Real-World Education for Modern Marketers

Join Over 606,000 Marketing Professionals

Start here!
N E X T
Text:  A A

How to Measure Those Three Little Words: Return on Investment

by   |    |  18,486 views

In this article, you'll learn...

  • The basics of calculating marketing ROI
  • A more nuanced and comprehensive way to calculate ROI

Sure, marketers like to hear those familiar three little words: "I love you." But, let's face it, no three words are more beautiful and pulse-racing to hear while on the job than "return on investment" (ROI).

Marketers are on the hook. Measuring ROI on each marketing activity is critical to proving our effectiveness and competing for scarce resources. Evidence that supports that notion: Nearly two-thirds of surveyed CMOs (63%) think ROI will be the primary measure of their effectiveness by 2015.

But what was profoundly revealing was that 56% of surveyed CMOs said they feel inadequately prepared to manage ROI. That means the pressure is on to calculate ROI in ways that yield accurate, supportable numbers, particularly pertaining to marketing expenditures.

CMOs who understand financial and statistical analysis are putting themselves on a level playing field with their brethren in other corporate departments, because those CMOs are using the same measuring stick to prove effectiveness.

Measuring ROI effectively will help marketers add value to their organizations and attract their fair share of resources.


Basic ROI Calculation

The math is basic, but the components aren't always easy to figure out. Any solid ROI calculation should measure revenue generated by the marketing campaign, profit margin on the items sold, and marketing-related expenses. Yes, profit margin can be difficult for organizations to quantify because it incorporates the cost of operations, but it's a necessary ingredient to calculate ROI accurately.

Keep in mind that to be able to quantify revenue generated "by the marketing campaign," those campaigns need to have been designed to capture responses in the form of revenue from sales.

In its simplest form, the calculation looks like this:

ROI = Campaign Revenue X Profit Margin / Cost of Campaign

That simple ROI calculation is a solid metric for getting a quick "temperature check" on campaign performance. The basic calculation works for basic comparisons, such as "Campaign 1 is doing better than Campaign 2" or "we improved this month over last month."

That formula is also useful for measuring pilot marketing campaigns, since those efforts are often launched with one-time investment costs or temporary extraordinary expenses that wouldn't apply once project efficiencies are captured in a full-blown campaign.

Though that basic ROI calculation is a good starting point and it keeps things reasonably simple, it lacks several key enhancements. The calculation sometimes doesn't factor in all of the costs associated with the campaign, and it doesn't consider the impact of control groups. Both of those are required to evaluate the actual revenue generated by a campaign.

The Advanced Version

Marketing will need to take a more advanced approach to ROI calculation if it wants to play in the same ballpark as the other departments in your company—and if it wants to be evaluated by similar financial measures.

Advanced ROI calculations should include all of the costs associated with having a marketing department: salaries, benefits, office space, computers, software, Marketing's share of the bills (heat, electricity, etc.), plus all of the direct campaign costs. All of those costs added together would give you the "cost of campaign" figure in the formula.

Also important in advanced approaches to calculating ROI is to factor in dollars generated and purchases generated from both the target group and a control group. Doing so will help you answer important questions, such as whether the average purchase dollar amount was higher from the target group or the control group... or whether the target group purchased at a higher rate than the control group.

If the average number of purchases generated from the control group and the target group is the same, consider the difference in the average dollars generated. Similarly, if the average dollar amount spent is the same between both groups, the most important factor is the difference in average number of purchases generated.

First, determine the number of sales from the target group that would have taken place without the marketing campaign by measuring the average number of purchases generated in the control group. Now, multiply that number by the average purchase amount ($) from the control group. Subtract that number from the total target group revenue to determine accurate campaign revenue. The equation looks like this:

Campaign Revenue = Total Target Group Revenue – (Average Amount of Control Group Purchases X Number of Transactions That Would Have Occurred Anyway)

That calculation factors out the transactions that would have occurred without the new marketing campaign, leaving the incremental revenue gain from the higher average transaction amount of the target group. Taken further, such calculations could enable a marketer to compute statistical tests of significance for the target and control group differences.

* * *

Hopefully, this article will help you apply the tools to excavate the important numbers that improve marketing ROI, and, coincidentally, reveal ways that you can get closer to the customer.


Join over 606,000 marketing professionals, and gain access to thousands of marketing resources! Don't worry ... it's FREE!

WANT TO READ MORE?
SIGN UP TODAY ... IT'S FREE!

We will never sell or rent your email address to anyone. We value your privacy. (We hate spam as much as you do.) See our privacy policy.

Sign in with one of your preferred accounts below:

Loading...
Jim Bergeson is president and CEO of Bridgz Marketing Group in Minneapolis, a BI WORLDWIDE company and member of ICOM, the 50-nation network of independent marketing communications organizations.

