Notes for owners · Digital marketing
How to track offline advertising
Offline is not unmeasurable. It is differently measurable: you give up the count of who saw it, and you build a response route the campaign can be read through.
The GullySales team · Updated 21 Sept 2026 · 7 min read
Offline advertising can be tracked. What it cannot do is count the people who saw it and did nothing. So the trade is simple: you give up the impression, and you build a route a responder can be read through. That route has to exist before the campaign runs, printed or spoken into the creative itself. A number nobody else uses. A code per site. A word the caller has to say. A web address that belongs to one panel. Then you read response against a baseline you wrote down first.
Two different questions, and only one of them is about response
Every offline campaign has to answer two things, and mixing them is where most arguments start.
The first is whether it ran. Did all nine hoardings go up on the date in the order? Did the spot air at the times on the log, and did the insert reach the pin codes on the plan? That is proof of execution, and it is a paperwork problem with a clean answer.
The second is whether it did anything. That one is never clean, in any medium. A digital dashboard hides the uncertainty behind a decimal point. Offline puts it in front of you, which is uncomfortable and more honest.
Do not let anyone tell you digital is the measured one
This matters because the comparison is usually rigged before it starts. An owner is shown a Meta report with 41 conversions and a hoarding with nothing, and concludes the hoarding failed.
Look at what that 41 is made of. The ad platform counts a conversion when it can connect a click or a view to an action. Google counts the same order from the same buyer. Add two platform reports together and you get a number bigger than your sales register. A share of every journey is missing anyway. Cookie lifetimes on iPhones are short, consent choices remove people, and a click arriving through WhatsApp lands with its tags stripped and is recorded as direct traffic. Some of what you are reading is modelled, which is a polite word for estimated.
So the real difference is not measured against unmeasured. It is a medium that reports on itself against a medium that does not report at all. Both need your own enquiry register to tell you the truth.
The seven routes, and what each one actually catches
| Route | What it catches | What it misses |
|---|---|---|
| A QR code per placement | Scans from a standing position, with the placement attached | Anyone in a moving vehicle, and anyone with an old phone |
| A phone number per site or creative | Calls made while the advertisement is in front of them | Everyone who searches your name and calls the website number |
| A promo code or code word | People who claim it at the counter or in the shop | Everyone too shy to ask, which is most of them |
| A vanity URL per medium | Typed visits, and everything they do on the site afterwards | Anyone who searches instead of typing |
| Asking the caller | The route no software can see | Whatever the caller misremembers, which is plenty |
| Branded search and direct traffic | Movement in how many people are looking for you by name | Which of the four things you ran caused it |
| Footfall and walk-ins | The response a clinic or a showroom actually lives on | Anyone who came in and did not buy |
Run two or three together, never one. The phone number and the counter question are the pair that earn their keep in almost every small business, and neither needs software.
Two of these deserve their own pages, because they are where campaigns go wrong in detail rather than in principle: QR codes and call tracking.
Branded search is a signal, not a scoreboard
When a hoarding works on somebody, what they usually do is nothing for a few days, and then type your name into Google. So watch the volume of searches that contain your name, your direct traffic, and the profile views on Google Maps, before the campaign, during it, and for a month after.
Treat it as a direction. The volume figures are estimates, and a news story or a job advertisement will move them. You cannot separate the radio burst from the hoardings if you ran both in the same fortnight. Rising branded search across an eight-week outdoor campaign is worth something. A precise percentage credited to one panel is worth nothing.
The baseline is the measurement
This is the part owners skip and then regret. Before anything is mounted, record four things weekly: enquiries by source, walk-ins, orders, and average order value. Write them down. Then hold everything else still while the campaign runs, and do not raise the Google Ads budget in the same fortnight the panels go up.
If you trade in more than one town, keep one of them as a control. Run the outdoor in Davangere and not in Shivamogga, leave both digital budgets alone, and compare the movement. It is not a laboratory. It is the closest a business of your size will get to an answer.
Write the trace on the media order
For example, a bathroom fittings dealer in Hubballi books six hoardings, an insert in four pin codes, a Kannada daily strip on Sundays and a local radio burst. Before signing, he adds a column beside the cost of each line: how this line will be traced. Six site numbers. A coupon word on the insert. A vanity address on the strip. On radio, nothing but branded search, and he writes that down as the answer rather than pretending otherwise.
At the end of the quarter he knows which two sites produced calls, which pin codes redeemed the coupon, and that both weeks of radio moved his name searches. Three of the six panels go. That is what tracking offline is for.
What to do next
Take your last media order and try to fill in that column line by line. Every line with no answer is a line you bought on faith. Faith is allowed, as long as you are choosing it rather than discovering it later.
If you want the routes set up and a baseline recorded before your next campaign runs, book the free audit. It is a 45-minute call about how enquiries reach you and what happens to them, and the written report that follows is ranked by what to fix first.