Is a 40 percent view rate good?

It is a fact about the video, not about the campaign. A high view rate on a badly targeted audience means you paid to entertain people who will never buy. Read it next to what happened to your enquiries and your branded search. A sudden drop in view rate is worth acting on, because it means the creative or the audience has gone stale.

Should I count view-through conversions in my cost per lead?

Report them, but on their own line. A view-through conversion means somebody saw the advertisement, did not click, and later converted inside a window Google chose. That is a reasonable clue and a poor invoice. Show the board two numbers: conversions from clicks, and conversions credited to views.

How do I know whether the television screen in the living room did anything?

You cannot trace it directly, because nobody clicks a television. Hold one city out of the buy, run the same weeks everywhere else, and compare branded search and enquiries between them. It is the only honest read available to a business of this size.

My YouTube campaign got thousands of views and no enquiries. What went wrong?

Usually the buy, not the video. Views are cheap to produce on a broad audience, and a broad audience has no reason to act. Check where the views came from by placement and geography first, then check whether the video ever asked for anything.

Do I need brand lift studies?

Not at a small budget. Those studies have spend minimums most Indian small businesses will not meet, and the same question is answerable with branded search volume, direct traffic and a holdout city. Spend the money on more weeks of media instead.

Notes for owners · Digital marketing

How to measure YouTube advertising

Views and view rate tell you about the video. Branded search, a holdout city and the question your receptionist asks tell you about the business. Decide which one you are buying before the campaign starts.

The GullySales team · Updated 21 Sept 2026 · 8 min read

YouTube will hand you views, view rate, cost per view, watch percentages and clicks. Those describe the video. To judge the campaign you need three other things. What happened to searches for your name during and after the flight. What happened to enquiries where you ran, set against an area you deliberately left out. And what people say when your own team asks them where they heard of you. Decide which of these you are buying before the first rupee moves, because the measurement has to be in place first.

A view is not a viewing, and the formats do not agree

On a skippable advertisement the Skip button appears after five seconds, and a view is only counted once somebody stays for thirty seconds or watches to the end. So a view is a real signal of attention. A short unskippable advertisement is sold on impressions instead, because there is nothing to skip and nothing to measure. An in-feed advertisement, the one that sits in the results list, counts a view when the person chooses to click it.

Three formats, three definitions, one column labelled "views". Compare view counts only within a format, never across them. And never let a seller put the three together into a single reach number.

What each number on the report is actually saying

Number	What it really measures	Where it misleads
Impressions	How many times it was served	Includes plays the viewer skipped at second six
Views and view rate	Whether the opening held attention	A high rate on the wrong audience is expensive entertainment
Cost per view	Auction pressure and audience scarcity	Falls when you broaden targeting, which feels like progress
Watched to 25, 50, 75, 100 percent	Where the story loses people	Tells you about the edit, not the buy
Clicks and click rate	Immediate intent	Video rarely produces clicks. Judging it on clicks alone kills good campaigns
Conversions from clicks	Real, traceable response	Under-counts, because a person sees it on television and searches on a phone
View-through conversions	Google's estimate of delayed credit	An assumption with a window attached, not an observation

The two campaigns that look the same and are not

A response campaign sends people somewhere. It gets its own landing page, its own phone number, and conversion tracking that works. You judge it on enquiries and cost per enquiry, and you can be strict about it. Marketing attribution is the service page if you want that route built for you.

An awareness campaign is buying memory. Judge it on cost per lead and it will look like a failure every time, so agree the measures before it starts. Searches for your brand name. Direct visits. Calls and direction requests on your Google Business Profile. And the share of new enquiries that arrive already knowing what you do. The method for reading those signals is in how to measure brand awareness advertising, and the same rules apply here.

The mistake is buying the second and reporting on the first.

The holdout, which costs nothing and settles arguments

Pick a city or a district you were going to run in and leave it out. Run everywhere else for six to eight weeks. Then compare the two on branded search volume, website visits and enquiries, before, during and after.

For example, a Karnataka home appliance brand running YouTube across Bengaluru, Mysuru and Mangaluru could hold Hubballi out of the plan. If enquiries and brand searches rise in three cities and stay flat in the fourth, the video did something. If all four move together, something else moved them, and it was probably the season.

Two conditions make this work. The held-out area needs a similar baseline, so do not hold out your smallest market. And nothing else can change there, so do not put a hoarding up in Hubballi the same month.

Ask the person, because video is remembered as video

Put one open question in your enquiry script and one field in the CRM: where did you hear about us. Not a dropdown of your own channels. A blank box the person on the phone fills in with what was said.

Television and YouTube are remembered vividly and described loosely. "I saw your advertisement" and "it came before a video my son was watching" are both YouTube. Read that field monthly against the weeks you were on air, and you will know more than the dashboard tells you. We record enquiries by source before any campaign starts for exactly this reason, so there is a before to compare the after against.

What the platform cannot see

Who else was in the room. Whether the person who searched your name at nine that night was the one who watched it at seven. Whether a skip at second six was rejection or a customer who already knows you. Whether the enquiry that came a month later was seeded by the video or by your neighbour.

None of this makes YouTube unmeasurable. It makes it measurable in aggregate and over weeks, like every medium that plays to a room rather than to a browser.

When the measurement is not the problem

If the video runs for two weeks, no measurement will tell you anything. Memory needs repetition, and a fortnight buys none.

If the landing page is the home page, conversions will be missed no matter how the tracking is built.

And if the campaign was bought as reach across India while you deliver in four districts, the report is accurate and the plan is wrong.

What to do next

Before the next flight, write down three numbers as they stand today: searches for your brand name last month, direct visits last month, and enquiries last month by source. That takes an hour with Search Console, your analytics and your enquiry register. Without it, whatever the campaign does will be arguable in both directions. If you would like that baseline set up and read with you each month, book the free audit and bring last quarter's numbers.

Related services