Rate this  

Overall rating

  • This has a 5 star rating
  • This has a 5 star rating
  • This has a 5 star rating
  • This has a 5 star rating
  • This has a 5 star rating
2 rating(s)

Add a Comment

Comments

  • by SergeKuznetsovFinUniversity Thu Mar 15, 2012 via web

    Two major problems remain unsolved:
    1/ how to separate/identify marketing investments and marketing costs:
    2/ how to properly calculate mRADR (marketing risk-adjusted discounting rate).

  • by Peter Green Thu Mar 15, 2012 via web

    Look, I understand that people in the ROI business want to make a living, and don't get me wrong I do think we need to measure marketing campaigns to understand better the effects. But this article is giving such a rosy picture of easiness, please... There are a couple of fundamental difficulties in measuring marketing campaigns, and one of them is baseline definition. To measure increase, you have to have a starting point that is relevant. In FMCG which is my area, this can be quite tricky. Promotions come and go, from your own brand and competition, and the shorter time period 'before campaign' you use, the more liable you are to make the wrong conclusion about a certain campaigns’ effect. And the longer timer period you choose other factors get into play, like seasons starts to play a role, etc etc. I hope you get the idea here. In the other end, you have the definition of costs, which is far from obvious. Do you include everything, from branding activities to POS material to price cuts to retailer incentives? It sure does make a difference in the evaluation. You need to be sure whether it's relevant to include those costs. Take retailer incentives. It's probable that you will have to pay them whether you do marketing campaigns or not, because that's part of the retailers way of working. And you will have sales only by being on-shelf, so there is a logic to it. Would you include listing fees when it’s about a new product launch. Normally you only pay them year 1, so should you periodicize them over the assumed lifetime of the product? I can go on for a while... but I will end with the example of the marketing department costs, down to dedicated electricity bills. Now that's one of the more ridiculous ideas I've seen in years. To begin with, I think it’s safe to assume that these costs are the same whatever type and quality of campaign that is produced. So it’s neutral in a benchmarking exercise. But it also kind of suggests that you could do without the marketing department, same as the marketing campaign cost. Of course you could, especially short term. You can also do without any advertising short term (it increases profit tremendously! - yet few brands stop advertising forever) or do without controllers short term, or HR people, or most types of people in the company, even CEOs. Long term, well I would suppose that if a company has a marketing department, it's because they need it. If not, don't calculate ROI on their salaries, just fire them! Thank you

  • by Michael Webster Thu Mar 15, 2012 via web

    @Peter;

    I think that the author only means to highlight how in principle you could measure the effectiveness of changing Marketing campaign A, by adding/subtracting cost x, to Marketing campaign B- holding everything else the same.

  • by Peter Green Thu Mar 15, 2012 via web

    @Michael,
    You may be right - but that's exactly my point. When "in principle" meets reality, it isn't that easy. Or rather, if you make it that easy, the results risk to lead to the wrong conclusions. Which I don't think is helpful.

  • by Cindy Erwin Thu Mar 15, 2012 via web

    What about measuring ROI for an at-need service business that doesn't focus on campaigns but rather on constant top of mind awareness? It could literally be years before someone needs our services--how do you tie that back to any one particular marketing or advertising effort (and its associated costs?) If anyone has the answer to that, I could make my boss very happy!

  • by Tim Altier Fri Mar 16, 2012 via web

    Hello,
    I am responding on behalf of Jim Bergeson, who is currently out of the country.
    Peter,
    Thank you for your comment.
    We didn’t mean to imply that calculating ROI or measurement in general is simple. We know it’s difficult. Our experience with a number of Fortune 1000 companies is that they are not tracking return on marketing dollars in their campaigns, primarily because it is difficult. For example, it took two years for us to convince a Fortune 100 client to use a control group so we could report the on the performance of the campaigns we were executing for them.
    I don’t think attributing overhead costs to departments is ridiculous at all – I live it every day, or literally every month end when my department P&L is published. I run the Analytics team at Brigdz Marketing Group. Of course my department is responsible for the easily trackable expenses like software licenses, salaries and computers; but also a portion of overhead costs that includes rent, utilities, phones, etc. are charged to my department. However, we are an agency and not a Fortune 100 company. I suspect a large corporation would roll the department level operating costs into the company’s cost of doing business. I think it is a matter of experience and perspective.
    There are companies who choose to hire IBM Global Services rather than have an IT department or Razorfish rather than have a web site team or Accenture rather than have an accounting department. There may be good reasons for companies to make such choices and likewise the choice to have, or not to have, a marketing department. The article just suggests assessing the whole picture to make an informed decision and provides a few principles as a starting point.

  • by Casey Carey Fri Mar 16, 2012 via web

    An interesting post and comment thread. This continues to be an embarrassing part of the CMO and marketing leadership profession, even with much more accountability being placed on the role. A couple of observations:

    1. Don't let perfection be the enemy of good. Agree with the CFO and exec team as to standard margin rates and allocations to be used for the calculations. Trying to get to the exact numbers is a fool's errand - this is marketing, not accounting.

    2. I am glad to see the inclusion of profit margin (Cost of Goods or Cost of Revenue) as part of the equation. Unfortunately, most of our peers do not understand this and think "R" stands for Revenue, not Return. This is especially true in the online world.

