How hard is it to book a doctor online? A test of 56 Indian hospital websites
Data collected on by Gaurav Phogat. Method, coding rules and dataset are published below.
I clicked “Book appointment” on 56 hospital websites across seven Indian cities and followed each path until the site either showed me a real time slot or asked for my personal details. Only 10 of the 56 showed a bookable slot first. Almost half (27) ended in a request form where you leave your number and wait for a call. Eight asked for a mobile OTP or a portal login before showing any time at all, and 10 had no working online path on the day I tested.
New to the topic? Start here: Hospital growth strategy sets out the reading order.
Last updated: 29 October 2026.
Every hospital website I have worked on has a “Book appointment” button somewhere near the top. It is the most valuable button on the site, and it is also the one that gets the least scrutiny after launch. Marketing teams track how many people click it. Far fewer teams check what happens after the click.
So I tested that part of online doctor appointment booking. Not the design, not the speed, just the path: how many steps it takes to see a time, what the site asks for, and whether the visitor ends up with a slot or a promise of a callback. I stopped before entering any personal detail on every site, so nothing here involved a real booking.
Key findings
- 18% showed a real slot first. 10 of 56 hospital websites displayed bookable time slots before asking for any personal detail.
- 48% ended in a request form. On 27 of 56 sites, “Book appointment” led to a form that collects your details for a callback. No slot, no confirmation on screen.
- 14% put an OTP or login first. 8 of 56 asked for a mobile OTP or portal sign-in before showing any time slot.
- 18% had no working online path. 10 sites offered only a phone number, WhatsApp or a callback, showed an “online booking unavailable” notice, or returned a server error.
- Three steps to a slot. On the 10 sites with live slots, the median was 3 steps from the homepage to seeing times (range 2 to 5).
- Fees were rarely visible. Only 5 of 56 showed the consultation fee before asking who you are.
- On a phone, 42% hid the booking button. On 23 of the 55 homepages I could measure on a phone-sized screen, no booking-labelled button was visible in the first screen.
How I did this
Sample. I searched Google India in a logged-in desktop browser for “hospital in [city]” and “best hospital in [city]” and read the first two pages of organic results. I took the first eight hospital-owned websites per city in order of appearance, skipping directories, aggregators, news sites, insurers, government department portals and any site already counted in an earlier city. The plan was ten cities. Google started showing a bot check after the seventh city, so the sample covers seven: Delhi NCR, Mumbai, Bengaluru, Chennai, Hyderabad, Kolkata and Pune. That gives n = 56 websites.
Hospital types. I labelled each site as national chain (8), regional chain (12), standalone private (29), charitable or trust hospital (6) or government (1). The labels are my own reading of each site’s footprint, so treat them as approximate. No hospital, doctor or company is named in this article or the dataset.
Path rules. From each homepage I clicked the most prominent booking button (header first, then hero). If that was only a phone number, I used the next booking link. I then took the shortest obvious route: a general or internal medicine doctor where offered, the first available doctor otherwise, and the earliest date with availability. I counted each click, dropdown choice or typed search as one step. I stopped at the first screen that either showed time slots or asked for personal details, and I never entered a name, number, email or OTP.
Mobile checks. I loaded each homepage at iPhone 13 screen size and recorded whether a booking-labelled button or link sat in the first screen, whether any booking button was fixed or sticky, whether a tap-to-call link was visible, whether a WhatsApp link existed, and whether the page scrolled sideways. One site blocked the automated mobile check, so mobile figures use n = 55.
Limits. This is a one-day snapshot from 29 October 2026, desktop for the booking path and phone emulation for the homepage checks. Slots depend on the doctor and date I picked. A form-based site may still call back quickly, and I did not test what happens after submission, so I cannot say anything about response times, OTP reliability or whether payment is taken at the end. Search results are personalised, so another person running the same queries would see a slightly different list.
What “Book appointment” actually leads to
The button says the same thing on almost every site. What sits behind it falls into six very different experiences.
n = 56 hospital websites, 29 October 2026. The eighth OTP-first site is counted under “no working online path” because its portal also said online booking was disabled.
The request form is the default in India, and I understand why. It is cheap to build, it feeds the call centre, and it lets the hospital steer a patient to the right doctor. But from the patient’s side it is not a booking. Ten of the 27 forms even ask for a preferred date, which makes the visitor feel they have booked something when they have only asked. One form’s phone placeholder used a US number format, a small sign that nobody on the team had tried it as a patient.
Live slots exist, and they are not hard to reach
The 10 sites that showed real slots were not all big chains. Six were standalone private hospitals, two were regional chains, one a trust hospital and one a national chain. Two ran the same off-the-shelf hospital information system booking screen, with dropdowns for hospital, specialty and doctor followed by a grid of 10 or 15 minute slots. Others used a hosted patient portal on a subdomain.
The median path was three steps: the booking button, a doctor, then a date. The fastest took two. The slowest took five, because the first date I picked had no slots and I had to use a “find next available date” button. Four of these 10 sites also showed the consultation fee next to the slots, which is the kind of plain information that saves a phone call.
Two other sites had slot systems but lost me before I saw a time. One regional chain site had a proper slot calendar, but every date I tried for the doctor its finder suggested showed no slots and fell back to a callback form. One national chain site defaulted the location to a city I was not in, and most booking buttons stayed greyed out until I changed it.
