How to improve your web searching
As a market researcher, you get to ask strange questions. And sometimes, your clients ask you strange questions. I was recently contacted by a very flustered group account director from a boutique advertising agency.
She said: “We’ve been asked to pitch for Brand X Coffee, but we’ve never had a coffee account. They started talking about ‘baristas’ and I thought they meant they had a legal problem! You have got to help me sound knowledgeable!”
With the first client briefing on the pitch in a week’s time, and no time or budget for market research, there was only one solution: the web. Maybe there would be sufficient free or cheap information on the web for us to cobble together a background briefing on all things ‘coffee’.
Googling ‘coffee’ gave us 347 million hits – so obviously some focus would be required! Here is where most people get into trouble. It is easy to go to a search engine and type in a few words, but extracting useful information can be more challenging. Ten years of searching the web for clients and 20 years as an information professional (market researcher) has given me a few ideas on the topic of searching secondary data.
The basics of searching the web include:
1. Learn more about search engines and searching. It sounds boring, but a few minutes of background reading will help you to understand more about web searching generally and will improve your effectiveness. I recommend www.monash.com/spidap.html as a great place to start.
2. Think before you search. The few minutes you spend working out what you really need to know will save you hours of sifting through useless stuff. Think of it as a brief that you are writing to yourself and use similar headings: What am I trying to achieve? What are the parameters? What are the mandatories? Etc.
3. Search efficiently.
Search instructions. Using syntax instructions such as “” and + can dramatically improve the accuracy of your searches. Quotation marks (putting quotation marks around a phrase, e.g. “Mark Fletcher”) means that you only get those results that include that exact phrase. Using + (putting + between words, e.g. Australian+coffee) means you only get those results that contain both words.
Multiple searches. Everyone uses Google, and it is great. But even the best search engine probably only covers 10% of the web, and different search engines give different results and have different ranking methods for displaying their results. Therefore, it makes sense to use at least two different search approaches, e.g. www.google.com and www.yahoo.com. Some people also recommend metasearch engines (which simultaneously search multiple search engines, e.g. www.dogpile.com and www.clusty.com). You should be aware, however, that the results from metasearch engines can be confusing and that some metasearch engines remove any syntax, such as “”, from your searches.
4. Be a detective. You will often find that a certain author/researcher, website or related topic will be cited in your web search. Be prepared to spend a few minutes directly searching on these new terms as they may lead you to useful information that did not emerge in your original search. I keep a blank piece of paper beside me while I am searching to draw a ‘mind map’ (www.en.wikipedia.org/wiki/Mind_mapping) of related topics that emerge as I search.
5. Look in the ‘invisible web’ (the 90%+ of websites not covered by search engines).
Searching. You should try some of the free invisible web search engines and directories such as ‘direct search’ www.freepint.com/gary/direct.htmand www.completeplanet.comjust to see what else is out there.
Databases. Proprietary databases such as ProQuest (you may have used this one at university) can also be good. My ProQuest subscription is not expensive and was obtained via the Monash University Alumni, but there may be other methods of subscribing.
However, the most important thing to remember is that data only becomes information when you can put it into the context of the decisions you need to make. It always takes time and thought to extract and organise relevant information from the web.
For example, my ‘Coffee’ report (57 points extracted from 19 articles/websites extracted from 300 million + ‘hits’) took me more than 10 hours to produce (you can see it by clicking here (PDF file) if you want to have a look at the final result).
Still, searching the web was cheaper and faster than commissioning market research, even if the ad agency still didn’t get the account!
|
