How to Calculate the ROI of a Business Website

Use profit and change—not visits and hope

A website can create value through new customers, better-qualified inquiries, larger or faster purchases, lower support effort, fewer scheduling errors, faster publishing, and reduced reliance on an old system. It can also influence customers who later call, visit, or buy through another channel. Build the model before choosing a website design scope so the team knows which actions and operational changes matter. A polished design without a baseline or measurement plan leaves the business guessing after launch.

Choose the measurement period and baseline

State whether you are evaluating the first 12 months, three years, or another period. Put every benefit and cost inside the same dates. Then choose a baseline: the old site's recent performance, a matched seasonal period, or a documented manual workflow. Do not compare a slow January before launch with a busy June after launch and credit the website for the season. Note other changes such as pricing, ad spend, staff, territory, inventory, promotions, and sales follow-up. A useful model is transparent about what it cannot isolate.

Collect the seven inputs

Google Analytics calls an action important to the business a key event; any collected event can be marked as one. That makes form completion, booking completion, or another meaningful action reportable, but it does not make the event a sale. Connect analytics to the lead or order record. If the business cannot match a person-level path for privacy or technical reasons, use aggregated rates and label the limitation. The setup steps in How to track website conversions help define the measurement layer.

A worked ROI example

The example is arithmetic, not a performance promise. Its weakest assumption may be that the new site adds six qualified opportunities every month. Test that assumption at three levels. If the conservative case produces only 0.5 extra customers per month, customer gross profit becomes $9,000; adding $1,680 in savings would not recover the $22,000 first-year cost. That does not prove the site is wrong—it may have a longer useful life or protect important risk—but it changes the investment conversation.

Use three scenarios, not one forecast

The spread is the insight. It tells the business which input deserves validation before signing. Interview sales about lead quality, inspect the old form and call records, and calculate real close rates by source. If the decision depends on the high scenario, reduce the initial scope or gather better evidence. SBA guidance on break-even analysis similarly treats the calculation as an estimate for planning, not a perfectly accurate forecast before costs and sales occur.

Do not double-count attribution

A customer may see a referral, search the company, read the website, return through an ad, and call from a saved number. Google describes attribution as assigning credit to touchpoints along the path to an important action and provides reports for comparing paths and models. No attribution model discovers a single philosophical truth. Choose a rule for the business decision, keep it consistent, and reconcile it with sales evidence. If ads and the website both claim the entire customer value, your combined marketing report will overstate return.

Include costs people commonly omit

• Discovery, research, copy, photography, design, development, and project management
• Data cleanup, content migration, redirects, integrations, accessibility and quality testing
• Domain, hosting or platform plans, apps, licenses, monitoring, maintenance, and support
• Internal interviews, content review, approvals, training, sales-data cleanup, and reporting
• Advertising or promotion required to create the modeled traffic, when it is not already in the baseline
• Later improvements promised by the business case during the selected period
• Financing, tax, and currency effects when material and confirmed by the appropriate adviser

If the website will be useful for three years, also calculate a three-year view. Include later support and improvements, and avoid simply multiplying first-year gains when demand or capacity will change. Discounted-cash-flow analysis may be appropriate for a large investment, but most small-business decisions improve immediately when assumptions, gross profit, complete cost, and time period are written clearly. More planning guides are organized in the website pricing and budgeting hub.

Create a monthly decision report

1. Record qualified visits and the pages or campaigns that attracted them
2. Count tested key events, then remove spam, duplicates, and internal activity where practical
3. Have sales classify inquiries with consistent accepted and rejected reasons
4. Connect opportunities to closed customers and collected value on a suitable delay
5. Update close rate, customer value, margin, and operating-savings assumptions
6. Compare actuals with low, expected, and high cases
7. Choose one action: fix measurement, improve a weak stage, increase a working source, or hold steady

Do not force a decision every week when sales cycles are long. Google notes that attribution-path reports can show time to a key event and the touchpoints involved. Match the reporting window to the customer's real path and maintain enough detail to investigate changes. The point of ROI measurement is not to decorate a dashboard; it is to decide whether to keep, change, expand, or stop an investment.