Sponsored Search: Question 5 - Keyword Determination

In the fast moving market of sponsored search, how do you choose your keywords for the near future? Is it possible to use past results in some way to determine the best keywords for the future?

To say it differently, before / today, we know what media are most effective. So we produce ads for these media and we fairly know how many people will see it, who will see it and how many will probably remember it.

On the web, to be seen, you can buy keywords. How would you choose them?
=> Do you believe you can launch a study to know what words are most commonly used to describe ... your brand ? your market ? or even your competitors ?
or
=> Do you believe the web is always ahead of every trend ?

Hi,

On that particular topic, i think there is a twofold responsibility:

- The marketing team inside the company should be able to determine which kind of keywords really fit the best the business. Indeed, the marketing team is responsible for developing the image of the brand, and that's the primary link with the customers. On that account, to determine the proper keywords, the company should focus on its marketing strategy and also on its users. It could organize focus group to know the trend on the specific market, it could also monitor its competitors to know how they choose to promote their products (since we know that the price of keywords is linked to the competition on the market)

- The ad service provider should have someone who is responsible for managing the online ads campaign, and to make sure that you spend your money in an efficient way. At google, these guys are called Account Strategist. Their goal are to maximize your investment and also Google revenues.

Cheers,
Laurent

I think there are also people working in companies with the task to care about search engine optimization, so called SEO. My opinion is that this can be a good investment because I have the experience that a good choice of keywords can save money which would have been spend for google adverts for example.

And I think it is not only about determining the keywords which fit the most to your products, you also have to look at your competitors and the customers. So together with a good planned google adverst campaign and skilled people of analyzing your SEO environment this can really help your business.

SEO in the search engine company is really focusing on optimizing the strategy both for advertisers and search engine company itself. Till now many useful tools have been developed for the advertisers, for eg., diagrams and tables of tracking the keywords frequency changing trend...
Regarding to past results, I think it's already been used by some search engine companies as data set for the analytical tools, it's an important method for trend forecast. Advertisers could choose to buy them as a value-added service. More such kinds of tools will be developed and the fee for them could be lower and lower.

In sponsored search, in experience with using Adwords before, as long as you have budget, you can have as many keywords as you want. In order to save budget, I thought Sysomos can be good source to know what kind of words are frequently used by internet users.

This might be a little bit off to the topic, I guess not only sponsored search result but also to bring your company or product web site on top at regular search result is also very important. I am pretty sure that you can not cover every keywords you can think of due to limitation of budget. So we have to put into account for regular search result when building web pages.

Using past results to determine the future has its limited. This way is good for the firm that stable and not very creative. If there wasn't any big change in last ten years, then we can suppose it won't get any big change in near future.
But when we talk about fast moving market of sponsored search, it's not easy to determine key words by past experiences. Past results can only provide a reference, to determine keywords, we need knowing ourselves very well, know our position very well, too.

I think the company usually buy the keywords which is relevant to their products or services for the purpose of making consumer associate their product with the core value and position of their products. In my opinion, the purpose of adWords is to make my product be easier to be searched or make my product more famous. But there are too many choices that I can't afford them all. So I think I will only buy the keyword which describe my product features well due to the limitation of budget.

I believe most of search engines all engage in researching the popular keywords in the future. But sometimes as things change so fast that they can't estimate everything accurately. So for the adWords buyers, I think the most important thing is to position themselves in the market accurately and distinguish their product features from others'. That will be more efficient when buying a adWords.

After reading this journal article on sponsored search, I got the feeling that search optimization were primarily for small-medium size enterprises. That is to say, if you focus on a hit product, eventually it will go viral by itself and take on its own life. There is no need to "game" the system and hedge bets for keyword bidding when you are absolutely confident that your product is a hit. Since consumer generated content is so ever expanding, the masses will determine what is "buzzing," not that the other way around. At best it can fan the flames but in doing so it becomes like any other market and susceptible to a bubble. In sum, many smaller companies use search words for leverage to augment sales, but the truly successful products (iPad, iPhone, etc) will organically surface through word of mouth and on the various social media sites. Instead of allocating resources for advertisment budget, one might be better off in creating a stellar product.

I think the issue of buying keywords is central.

Whenever I'm looking for information on products or a market, the first thing I do is google it. Then, most often, I don't go further 4-5 websites that match my enquiry. So if a brand doesn't appear in one of these, I won't buy it for sure.

As a result, a brand has to make it knows what keywords researchers will use most frequently to be on the top of the research responses. In my opinion, the best way to do it is:

1. Brainstorm amongst the company. The brand looking to buy keywords has to think internally what are the best keywords. Noone knows a company better than its employees. As a reult, a survey has to be launches internally to determine the employees thought.
2. Next step is to make a survey amongst the potential buyers. This can be made on the street if the products focuses on a large population. Question would be "what comes to your mind when we mention this brand?". The first responses are the keywords the company has to buy.
3. Last but not least, the company has to be dynamic. Research engines offer a lot of tools to see which keywords were most used by researchers. The company should adapt to the researches and evolve to match the consumers' keywords best.

I find out a website can watch the ranking.
http://www.googlerankings.com/
I think making some survey will be helpful.
Sometimes we don't only base on the past, we can also base on others.

I think another tactics that has been widely in use at a companies expert in online marketing like Groupon or PriceMinister has been to constantly surf on short-lived but highly mediatised trends in the news benefiting from the ponctual surge in research for the lexical field of the concerned news. MacDonald's launches a new burger, Groupon will buy the keyword New burger and ads featuring discount on restaurants and fast food are going to show up on front page of the search engine!