Better than Free

Let’s face it. People like free stuff. They’ll listen to you pitch your product if it means they get a t-shirt at the end, and putting a “free!” label on your ad should make your customers click through to get their prize at the bottom of the box.

But it turns out that “free” is more complicated than we give it credit for.

If you just look at a straight comparison, “free” doesn’t come off very well. It beats out “$0” and “gratis,” for certain, at a decent margin. But when you start to look at its other synonyms, “free” begins to fall short. “Complimentary” beats out “free,” appearing in over three times as many successful ads. And “no cost” then beats out everything by an even larger margin, with over four times as many successful compared to unsuccessful ads.

 

The raw breakdown, then, is:

  1. No cost
  2. Complimentary
  3. Free
  4. gratis
  5. $0

But why is this? Why would “complimentary” and “no cost” beat the ever-recognized “free”?

To test this, we need to look at expanding the keywords. A comparison of “free shipping” and “complimentary shipping” shows that complimentary still wins out. But “no cost shipping” doesn’t exist in successful ads at all. In fact, “no cost” is never paired with a noun. Rather, it expands into “at no cost” or “at no cost to” and so on. So while “no cost” is the best out of all when compared directly, it’s far less versatile than “free” or “complimentary.”

So then we have “complimentary” versus “free.” Both can be used in conjunction with nouns, like “complimentary breakfast” and “free consultation.” And in all of those pairings, complimentary is the overwhelming winner. The exception to the rule is “complimentary consultation,” where “free consultation” is three times more likely to be found in successful ads.

But the pairings for “complimentary” are limited. It seems if it’s not a breakfast, or shipping, or consulting, no one uses it. And that’s where the deceptiveness of “free” comes in; its versatility as a word means it is used in more ads, which in turn can create a deceptive downturn in data. So while “no cost” and “complimentary” may be better converting, they don’t and can’t apply to all of the situations that “free” can, making it still a viable choice for advertising, when anything else doesn’t sound quite right.

And there’s your complimentary Convertasaurus results, at no cost to you.

Posted in Case Studies
  • Can you guys say a bit more about your methodology? Since you’re behind Spyfu I assume your “successful” ads are the ones with more longevity? Or more numerous? (ie assuming that profitable ads stay and unprofitable ads go away).

    Ian

    • You’re on the right track, Ian. “Successful” ads hold steady and show the advertiser continued to invest in it. Their high rankings combined with competitiveness (going against many advertisers) and longevity build them up as successful ads.

      • I think you ought to be a bit clearer about that when you talk about “successful” ads. When most people read the phrase “successful ads” they think the ad has a high CTR or ROI. You’re presuming that from the competitiveness and longevity of the ad. It’s a reasonable presumption, but not watertight.

        Ian

        • Actually, @SpyFu is correct in how we determine successful ads for SpyFu — but we do something different for convertasaurus.

          On SpyFu we try to figure out which ads are most profitable for your competitors. We know that people waste money on Adwords, but they don’t waste a *lot* of money for a *long* time. We also know that one of your competitors may not be smarter than you are — hopefully they aren’t smart at all. But, when *many* of you competitors are *all* betting on the same keywords, and you aren’t, it might be time to take notice. In other words, there’s wisdom in the market, but not necessarily in a single competitor.

          On Convertasaurus, we take that *market* wisdom to the extreme. Because we don’t have to limit ourselves to reverse engineering the secrets of *your* competitors, we can look at the entire Adwords market. And in a way our algorithms can get simpler.

          Basically, we take advantage of Google’s “greed” to figure out which ads get more clicks. Because the ads that Google chooses to display is based on eCPM, we know that *all other things being equal* Google will display the ads with the highest CTR most often and in the most prominent position.

          Our job, then is to control for the *all other things being equal* — quality score, bid, etc. That’s difficult, but the sample size billion-ish ads is sort of the secret weapon.

          So, that’s what we did.

          But, the proof, quite honestly is more in the output. When we try to make something “smart” with big data, when it goes wrong, it is really, really obvious to your gut. You’d see a pattern of wrongness immediately. I’ve done this a lot. Tried big computations and put them to the gut-check test. Failed many, many times.