OTP walls before the patient sees a time
Eight sites asked for a mobile number and OTP, or a portal login, before showing any slot. Three of the eight national chains did this. I see the logic: it ties the booking to a verified number and cuts no-shows. The cost is that the patient has to commit before they know whether their doctor is free this week. One flow then listed four steps ending in “Confirm & Pay”, and another asked for a mobile OTP followed by title, names, gender, date of birth, email, country, state, city and pincode before reaching the doctor list.
Compare that with the sites that showed slots first and asked for a mobile number only after the visitor picked a time. Same verification, but the patient sees value before the ask.
How much a request form asks for
n = 27 request forms. Fields counted as every visible input, dropdown or date picker on the booking step.
The median request form had five fields: usually name, phone, email, a department or doctor, and a message. That is reasonable. The two longest had 12, including address, country, state and city, a report upload and a quiz-style anti-spam question. Two forms opened with a marketing consent box already ticked. Under the DPDP Act, consent has to be a clear affirmative action, so a pre-ticked box is worth fixing whatever the form’s length.
When the button goes nowhere
Ten sites had no working online path on the day. Six offered only a phone number, WhatsApp or a callback, and on one of those every “Book appointment” button was simply a phone link. Three showed a notice that online booking was temporarily unavailable or coming soon. One booking link returned a server error page. Eight of these ten were standalone private hospitals, which is 8 of the 29 standalones in the sample.
| Hospital type | Sites | Live slots first | Request form | OTP or login first | No working path | Other |
|---|---|---|---|---|---|---|
| Standalone private | 29 | 6 | 15 | 0 | 8 | 0 |
| Regional chain | 12 | 2 | 7 | 2 | 0 | 1 |
| National chain | 8 | 1 | 2 | 3 | 1 | 1 |
| Charitable or trust | 6 | 1 | 2 | 2 | 1 | 0 |
| Government | 1 | 0 | 1 | 0 | 0 | 0 |
A phone-only path is not wrong for every hospital. Many Indian patients prefer to call. The problem is a button labelled “Book appointment” that silently dials a number, or a booking page that has said “coming soon” long enough to become part of the site.
On a phone, the button is often below the fold
n = 55 homepages at iPhone 13 size, 29 October 2026. One site blocked the automated mobile check.
The basics were fine. All 55 homepages had a proper mobile viewport tag, and none scrolled sideways. But on 23 of the 55 there was no booking-labelled button in the first phone screen, and only 24 kept a booking button fixed while scrolling. Tap-to-call was far more common than booking, which tells you where most hospitals still expect the conversation to happen.
What this means for hospital marketers
None of this needs a new website. It needs someone to walk the path as a patient once a month and fix what they find. Here is what I would do, in order.
- Name the button for what it does. If it opens a callback form, call it “Request a callback” or “Request an appointment”. If it dials, label it with the number. Save “Book appointment” for flows that end in a slot.
- Show times before you ask for a number. If your HIS or portal has slots, let visitors see them first and verify the mobile number after they pick a time.
- Show the consultation fee where the slot is. It is factual information, and it removes the most common reason to call. Keep fee text plain: no discounts, offers or “best price” claims, which sit badly with NMC and ASCI rules.
- Cut the form to five fields or fewer. Name, mobile, department and preferred date are usually enough for a callback. Untick consent boxes by default and link your privacy notice.
- Remove dead ends. Search your own site for “coming soon”, “temporarily unavailable” and broken booking links. Default location from the visitor’s choice, not a guess.
- Fix the first phone screen. Put one booking action and one call action in the first screen and keep them sticky. Test on a mid-range Android phone on mobile data, not on the office Wi-Fi.
- Measure the path, not just the click. Track button clicks, slot views, form starts and submissions as separate events so you can see where patients drop.
If you want a wider view of how Indian hospital websites perform beyond booking, my Indian hospital website benchmark covers tracking and privacy, Core Web Vitals, accessibility and structured data. For what happens after a form is submitted, see CRM lead ownership and routing for multi-hospital groups. Before you change consent wording, read the DPDP Act for hospital marketers, and for fee and offer language, hospital advertising rules under NMC and ASCI.
Download the data
The full dataset (CSV) has one row per website: city, hospital type, booking outcome and flow, steps to first time information, whether exact slots appeared before personal details, the step where details were asked, OTP or login gates, form field counts, fee visibility, the mobile checks and a short note on each path. It contains no hospital or person names. Collection date is in every row. It is free to reuse under CC BY 4.0 with credit to Gaurav Phogat and a link to this page.
How to cite this study
Phogat, G. (2026). How hard is it to book a doctor online? A test of 56 Indian hospital websites. gauravphogat.com. Data collected 29 October 2026. https://gauravphogat.com/hospital-opd-booking-flow-study/
Questions people ask
Sources and further reading
- CC BY 4.0 (creativecommons.org)
In my test of 56 hospital websites across seven cities on 29 October 2026, 10 (18%) showed real bookable time slots before asking for personal details. Another 8 asked for a mobile OTP or portal login before showing any slot.
Most Indian hospital sites use request forms that feed a call centre, so the phone number is the booking. In this study 27 of 56 sites ended in such a form, with a median of five fields.
On the 10 sites that showed live slots, the median was three steps from the homepage: the booking button, a doctor and a date. The range was two to five steps.
Showing the fee as plain factual information is generally fine and saves patients a call. Only 5 of 56 sites in this study did so before asking for details. Avoid discounts or superlative claims, and check NMC and ASCI rules with your compliance team.
Label the button for what it does, show slots before asking for a mobile number where your system allows, keep request forms to about five fields, remove dead booking links and keep a booking button visible in the first phone screen.
Read my takes first in Google Search
