How much to charge for a dashboard
How much to charge for a dashboard comes down to one question: what decisions will the client make with it, and what does getting the numbers cost them today? Nobody buys a dashboard for the charts. They buy it so the owner stops waiting until the end of the month to see which channel is losing money, so a manager stops spending Monday morning pasting exports into a spreadsheet, or so the team argues about what to do instead of whose numbers are right. Price the build on a share of that value, add lines for each data source and for the work of cleaning and defining the data, and quote a monthly plan for upkeep. Do not price it on the number of charts or on hours.
Dashboards are good work to sell with Claude Code because the first version comes together fast. Point Claude at the data, describe the views, and you can show something real early. That is also the trap. A dashboard looks finished the moment the charts render, so builders price it like a quick job, and then the client finds a number that does not match their accounting software and the real work starts, unpaid.
Why the charts are the wrong yardstick
A dashboard with three charts can be worth far more than one with thirty. If those three numbers tell an owner where to spend next month's budget, they matter. Thirty charts nobody opens matter to nobody. Charging per chart rewards you for clutter and punishes you for doing the hard thinking about what actually needs to be on the screen.
Hours are worse. Claude Code makes the build itself quick, so pricing on hours turns every bit of speed into a pay cut. I covered why in why I stopped charging hourly for AI builds. The real costs of a dashboard sit around the charts: getting access to the data, fixing it, agreeing what each number means, and keeping it right when a source changes.
The CCC Dashboard Quote
Five parts of a dashboard quote
| Part | What it prices | Question to ask the client |
|---|---|---|
| Base build | The value of the decisions and the reporting time it replaces | Who builds these numbers today, how long does it take, and what do you decide with them? |
| Data sources | Each system the data comes from | Where does each number live today, and can we get access to it? |
| Cleaning and definitions | Fixing messy data and agreeing what each number means | When two reports disagree, which one do you trust, and why? |
| Access and sharing | Logins, who sees what, and scheduled reports | Who needs to see this, and should everyone see everything? |
| Keep plan | Monthly checks, fixes and small changes | What happens if a number on this is wrong for a week and nobody notices? |
Base build. Work out with the client what the reporting costs today: how many hours someone spends pulling exports and building the report, multiplied by what their time costs. Then ask what they would decide differently if they saw the numbers daily instead of monthly. Price the build at a fraction of one year of that value, so the client can see it pay for itself.
Data sources. Each place the data comes from is its own line: the payment system, the CRM, the ad accounts, a shared spreadsheet. Each one has its own access, its own limits and its own way of changing without warning. Ask for access before you send the final number. If a source needs a paid API, the key lives on the server, never in the page, with a spending limit, and that setup is part of the quote. How to connect Claude Code to a database shows what one connection involves.
Cleaning and definitions. This is the line builders forget and clients never expect. The same customer appears twice under different emails. Refunds are counted in one system and not the other. Two managers mean different things by "active customer". Agreeing those definitions in writing and making the data match them is often the biggest part of the job, so price it on what you find, not on a guess.
Ask these before you quote
- Which three numbers would you check first every morning? If they cannot name them, the first paid step is working that out with them, not building.
- How fresh do the numbers need to be? Daily, hourly and live are different jobs with different costs. Most owners need daily.
- Who sees it? One owner is simple. A team where each person should only see their own region or clients adds an access line.
- Where does it live? A hosted page with a login, a view inside a tool they already use, or a scheduled email report. Settle it before you quote.
- Who owns the data accounts? Access should run through accounts in the client's name, so the dashboard keeps working if you part ways. How to hand off a Claude Code project covers the checklist.
More questions that save a bad quote are in client discovery questions before you quote.
Always quote the Keep plan
Dashboards break quietly. A software tool renames a field, an ad platform changes what it reports, the client adds a new sales channel, or a login expires. The page still loads, the charts still render, and the numbers are wrong. That is worse than a dashboard that is down, because people make decisions on it.
Quote a monthly plan in the original proposal. Say what it covers: checking the numbers against the sources, fixing breaks when a source changes, keeping access current, and small changes to existing views. Say what it does not cover: new data sources and new dashboards, which are quoted separately. How to price a maintenance retainer covers the structure. A dashboard the owner checks every morning is one of the easiest retainers to justify, because they notice the first day it is wrong.
How to present the dashboard quote
- Lead with the decision in their words. One line on what they will be able to see and decide, then the price.
- Itemize it. Base build, each data source, cleaning and definitions, access and sharing, then the Keep plan. Itemized quotes get trimmed, not rejected.
- Gate payments on proof. A deposit to start, a payment when the numbers match their own records for an agreed period, and the final payment once the team is using it. How to structure milestone payments for an AI project covers the schedule.
- Put the definitions in the proposal. A short list of what each number means, signed off by the client, ends the "this number is wrong" argument before it starts.
If you have not built one yet, how to build a dashboard with Claude Code walks through the build so you know what you are quoting.
Short, practical drops on skills, MCP, agents, prompts, and more. No spam, unsubscribe anytime.
Frequently asked questions
How much should I charge for a custom dashboard?
Price it on what it changes for the business: the reporting hours it replaces and the better decisions that come from seeing the numbers sooner. Charge a fraction of one year of that value for the build, add a line for each data source, a line for cleaning and defining the data, and quote a monthly plan for upkeep.
Should I charge per chart for a dashboard?
No. A few charts that drive real decisions are worth more than dozens nobody opens. Per-chart pricing rewards clutter and ignores the data access, cleaning and upkeep that make up most of the work.
Should I charge a monthly fee for a dashboard?
Yes. Data sources change, fields get renamed, logins expire and the client adds new channels. A monthly plan covers checking the numbers, fixing breaks and small changes, with new data sources and dashboards quoted separately.
What is the difference between a dashboard and an internal tool?
A dashboard is mostly for reading numbers and making decisions. An internal tool is where the team does work: entering, approving or editing data. If the client's team needs to act inside it, price it as an internal tool.
What if the client's data is a mess?
That is common, and it is part of the job. Look at the data before you send the final number, price the cleaning and definitions as their own line, and get the client to sign off on what each number means.
Last reviewed by Duncan Rogoff on October 10, 2026
