Online Research

There’s no question that search engines have made the process of gaining knowledge, and researching products and services, extremely easy — often nearly effortless. The search bots do the work for us, helping us find the answers we’re seeking in the blink of an eye.

So what’s not to love about search?

The thing about search engines is that the algorithms “reward” the purported wisdom of crowds – particularly since there’s more social interaction on websites than ever these days. It’s one thing for developers to optimize their websites for search – but there’s also the behaviors of those doing the searching and interacting with those same websites and pages.

Whether it’s tracking how much time visitors spend on a page as a proxy for relevance, or how visitors may interact with a page by rating products or services, the bots are continually refining the search results they serve up in an effort to deliver the highest degree of “relevance” to the greatest number of people.

But therein lies the rub. Popularity and algorithms drive search rankings. If people confine viewing of search results to just the first page – which is what so many viewers do — it limits their exposure to what might actually be more valuable information.

Over time, viewers have been “trained” to not to look beyond the first page of online results – and often not beyond the top five entries. That’s very convenient and time-efficient, but it means that better information, which is sometimes going to be found in the middle of search results rather than at the top, is completely missed.

As we rely more on ever-improving software, it’s tempting to assume that the search algorithms are going to be more and more airtight – and hence more effective than human-powered expertise.

But that isn’t the case – at least not yet. And a lot of things can slip through the gap that exists between the perception and the reality.

It turns out, there are some predictive factors.

One of the nice things about the proliferation on online surveys in recent years is that, over time, we’ve come to understand survey response dynamics much better.

Of course, predicting response rates with flawless precision is impossible due to the individual attributes of each individual survey, the sample composition and so forth. But thanks to a 2015 compilation of “bottom-line” information by content marketing specialist Andrea Fryrear, the following points are good ones for marketing personnel undertaking market survey work.

Surveys aimed at “internal audiences” outperform external ones.

Targeting an internal audience such as a company’s own employee base is likely going to generate higher response rates (in the neighborhood of 35% to 40%, give or take). For surveys of an external audience, it’s more like 10% or perhaps even lower.

The reason is simple: Surveys aimed at internal audiences are likely very-well targeted, whereas with an external audience, often it’s difficult to reach only the right type of respondents. At least some of them will turn out to be poor targets.

Additional motivating factors.

Other factors that can influence survey response rates include:

Customer loyalty – People who feel a connection with the brand conducting a survey tend to be more likely to participate.
Brand recognition – Surveys that focus on well-known brands will typically outperform ones from an unknown source or dealing with unfamiliar brands.
Perceived benefit – The “WIIFM” factor. For example, response rates can soar even higher if the respondent population is motivated by serious incentives. I recall getting more than a 60% response rate on a mail survey and an external sample because the monetary incentive was a $2 bill.
Demographics – The reality is that certain segments of the population are more likely to respond to surveys than others. Think everything from age and gender to ethnicity and geographic location.
Survey distribution – Certain audiences are used to interacting on social media … others online … still others offline. Chances are, you already know which type of research targets those are within your target markets, and it should influence your choice of survey delivery.

Survey length can make or break your response and completion rates.

To achieve the highest response rates, ideally surveys should take five minutes or less to complete. Ten minutes or less is probably OK, too. But anything longer than that will likely have deleterious effect on your response rate.

How many questions does this mean? On average, respondents can complete five closed-ended questions in a minutes’ time … but only two open-ended ones.

Survey reminders? Yes.

Particularly with online surveys, it’s a good idea to send reminder notices to those who haven’t completed surveys as you get closer to the cut-off date. Sending two or three reminders is a good rule of thumb … and try sending them at different times of the day or different days of the week to that you can reach as many different prospective respondents as possible.

Learning from the experience of the thousands of surveys administered every month should make it easier for marketers to ensure their next survey will generate successful results instead of flame out. There’s really no reason for failure considering the wealth of “experiential information” that’s out there.