Skip to content

4 ROI Myths

2010 February 2
by Greg

Everybody wants value for money.  Accountability is a crucial element of any business relationship.  Shouldn’t there be a simple formula to evaluate return on marketing investment?

Unfortunately, there isn’t one and there never will be.  Running a business requires judgment, skill and experience.  Any simple rule of thumb is bound to be either so narrow that it misses the big picture or so general that it’s meaningless.

That doesn’t stop people from trying though, which is why we’re inundated with ROI myths.  Here are four of the most damaging:

Response = ROI

There are many ways to improve response.  Put a scantily clad woman on a banner and watch CTR’s skyrocket.  That doesn’t mean that there is a return on investment, it just means that you got people to click.

CTR’s for banners are usually just a few percent at most.  Rich media and e-mail can be considerably higher (I’ve gotten as high as 20% for e-mails).  However, even in the best cases, the overwhelming majority do not immediately respond.

Focusing solely on direct and immediate response ignores ensures that you will miss the opportunity to communicate effectively with consumers.  Many purchases require significant investment and consumers put a great amount of thought into them.  A Pavlovian response is of dubious value.

Effective brand communication is not a pick-up line. The goal is to build a relationship, not a one night stand.

Sales = ROI

Sales numbers can be deceptive as well.  It’s easy to flood sales channels for a quarter or two.  Current sales say little about long term prospects.

Competitive marketing activity also needs to be taken into account. Many companies will see an uptick in sales as their competitors go dark in preparation for a big launch campaign.  Should the calm before the storm be seen as delivering effective ROI?

Moreover, focusing exclusively on sales promotions may produce results in the short term, but the brand will decay in the medium and long term, making future sales promotions less effective.  This effect is very real and can be measured econometrically.

As this paper from the Marketing Science Institute shows, two thirds of return on brand investment shows up in long term performance. That doesn’t make the return any less real, just less immediate.

Suppliers should be responsible for ROI metrics

Unless a supplier has adequate access to internal client data, any ROI estimate will be incomplete, which is why I’ve argued that most of the talk about digital ROI is a waste of time.  Unless all marketing activity is taken into account, any evaluation is more likely to confuse than to explain.

To really do ROI right for an integrated marketing program, sophisticated mathematics, comprehensive data and expensive software need to be utilized.  It’s not a part time job.  Suppliers are responsible for delivering a product, not evaluating marketing effectiveness.

Moreover, the success of the outdoor advertising industry shows that marketers are more than able to evaluate effectiveness even when media metrics are not easily obtained.

One Size Fits All

Different categories have different needs, different companies have different strategies and different campaigns have different objectives.  You can’t evaluate a one channel campaign for an e-commerce site the same way you do a multibillion dollar ad budget.

That’s why we have marketing briefs, so that campaigns will be executed according to specific objectives..  To assume that every marketing program should be evaluated the same way is just foolish. (For a more in depth discussion of various approaches, see this ROI post).

How to Approach ROI

ROI is an evaluation, not a calculation.  Business isn’t “paint by numbers,” but requires judgment. Simple rules are for simple people.

While there is no general formula for assessing ROI, every serious effort should include:

Objectives: Both brand and sales objectives need to be taken into account.  There’s no sense in pumping up sales while your awareness numbers plummet.

Competitive activity: An often overlooked factor in marketing effectiveness is competitive activity.  Share of voice can usually explain results better than a simple budget analysis.  That’s why many marketers closely monitor their aggresivity index (expenditure share / market share)

Integration: There’s very little utility in evaluating one part of a marketing program in isolation.  To evaluate media mix, techniques such as factor analysis and multivariate modeling can point the way.  An effective integrated marketing program works synergistically, so there’s no point in acting as if any one part is working on its own.

ROI is a complex issue and shouldn’t be taken lightly.   But with some effort and good sense, myths can be avoided and results delivered.

- Greg

19 Responses leave one →
  1. February 2, 2010

    Nice post.
    Thanks for keeping this simple.

    [Reply]

    Greg Reply:

    Paul,

    Thanks. I’m glad you liked it.

    - Greg

    [Reply]

  2. February 3, 2010

    Thanks for the post Greg. Tracking ROI for marketing efforts is always a little fuzzy. It is much easier to track a specific marketing initiative, but even those results are not 100%. If a business wants to track the success of an email blast for an example, they can use database tracking programs to see how many sales they actually do get after sending out the email blast from the email recipients. However, there might be other factors that are actually involved with a particular sale from an email recipient. Maybe a recipient saw an advertisement or got some positive feedback about someone’s company so that when they received the email blast, their interest was already peaked. The longer the sales cycle, the more likely it is that other variables could be involved. No tracking software can account for these possibled variables. Now add social media marketing to the mix and ROi tracking results will be even more skewed. How can you really determine why a particular email recipient of your specific campaign became a customer? You can’t. Any results obtained from tracking will only be estimates. Why are people still searching? Because they have to. They gave to either justify to management or themselves if they are a business owner, that their marketing efforts are paying off. Your view, which I obviously agree with, is not a popular one – but in my opinion, it is the only realistic one.

    [Reply]

    Greg Reply:

    Julie,

    Thanks for your input. However, I’ve found the opposite is true, that you can gain insight easier by tracking the whole budget over time and using econometrics.

    There’s too much noise in one campaign or to know if the results are seeing are real. There are many cases when online agencies have presented their “ROI analysis” to clients only to find that the peaks in response correlated much more closely with the TV budget than the supposed optimization.

    The point isn’t whether to show whether marketing works (if the client has a budget, they believe that marketing works), but to understand why and continually optimize.

    - Greg

    [Reply]

  3. February 3, 2010

    Great post! Nice and simple. Too often people don’t look at longevity (sustained impact) for ROI and just the “sudden impact” and they base organizational strategy on this “sudden impact.” I’ve known a few such companies!