          But Convertasaurus is different. I can’t outsmart it. If my brain is a sufficient Turing test, then Convertasaurus passes. 😉

          • Well, “yes and no” on the gut feel thing. If convertasaurus always agrees with your gut feel then you don’t need it. You can just go with your gut feel. Surely its greatest value is in highlighting surprising insights that you wouldn’t have been able to figure out with your gut or any other intestinal part.

            My main point though is that it would be good to see this algorithm more visible – have a “how convertasaurus works” section on the site. Then it’s transparent and you won’t have people drawing the wrong conclusions. And we’ll probably even more impressed.

            Ian

          • Hey Ian,

            Thanks for the feedback on the “how convertasaurus works” section. I think you’re definitely right. I think we will do that at some point. Right now, our content plans are to release more of these research-driven articles followed by an ebook. But, a nuts and bolts whitepaper on how it works would be cool.

            I see what you mean on the gut thing. But, while Convertasaurus confirms our expert *gut* as English-speaking- professional-marketers-versed-in-conversion-rate-optimization, it builds on our knowledge adding resolution to gray areas.

            Here’s an example (ripped from a future blog post):

            1. Incredible selection
            2. Massive selection
            3. Largest selection
            4. Big selection
            5. Huge selection
            6. Wide selection
            7. Great selection
            8. Fantastic selection
            9. Nice selection
            10. Good selection

            This list is in order from best to worst. Putting this list in order by myself, my *gut* would tell me that “Largest selection” probably beats “Good Selection” and that “Incredible Selection” is better than “Nice Selection”. Really, I could probably do a pretty good job that is *close* to the list above.

            But, Convertasaurus helps not only validate my wisdom, but it gives me additional resolution. I’m not sure I’d know where to place “Incredible Selection” relative to “Massive Selection” or “Wide Selection” vs “Huge Selection” — or really even “Huge Selection” vs “Massive Selection” (because “huge”, to me, seems almost as big as “massive” — and shorter which tends to be better, so… tough one).

        • That’s a fair point, Ian. We’re putting a lot on that presumption. Fortunately, we have seen good correlations with the same ads (and with SpyFu: keywords) that we’ve dubbed successful. True, it’s mostly anecdotal from those happy to share their own user data, but it’s a step in the right direction. You are right that it’s not watertight.

  • Svetlana

    It would be interesting to know the demographics and psychographics. Which groups prefer which terms.

  • Dan White

    It’s not that people don’t want “free” stuff, it’s that most have been taught that nothing of value is truly free; except air to breathe so far. However if you eliminate the doubt by revealing why its free and even how doing this benefits you, your prospects can switch from suspicion about the offer to making a decision as to whether accepting your offer is worth it to them. In an ad, making the word free, or even the other two that you propose, the anchor for it is a mistake IMO. I always think of using those words as a booster shot, much like the highly successful method of Ron Popeil.

  • Interesting that “complimentary consultation,” beat out “free consultation”

    With the versatility and frequency of the word Free do you suspect the data is accurate?

    My Hypothesis is that “Free” has a stigma of not actually being free, there’s always a catch.

    • Sidra

      We tend to agree with your hypothesis, Bobby. We’ve seen this pop up in a few different cases where “free” just carries the burden of having been tossed around in spammy offers. People have their guard up with anything “free.”

      This author does call this one out an exception. Where “free consultation” beats out “complimentary consultation,” I would *guess* that it’s the familiarity of that combined phrasings. “Complimentary consultation” feels a bit clumsy.

      For what it’s worth, “free consultation” also beats out “no obligation quote” which carries a similar meaning. As for the accuracy of that phrase working so well, we looked at 154,297 ads by 9,545 different advertisers containing this phrase to come to this conclusion.

  • We tend to agree with your hypothesis, Bobby. We’ve seen this pop up in a few different cases where “free” just carries the burden of having been tossed around in spammy offers. People have their guard up with anything “free.” In this exception, where “free consultation” beats out “complimentary consultation,” I would guess that it’s the familiarity of that combined phrasings. “Complimentary consultation” feels a bit clumsy. “Free consultation” also beats out “no obligation quote” which carries a similar meaning. As for the accuracy of that phrase working so well, we looked at 154,297 ads by 9,545 different advertisers containing this phrase to come to this conclusion.