Prompt tracking for brands
Prompt tracking is not a pile of screenshots. It is a frozen list, a schedule and a history you can read next month.
Built by StyleLine FZE in Sharjah. Email support@mentionclock.com. We reply within one business day.
Choosing a tool is a different job from a single method. The dated vendor scorecard is best ChatGPT brand monitoring tools.
Prompt tracking for brands is rank tracking for chat. Without history you cannot tell if last week's article did anything.
Build the list
Start from sales calls, "best X" queries in Search Console, and the prompt ideas generator. Cap the list at a number you will actually read. Twenty good prompts beat two hundred you ignore.
Tag them
Awareness ("what is X"), consideration ("best X for Y"), competitor ("X vs Z"). Mention rate will differ by tag. That tells you whether you have a famous name and a weak comparison story, or the reverse.
Review
Once a week, open the runs that dropped. Read the cited pages. Assign one URL to improve. Leave the rest of the list alone.
Mention Clock stores the list, the engines and the history. Free covers 5 prompts. Starter at $29 a month covers 25.
A worked week on a real project
Prompt tracking for a brand is a frozen list, a schedule, and a history. Mention Clock stores that list on a project and runs it through the engines on the plan. Free is $0: 1 project, 5 prompts, 2 engines, weekly, no card. Starter is $29 each month ($290 billed yearly): 25 prompts, 3 engines (ChatGPT, the second chat engine, and Perplexity), 3 rivals, Slack or Teams. Pro is $79 a month or $790 a year: 100 prompts, every 3 days, 10 rivals, and the cited-source report. Agency is $199 a month or $1,990 a year: 15 projects, 400 prompts, client workspaces, and a white-label PDF. Yearly prices are 10 months, which is two months free. The upcoming engine is being added. These figures are the published pricing card. Nothing new is added here.
Here is a calculation, not a case study. A brand on Free picks 5 prompts that cover at least three shapes: one "best of" question, one problem question, one alternatives question, and two that name a constraint such as team size or budget. The project runs weekly on 2 engines. After four weekly runs you have 5 prompts times 2 engines times 4 weeks, which is 40 answer rows. That is enough to see whether a name appears at all. It is not enough to split awareness from comparison. When the same brand moves to Starter, the cap becomes 25 prompts and 3 engines. A 6, 6, 5, 4, 4 mix of category, problem, alternatives, comparisons, and specific needs fits that cap. Leave the wording alone for those four weeks. A renamed prompt is a new series, not a continuation.
Open the runs that dropped. Read the cited pages. Assign one URL to improve and leave the rest of the list alone. The top impression page in the 26 Oct to 1 Nov 2026 window for this product is ChatGPT brand monitoring. Use it when the question is specifically ChatGPT. Use the switch form when you already have a prompt export from another tool and want it on a Mention Clock project without retyping.
Tag the list as you add it. Awareness is "what is X". Consideration is "best X for Y". Competitor is "X vs Z". Mention rate will differ by tag. A famous name with a weak comparison story shows up as strong awareness and weak competitor tags. The reverse is a product people find only when a rival is already in the question. That split is the point of tagging. A single blended percentage hides it.
Who it is for: a brand owner or an in-house marketer who can name the buyer and the rival. It is not for an intern collecting screenshots. Screenshots have no schedule and no cited-page history. It is not a promise that a rewrite will move a model. Search-backed answers can reflect a new page within weeks. Answers that do not search wait on a model update. Judge a month of weekly runs before you call a page a win or a loss.
Review the list each quarter or when you launch a product line. Skip branded-only lists that only ask for your name. Skip ten near-duplicates that differ by one adjective. If people use both a product name and a company name, track both as variants in the same project so share of mentions stays honest. Keep another language in another project. Mixing English and German in one share number hides both.
The Monday recap and the drop-out mail are on every plan, including Free. Slack or Teams posts start on Starter. The cited-source report starts on Pro. If you only need to know whether five buyer questions name you, stay on Free until the list is obviously too small to read.
What you leave alone
After the fourth weekly run, resist editing the winners. Assistants are sensitive to small wording changes. "Best CRM for a 10 person sales team" and "best CRM for small sales teams" can name different products. Treat the prompt like a query you would not rename every week. When you must edit, note the date in the project so a drop is not mistaken for lost visibility. The Monday recap is the read. The drop-out mail is the exception list. Everything else waits until the quarter review.
Share of mentions needs the rival set you actually lose deals to. Free includes 0 competitors, so share is not on that plan. Starter includes 3. Pro and Agency include 10. Adding a rival you never meet inflates the denominator. Removing the one you do meet makes the chart look calm. Write the three names down before you upgrade, then leave them for the same four weeks as the prompts.
The cut from a long brainstorm down to the questions you will rerun is how to pick the 20 prompts that matter.
FAQ
Should product names and the company name both be tracked?
Yes, as variants, if people use both. Keep them in the same project so share of mentions stays honest.
What about other languages?
Separate projects. Mixing English and German in one share number hides both.