    Thanks again for the post!

    Best regards,
    Gina

    [Reply]

    Greg Reply:

    Gina,

    Thanks. I’m glad you liked it.

    I think it’s natural for people to have a bias for accessible information over complete information.

    - Greg

    [Reply]

  4. February 3, 2010

    Greg; Another great post which should help to dispel some of the myths. Also good comments, as usual. I think we all need to understand that calculating the exact or precise ROI of something like a marketing campaign is a very difficult task indeed, but calculating the approximate values is much easier. I guess that’s probably true of most things, but in the case of ROI, trying to be precise means digging in deeper and deeper into more abstract and difficult to calculate values. Just how does on find out which campaign produced this lead who arrived on our site via a Social Media link (left behind in a comment), but who in fact had heard of us from before and was just now being reminded to visit our site?
    The link at the end of this comment is to an index page on our blog which contains 3 posts on the subject of calculating ROI which your readers may find useful. The 1st post describes how to calculate the ROI of your website as a whole, the 2nd provides a list of the 10 best free ROI calculators on the web, and the 3rd explains how to build your own ROI calculator, for use, for example, in calculating ROI of Social Media Campaigns.
    http://bit.ly/bKPKRp

    [Reply]

    Greg Reply:

    Eric,

    You’re right. It’s always better to be roughly right than precisely wrong.

    One thing that I think gets lost is that for an ROI process to have value it must be able to predict future performance. That’s how you can improve your marketing, by identifying factors that raise performance.

    - Greg

    [Reply]

  5. February 4, 2010

    Greg – nicely put – roughly right is better than precisely wrong! And totally agree with you about using the numbers to predict future performance. We like to throw in the old adage of “50% of my marketing budget is wasted – I just don’t know which half!” and then add by using ROI you can solve the problem. I added this statement as to our way of thinking, using ROI is the only way to focus on what is working and to stop what isn’t.

    [Reply]

  6. February 17, 2010

    Solid list and some superb posts. Timely because I just posted a question on a board looking for differing ideas about how we help clients understand the difference between “Response” and “ROI”.

    Not in agreement on “2/3 of brand impact is in the future”. That idea (even if true for some cases) is too often the excuse offered by people who’ve created ineffective advertising.

    So, I’d suggest that, as you suggest about how to calculate ROI, the potential future value of brand activity depends on your company, your needs, and your products.

    [Reply]

    Greg Reply:

    Doug,

    Thanks. I agree that many people use the idea of future payoff as an excuse, but that too can be evaluated.

    - Greg

    [Reply]

    Doug Garnett Reply:

    The difference between “can be” evaluated and “is” evaluated is perhaps the real challenge.

    In large part, the training to become an ad professional is, um, not ideal for learning to do analysis with numbers. Hence, I don’t find that it’s often done. (Says the unusual ad guy with a master’s in math.)

    To your point, the companies who realistically project future value find they use their ad money with MUCH higher impact.

    Thanks for the great ideas…

    [Reply]

    Greg Reply:

    Doug,

    Thank you!

    btw. I also agree with your assessment of math skills in the industry. Even major global networks that have econometric divisions don’t integrate the skills very well. It’s a big problem.. I wrote another post about it if you’re interested.

    - Greg

  7. March 5, 2010

    ROI should be measured in terms of customer lifetime value and not against individual campaigns or channels. In order for this to happen, marketing must be looked at as a strategy with specific business goals.

    [Reply]

    Greg Reply:

    Carla,

    Absolutely! Thanks for commenting.

    - Greg

    [Reply]

  8. Steve Burris permalink
    March 18, 2010

    Meaningful measurements depend upon one’s business model and it’s “most wanted response” but i sure like your insight … Marketing Science Institute shows, two thirds of return on brand investment shows up in long term performance. That doesn’t make the return any less real, just less immediate.

    Keep up the good info Greg

    [Reply]

    Greg Reply:

    Steve,

    Thanks. I’m glad you liked it.

    - Greg

    [Reply]

    Doug Garnett Reply:

    This gives me a better way, perhaps, to discuss the challenge that my more entrepreneurial clients face with agencies that only offer traditional brand approaches to them.

    Essentially, it takes long-term financial stability before a company can afford to wait for a future return on 2/3 of their spending. What I find with new & growing businesses is that agencies don’t have the business planning savvy to separate two categories:

    - The company for whom investing in a long term return is the best use of advertising money.
    - The company who must see more immediate return from their advertising dollar.

    In the first case, traditional brand advertising may be their best option. In the second case, traditional brand advertising may well put the company out of business.

    Does anyone know of research showing when/in what situations it is a higher return to wait for the long-term 2/3′s and in which situations planning for the future 2/3′s is less financially effective in the long term even if you can afford it? (Wow, that’s a convoluted question. Probably someone knows how to articulate it better than I.)

    Guess what I mean is that logically, merely because you CAN afford the investment in the long term doesn’t necessarily mean that the long term return on the money is the BEST place for that money. This seems to depend on your specific circumstances.

    …Doug

    [Reply]

    Greg Reply:

    Doug,

    I’m not sure your logic is quite right. Just because long term benefits exist doesn’t mean that you have to sit around waiting for it to work.

    For instance, you don’t expect to get all your money out of a computer the first day. You expect to use it every day for years. The long term doesn’t steal from the short tern, it augments it.

    Moreover, if speed was the only consideration, advertisers would buy only TV and no direct advertising. It’s the fastest way to build coverage and see immediate results.

    However, in the final analysis it’s how everything works together that’s important. The media multiplier is well documented and extremely powerful.

    - Greg

    [Reply]

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS

CommentLuv badge