Using Adwords for research


Sometimes, including on this forum, I read advice to use Adwords for market research. Looks simple: find a few terms that describe your product idea, see how many people are searching for it, put up an ad, see how many people click etc.

I am trying to find a niche in the analytics/reporting/bigdata space. So far, I learned that most keywords cost upwards of $2/click. But even as they get to the landing page, it is difficult to say if they are genuinly interested or are just tyre-kickers.

If anyone here has actually done this, I would much appreciate them sharing the experience, particularly about:

  1. Does click-through-rate matter? you would probably get 1-2% CTR if you advertise snow to the Eskimos. what kind of CTR you are looking for to say a definite no or yes, if at all?
  2. Without having a website, except a couple of landing pages, what techniques did you use to determine genuine interest?
  3. if all you had was a contact form or phone ## on the landing page, did people actually contact you?

Adwords Research Pay Per Click

asked Nov 7 '12 at 19:35
Sergei Veinberg
429 points

1 Answer


I'd say that a CTR of 1-2% is pretty good. But you're right that on its own, it's not enough to rely on when researching a market. All a click-through means is that your ad was interesting, and what you want to measure is how interesting your idea as explained on the landing page is.

One thing I've heard recommended is to put a form on the front page saying "Sign up and we'll tell you when the beta is ready". Just ask them for their email address, absolutely nothing more.

That's much more less effort for the site visitor than calling a phone number or filling in a more complicated "contact" form, but it does make sure than anyone who uses it is genuinely interested -- at least, interested enough to run the risk of you spamming them later on!

answered Nov 8 '12 at 04:34
Giles Thomas
1,540 points
  • Not everyone knows what a beta is. It's great advice, just make sure you get the messaging right as it will greatly affect conversions. A bit of A/B testing on the landing page wouldn't hurt if you have time. – Joel Friedlaender 10 years ago
  • Thanks Giles. I actually want to have a conversation with them, understand who they are, what they are after - see if the whole opportunity is worth pursuing. – Sergei Veinberg 10 years ago
  • That's a good idea. When they sign up, you can send them a thank-you email in which you can ask them few questions - what they plan to use your product for, etc - and use that to start the conversation. The percentage of people who reply to that might be an interesting metric too. – Giles Thomas 10 years ago
  • so far no one signed up - but then again, although the CTR is around 2%, i only collected about 20 clicks on the best themes. I am thinking now to focus on these best themes, and on the landing page have a few links, eg for visitors of type A, of type B etc. they will click and i will get some idea of who they are. – Sergei Veinberg 10 years ago
  • 0 signups out of 20 is definitely on the low side for signups, I'd aim for at least 2 or 3. You could consider different landing pages with text customised so that it relates specifically to the text in the ad the user clicks. Ultimately, though, if someone's not willing to give you an email address now, they're probably not willing to give you money later -- so if you're testing an idea you should certainly take that into consideration. – Giles Thomas 10 years ago

Your Answer

  • Bold
  • Italic
  • • Bullets
  • 1. Numbers
  • Quote
Not the answer you're looking for? Ask your own question or browse other questions in these topics:

Adwords Research Pay Per Click