By Mehdi HassanWhat is dashboard copywriting?
Dashboard copywriting is the writing on a dashboard: the headline, chart titles, metric labels, axis units, tooltips, empty states, errors and alerts. It's the part of dashboard design that turns numbers into meaning. A perfectly drawn chart with a vague title still leaves the reader asking, "So is this good?"
Almost no dashboard guide covers this, which is odd, because most dashboard confusion is a words problem. In our dashboard design guide we argue for writing the headline insight before designing anything. This article is the practical follow-up: what to write, where, and how to keep it accurate as the product grows.
Copy element | Its job | Weak | Strong |
|---|---|---|---|
Headline | Give the conclusion | Overview | Signups are up this week, led by organic search |
Chart title | Say what the chart shows | Chart 1 | Weekly signups, last 12 weeks |
Metric label | Name the number | conv_rate_v2 | Trial to paid conversion |
Period and unit | Frame the number | (none) | This month · USD |
Tooltip | Define the metric | (none) | Paid accounts divided by trials started in the period |
Empty state | Explain and guide | No data | Connect your store to see sales here |
How should you write the text on a dashboard, step by step?
Write the text on a dashboard in the same order a reader meets it: headline first, then card labels, chart titles, definitions and finally states. Doing it in that order before visual design means each later piece supports the earlier ones, and you find out early if a card has nothing to say.
Write the user's question
One sentence in their words, such as "Are my campaigns bringing in customers?" This is the brief for all other copy.
Draft the headline pattern
Metric, direction, comparison and reason where you have it: "Signups are up on last week, led by organic search."
Name every metric
List each number with a plain label, its period, its unit and whether higher is better.
Title every chart
A descriptive title for every chart, and an insight subtitle where the product can generate one.
Write definitions
A one-line tooltip for every rate, score or ratio, matching how it's actually calculated.
Write the states
Empty, loading, zero-results, error and no-permission copy for every card and table.
How do you write chart titles that explain the numbers?
Write chart titles that say exactly what is plotted, including the measure, the breakdown and the period, and add an insight line when you can state the finding. Descriptive titles keep a chart accurate as the data changes; insight titles save the reader from working out the conclusion themselves.
- Descriptive title: "Monthly recurring revenue, last 12 months." Always correct, even when the trend changes.
- Insight subtitle: "Up for the third month in a row." Only works if the product generates it from the data or a person writes it for a report.
- Avoid labels as titles: "MRR" alone doesn't say the period or the breakdown.
- Avoid clever titles: "Money, money, money" is memorable once and confusing forever.
- Put units and periods in the subtitle if the title gets long: "USD · Monthly · Excludes refunds."
In static dashboards, descriptive titles are the safe default, because an insight written at design time will be wrong next month. In reports and in products that generate summaries from the data, insight titles make the dashboard feel like it's talking to you. Choosing the right chart to support the title is covered in how to choose a chart type.
How should you write dashboard labels and units?
Write dashboard labels in the words your users already use, keep them short, and make each label mean exactly one thing everywhere in the product. A label is a promise about how the number is calculated. If two screens use different names for the same metric, or the same name for different metrics, people stop trusting both.
- Use the customer's vocabulary. "Orders" for a store owner, "Bookings" for a clinic, even if the database calls both "transactions".
- Prefer nouns to jargon. "Customers who left" may beat "Churned accounts" for non-specialist audiences.
- State the period. "This month", "Last 7 days", "Since you joined". If one control sets it for the page, make it visible near the top.
- State the unit once. Currency and time units in the subtitle or column header, not repeated on every value.
- Say when data is fresh. "Updated 5 minutes ago" stops people acting on stale numbers.
- Keep a glossary. One shared list of metric names and definitions used by product, design, support and docs.
Labels on KPI cards carry extra weight because they're read first. Our guide to KPI card design covers labels together with comparisons and deltas, and our guide to UX writing for SaaS covers product-wide microcopy.
What makes good microcopy for dashboard tooltips and definitions?
Good dashboard microcopy in tooltips defines the metric in one sentence, says how it's calculated in plain words and mentions anything that's excluded. It's the answer to the question a careful reader would otherwise ask support. Keep it short, because tooltips are read in a moment, often while hovering over the number.
Hypothetical example: rewriting tooltips for a churn card
For example, suppose a subscription product's dashboard has a card labelled "Churn" with the tooltip "Churn rate."
A useful rewrite: label "Customer churn, this month"; tooltip "Share of customers at the start of the month who cancelled during it. Excludes customers who paused." That one sentence answers which customers count, which period and what's left out.
If the product also reports revenue churn, the label must say so: "Revenue churn, this month," with its own definition. Two numbers called "Churn" on the same dashboard are a reliable source of arguments in leadership meetings.
Definitions also need to stay true. When the calculation changes, the tooltip has to change in the same release. That's easiest when the definition lives next to the metric in code or in the shared glossary, rather than being typed into a design file once and forgotten.
How do you write empty states, errors and alerts on a dashboard?
Write empty states, errors and alerts so they say what happened, why if it's useful, and what to do next, in that order. These are the moments when a dashboard is most likely to confuse people, and where a short, honest sentence does more than any visual design.
Situation | Weak copy | Better copy |
|---|---|---|
New account | No data | Your sales will appear here once your store is connected. Connect store |
Filters too narrow | No results | No orders match these filters. Clear filters |
Query failed | Error 500 | We couldn't load this chart. Try again |
Partial data | (nothing shown) | Data for 2 of 3 accounts. One account needs reconnecting |
Threshold alert | Warning | Usage is close to your plan limit. View usage |
The same principles apply to form errors and reassurance text across your product. Our guide to microcopy that converts covers those patterns, and accessible forms covers how to make error messages work for screen reader users.
How should AI-generated insights be written on a dashboard?
AI-generated insights on a dashboard should follow the same rules as a human-written headline: specific, tied to a visible number and honest about uncertainty. They should also show where the claim comes from, so the reader can check it against the chart, and never state a cause the data doesn't support.
- Anchor to a number: "Signups fell compared to last week" should sit next to the signups chart that shows it.
- Separate fact from guess: "Signups fell" is a fact; "likely because of the pricing change" is a hypothesis and should read like one.
- Link the source: let people open the filtered data behind the sentence.
- Allow dismissal and feedback: people should be able to hide an insight that isn't useful.
These patterns overlap with how AI features earn trust in general. Our guide to AI product UX covers confidence, sources and controls, and our guide to AI chat interface design covers writing for conversational interfaces.
