Guide6 min readUpdated
Website Analytics for Design: What to Measure and Fix
Website analytics for design decisions: learn which pages people leave, how to find enquiry form drop off and what a small business should measure.
A photographer's website in Kolkata gets steady visits, but only a few enquiries each month. The owner thinks the portfolio needs more photos. Then she looks at her visit data and sees that most people reach the Contact page and leave without sending the form. The portfolio was fine. The form was the problem.
Website analytics for design is the habit of reading simple visit data, such as pages, clicks and exits, to decide which part of the design to improve first. It does not need a statistics degree. It needs a few questions and a weekly routine. This guide uses concepts only, so it works with any analytics tool you pick. Check the current options of any tool before you choose one.
What should a small business measure on its website?
A small business should measure five things: device type, landing pages, exit pages, button clicks and form completion. These five answer most design questions.
| What to measure | The question it answers | Design area it points to |
|---|---|---|
| Device type | Do most visitors use a phone? | Mobile layout, tap size, speed |
| Landing pages | Where do people first arrive? | The pages that need the strongest first screen |
| Exit pages | Where do people leave? | Pages with a missing next step |
| Button clicks | Do people press Call or WhatsApp? | Placement and wording of buttons |
| Form completion | How many start and finish the form? | Form length, fields, errors |
Most Indian visitors use a mobile phone, so check your own device split and look at your site on a mid-range phone first. For this approach, see mobile-first website design.
Do not track everything. A long report with 30 numbers hides the three that matter. Pick the five above, and add more only when you have a question that needs them.
Which pages do people leave, and is that a problem?
Leaving is only a problem on a page where the visitor should have continued. People always leave from some page, so an exit is normal on the thank-you page and a warning on the pricing page.
Two terms help. An exit is the last page a person sees before leaving your site. A bounce is a visit where the person saw only one page and left. A blog post with a high bounce can be fine: the reader found the answer and went away. A homepage with a high bounce is more worrying.
Ask three questions about each page that people leave a lot:
- Is this the end of a good journey? A thank-you page or a confirmed booking is a good end.
- Is there a clear next step? If the page has no button, link or contact option, the visitor has nowhere to go.
- Did the page match the promise? A visitor who clicked an advertisement for "wedding photography" and landed on a general page may leave at once.
Use the page checklists in homepage design checklist and call-to-action button design to check the next step.
How do you find enquiry form drop off?
Enquiry form drop off is the gap between people who reach your form and people who send it. To see it, count three steps: people who open the page, people who start typing and people who press submit.
Imagine a coaching institute with an enquiry form of 9 fields. Many people open the page. Fewer start typing. Very few submit. The gap between "open" and "start" suggests the page does not make people want to continue. The gap between "start" and "submit" suggests the form itself is the problem.
Common causes and first fixes:
- Too many fields. Remove everything you do not need for the first reply. Name, phone and one question is often enough.
- A required field that people avoid. Make email optional if you mostly reply on WhatsApp.
- Poor error messages. If a wrong phone number clears the whole form, people give up. Keep what they typed and explain the error.
- No sign of what happens next. Add one line: "We reply on WhatsApp within a day."
- A form that is hard on a phone. Use the number keypad for phone fields and large tap areas.
See form design best practices for the field-by-field details, and website contact page and WhatsApp CTA for choosing between a form and a chat button. Some businesses find that a WhatsApp button gets more enquiries than a form. Your own data will tell you.
How do you turn a number into a design decision?
Turn a number into a decision by writing three things: what you saw, the most likely cause and one small change to try. Then change one thing at a time.
| What you see | Possible cause | First change to try |
|---|---|---|
| Many mobile visitors leave the homepage | Slow load, small text, a heavy first screen | Reduce image size, enlarge the headline and button |
| Visitors reach the pricing page and leave | Prices unclear, no next step | Add a clear "Get a quote" button and what is included |
| Few clicks on the WhatsApp button | Button hidden or wording vague | Move it up, use "Chat on WhatsApp" |
| Many open the form, few submit | Too many fields or errors | Cut fields, improve error messages |
| People land on an old page | Search results or links point there | Update that page or redirect it |
A page speed problem is also a design problem, as explained in how design affects website speed.
When you have a change ready, wait a few weeks and look again. If you want to compare two versions properly, read what is A/B testing in design. With low traffic, a test is unreliable, so a before and after comparison is more honest.
What is a simple weekly analytics routine?
A simple routine takes 20 minutes once a week. Do the same steps each time so that you notice changes.
- Open the last 7 days of data.
- Check the device split.
- List the top 5 landing pages.
- List the pages people leave most, and mark the ones that should not be an end.
- Note how many people clicked Call or WhatsApp, and how many sent the form.
- Write one sentence: "This week I will fix ___ because ___."
Keep a note of each change and its date, such as "10 Nov: shortened the form to 3 fields". Without a date, you cannot link a result to a cause.
Respect your visitors. Tell them in a privacy note what you collect, and collect only what you need. This is general guidance, not legal advice.
What can analytics not tell you?
Analytics cannot tell you why people behave as they do, and it cannot show people who never arrive. A drop at the pricing page may mean the price is high, the layout is confusing or the person was only browsing.
To find the why, ask. Call three recent visitors or customers and ask what they were looking for. Watch two people use the site on their phones, as in usability testing on a small budget. Use a UX audit checklist to inspect the weak pages.
Small numbers also mislead. If a page gets very few visits, one person can change the pattern. Treat small numbers as hints, not proof.
If you would like help reading your numbers and turning them into a clear redesign plan, see my website design work or get in touch.
Frequently asked questions
Which website numbers matter most for a small business?
+
Start with device type, landing pages, exit pages, clicks on your call or WhatsApp button and enquiry form completion. These five show where visitors get stuck.
What is a good bounce rate?
+
There is no single good number. A blog page where readers find the answer and leave is fine, while a homepage where people leave at once needs a closer look. Compare each page with its own purpose and with its own past results.
How often should I check my website analytics?
+
Once a week for 20 minutes is enough for most small businesses. Write down what you changed and on what date, so you can see what worked.
Can analytics tell me why visitors leave?
+
No. Analytics shows where people leave, not why. Pair it with short conversations and watching two or three people use your site on a phone.
Working on something like this?
Tell Tithi what you are planning. You get a clear scope, a fixed price and a timeline before any work begins. I reply to every enquiry within 24 hours.