits work.

Permissions#

Everyone in a workspace can run tools; there is no per-tool role restriction. Access is decided by plan, not by role:

Team roles and what they change are covered in Roles and what each can do.

Navigation Path#

Dashboard → (section icon in the left rail) → (tool name in the tool list)

Three ways to get to a tool:

• Click a section icon in the left rail, then a tool in the list that appears. Clicking a rail icon opens that section's first tool immediately, so you always land on something you can use.
• Hover a rail icon to get a compact flyout listing that section's tools, and click straight through.
• Press Ctrl + K (Cmd + K on a Mac) and type the tool's name. See The command palette (Ctrl+K).

Step-by-Step Guide#

1. Open the tool. The screen shows a short description, an input row, and an illustrated explainer below. The explainer is reading material, not part of the run. See In-app tool explainers.
2. Set your global filters, if the tool uses them. The country picker in the topbar defaults to United States. On time-series tools a Range control appears next to it. Keyword Explorer adds MIN VOLUME and INTENT filter pills above the input. Filters are part of what makes a run unique, so changing one and running again is a genuinely different run. See Common inputs and validation.
3. Type your input. One value in the main field: a domain, a URL, a keyword or a topic, depending on the tool. The placeholder shows the expected shape, for example e.g. nike.com. A few tools take two values, such as Gap Finder (yourdomain.com and competitor.com) and Competitor Battle.

On a tool that takes a domain there is a second path: a Run on my website button beside the run button, carrying your connected website's name and the same cost as the run button itself. One click fills the field and starts the run, so you never have to type your own domain. It appears once a website is connected, follows the website switcher, and is absent on tools that take a keyword or a topic, where your domain would not be a sensible input.

1. Press Run, or just press Enter. Enter in the input field does the same thing as clicking the button. The button label names the action and, on most tools, the cost: Analyze Domain · 6 reports, Audit Page · FREE.
2. If you have run this before, choose. When the same tool has already been run on the same query, a dialog appears headed You already ran this, with the sub-line "Open the saved result for free, or run it again for fresh data." It shows the query, how long ago it ran, and what it cost then.
• Open saved result returns the stored copy and charges nothing.
• Run fresh (N credits) runs it again and charges the current rate.
• Cancel, Escape, or a click outside closes it without running anything. This prompt does not appear during a workflow run, because a workflow states its total cost up front.
3. Wait. A dialog opens over the tool showing Analyzing… this can take a few seconds for live data. with a spinner. Most tools finish in a few seconds. Heavy tools take longer; the long-form article writer tells you to expect 60 to 90 seconds, and the workflows quote 30 seconds to 2 minutes.
4. Read the result. The report replaces the loading state in the same dialog, headed with the tool's name. A source badge appears at the top, then a summary, then the category tabs. See Reading a result.

The field may already be filled in. When one tool sends you to another, what you were working on comes with you: run On-Page SEO on nike.com, follow a recommendation to Domain Overview, and the domain is already there. This works from any hand-off, including a recommendation's Fix this in link and a pasted #/ link.

What travels depends on what the destination asks for. A domain reaches tools that want a domain; a keyword reaches tools that want a keyword. A keyword tool is never seeded with a domain, because asking how many people search for "nike.com" would spend a credit on a question nobody asked. Anything you have already typed is left alone, and clearing a filled field keeps it clear.

Every tool reports this way, so a result looks and behaves the same wherever you ran it. Close the dialog with the X or Escape. A click outside it does nothing, and that is deliberate: what is on screen is a finished report you spent credits on, and a stray click beside the card used to throw it away mid-read. Either way the report is not lost: a View results button appears where the report would have been, and reopens it. Moving to another tool closes the dialog too, and the button is waiting when you come back.

1. Do something with it. Every result carries an export row with Export PDF, Export Excel, Export CSV, Export JSON and Share public link, and a recommendation panel underneath. See Exporting results and Get Recommendations.

Your run is saved automatically. It appears in the Library, in Saved Work, and it is what the "You already ran this" prompt offers you next time. Saved results are kept for 90 days. See Saved Work.

What a run costs#

There is no confirmation step before a paid run. Clicking Run starts the tool and the credit is charged. The two things that protect you from an accidental charge are the credit figure printed on the button and the "You already ran this" prompt.

Costs are per tool and range from 0 to 8 credits. The full table is in Tool catalog and credit cost. Credits are deducted when the run succeeds, and reopening a saved result never costs anything.

Troubleshooting#

FAQs#

Do I get charged if the tool fails? No. Credits are recorded when a run produces a result. A failed request is logged as an error, not as usage.

Does pressing Enter cost more than clicking the button? No, they are the same action. Enter in the input field runs the tool once.

Can I run two tools at the same time? Yes. Each screen keeps its own result, and switching sections does not cancel a run in progress. The workflows already run several tools in parallel for you.

What happens if I close the tab mid-run? The request continues on the server and the result is still saved, so you can usually find it in the Library. The long-form article writer explicitly asks you not to close the tab, because its progress is polled from the page.

Why does the same tool sometimes ask "You already ran this" and sometimes not? The check matches the tool, the query and your filters. Change the country, the range or a filter and it is a new run, so no prompt appears. The prompt is also skipped inside workflows and inside the recommendation panel.

Can an administrator turn a tool off? Yes. A tool can be disabled at the platform level, in which case it returns This tool is temporarily unavailable. Please try again shortly. for everyone until it is switched back on.