How to track Punjabi and English leads so you know which one pays

The principle: one campaign, one trail

Every campaign should leave its own trail from the first click to the booked job. For a two-language business, that means each language gets its own:

• Budget line, so spend can be compared fairly
• Landing page address, such as /services/kitchen-renovation/ for English and /pa/services/kitchen-renovation/ for Punjabi
• Tracked phone number, shown only on that language's ads and pages
• Tagged links, so analytics knows which ad sent each visitor

When one of those is shared, part of the trail goes cold. The commonest culprit is the phone number: a Punjabi landing page that shows the same number as everything else makes Punjabi calls invisible.

Step 1: Set up separate page addresses

Google's guidance for multilingual sites recommends a separate URL for each language version. That helps search engines, and it helps you, because your analytics can then report English and Punjabi visits separately with no extra effort.

Keep the structure simple. A /pa/ folder for Punjabi pages is easy to understand and easy to report on. Avoid pages that switch language with a button but keep the same address, because they're hard to track and hard for Google to index.

Step 2: Give each language its own phone number

Call tracking numbers are local numbers that forward to your main line and log each call. You can get them from a call tracking provider at modest monthly cost per number. For a two-language test, you need at least:

• one number for English ads and pages
• one number for Punjabi ads and pages

Your main number stays on your Google Business Profile, vehicles and signage, so your listings stay consistent. The tracked numbers appear only where the campaign sends people.

If you also want to compare areas, such as Newton against Guildford, you can add numbers per area. Start with language, though. Two numbers you actually check are worth more than eight you ignore.

Step 3: Tag every ad link

Analytics tools read short tags added to the end of a link, called UTM parameters, to know where a visitor came from. Google Analytics explains how to collect campaign data with custom URLs. A tagged Punjabi Meta ad might end with something like:

?utm_source=meta&utm_medium=paid&utm_campaign=renovation-pa

and the English version with renovation-en. Pick a naming rule and stick to it. The language code at the end of the campaign name is what lets you filter later.

Step 4: Keep a weekly booked-job sheet

Tracking tells you who called. It doesn't tell you who booked. For that you need a simple sheet, filled in weekly by whoever handles bookings:

Ask whoever takes the enquiry to note the language the customer used. That catches the customer who saw a Punjabi ad and then called your main number. It takes ten seconds per enquiry and it's the most valuable data you'll collect.

Step 5: Read the numbers at 60 to 90 days

Give each language a fair run before judging. As a rough guide, wait for at least 20 to 30 enquiries in each language. Then calculate, for each language:

• cost per enquiry = spend ÷ enquiries
• booking rate = booked jobs ÷ enquiries
• cost per booked job = spend ÷ booked jobs
• revenue per dollar spent = revenue booked ÷ spend

A worked example

A Surrey landscaping company runs both languages for 90 days:

Both campaigns are working. Punjabi is cheaper per booked job, and the booking rates are close, so handling isn't the issue. The sensible next step is to move some budget into Punjabi and watch whether its cost per job holds as spend rises. These figures are an illustration of the arithmetic; your test will produce your own.

Reading the warning signs

Some patterns point to a specific problem:

• Punjabi enquiries are cheap but rarely book. Look at handling first: callbacks, quotes, who answers. Our guide to handling enquiries in both languages covers this.
• Punjabi page visits are high but calls are low. Check the page itself: Is the phone number visible on a phone screen? Was the copy written by a native speaker? Does it match the ad?
• Both languages cost more per job than a job earns. The problem is upstream: targeting the wrong areas, the wrong searches, or a price point the ads can't support. Revisit your area plan.
• Lots of "either" enquiries. Staff aren't recording language, or customers are skipping the tracked numbers. Retrain on the sheet before drawing conclusions.

Common mistakes

• Putting the tracked number on your Google Business Profile. It can confuse your listings. Keep your main number there.
• Judging after two weeks. Small numbers swing wildly.
• Comparing cost per click. It tells you nothing about booked work.
• Forgetting walk-ins and referrals. Note them in the sheet so you can see the whole picture.
• Changing everything at once. Change one thing per language, then measure again.

Where this fits

Tracking is the backbone of our Punjabi and English ad campaigns, and it's the reason our reports lead with cost per booked job. If you'd like help setting it up, see pricing, browse the other guides, or call, text or email us.