    3. I am disappointed to see the author failed to include campaign expense in the numerator. The equation is ROI = ((Campaign Revenue X Gross Profit Margin) - Cost of Campaign) / Cost of Campaign. Return must include the cost of the campaign.

    4. Hold out tests are awesome, though hardly anyone does them. The important idea here is the notion of incrementality, i.e. if you did absolutely no marketing, how much business would you have? And, no the answers is not zero. While an impossible question to answer, as marketing leaders we need to look at our contribution to the business with this as the starting point. Once you do, ROI and creating value becomes much more clear is the main idea that drives what you do and why you do it.

    Read more: http://www.marketingprofs.com/articles/2012/7323/how-to-measure-those-three... -

  • by Tim Altier Fri Mar 16, 2012 via web

    Casey, thank you for your comments. All points well taken. Some thoughts on item 3. I suggest that your ROI equation is the same as ours except your calculation is centered about 0 (zero) and ours is centered about 1 or 100%. Let's say a campaign costs exactly the amount it returned (i.e. breakeven). Your calculation yields 0, ours 1. Second case - the campaign generates $0 (zero) revenue. your calculation yields -1, ours yields 0. What I like about our form of the calculation is that it provides some very nice ratios. If the campaign returns 50 cents on the dollar, our form yields .5 or 50%. If I breakeven on the campaign, you might say I am made whole again which is expressed nicely as 100%. If I double my money, i.e. the campaign returns $2 for every dollar invested, our form of the equation yields 200%, triple yields 300%, etc. None of this is worth bickering over though; it's simply a matter of preference and style. And frankly, we use the form of the equation that we do because it is what most of our largest clients use. :)

    Thanks again for the thought provoking discussion!

    (I am responding on behalf of Jim Bergeson, who is currently out of the country.)

  • by RICHARD EFFAH ANNAN Sun Mar 18, 2012 via web

    I've read the article and all the comments and they are all good, I must say. We're all learning! There's ONE notable point I want to say:

    We have numerous methods in calculating ROI, each having its pros and cons. Therefore, each method one chooses would definitely have an accusing finger pointed at it. It doesn't however mean that the method chosen is bad.

    What is more important should be whether or not the method chosen helped achieve its intended results for an informed decision and action to be taken. Remember, a method chosen for one dept/company may not equally be good for another dept/company.

    We all try to achieve perfection but it doesn't also mean good is 'bad'. The author tries to move from 'A' to 'B' but it doesn't mean there couldn't be 'C' or 'D'. and if you moved from 'B' to 'C', it doesn't mean 'B' is bad. Again, criticising method 'A/B' backlashes us to move to'C/D'. In this wise I compliment the author and the critics.

    We should also note that even the Accounting Profs give rooms for errors, ommissions, miscellaneous, inaccuracies, etc,etc in their methods of calculating.

    Well done and thank you all for this 'academic-professional' discussion.

  • by RICHARD EFFAH ANNAN Sun Mar 18, 2012 via web

    Adding to my comment above, in sum, let's move forward, let's be flexible and let's choose the method which best fits. Thank you!

  • by Thorsten Strauss Sat Mar 24, 2012 via web


    @Jim

    Sorry to say but there is another issue with you article:

    QUOTE : “Advanced ROI calculations should include all of the costs associated with having a marketing department: salaries, benefits, office space, computers, software, Marketing's share of the bills (heat, electricity, etc.), plus all of the direct campaign costs. All of those costs added together would give you the "cost of campaign" figure in the formula.”

    That is not totally correct. Because you still occur some of these costs, if you were not to run the campaign. That means there is a certain amount of fixed or semi fixed costs usually called overhead such as : office space, heat, electric and variable costs : software, computers, salaries,…

    A campaign should only be allocated the full variable plus an appropriate of share of the fixed costs. That requires activity based costing and tight project cost / time allocation control.

    From my own experience trying to implement a project measurement tool in a design department I have to say that there is a cultural barrier between creative people and exact cost allocation measurement processes. What works well among (software) engineers does not apply in design. In theory is sounds nice and easy, in practice people work parallel, iterative and collaborative. Good luck with accuracy here.

    Concluding, the challenge in marketing ROI is not how to calculate it, but being able to measure costs accurately (the easier part, but still very challenging) and revenue or profit (which can be the near impossible part).

    And as final thought, do not forget to measure the loss of efficiency and impact of the measurement process itself as part of overhead. I found that it can be more disturbing and costly to efficiency to measure costs on project basis than letting cost measurement go and focusing on measuring campaign results and doing a good job.

    Don’t get me wrong, I am a big fan and advocate of metrics in marketing but as somebody said before, you are making it sound a bit to easy for practitioners.

    Best regards,
    Thorsten Strauss

  • by Jim Bergeson Tue Mar 27, 2012 via web

    @Thorsten

    There are as many ways to do cost accounting and measurement as there are companies out there - as you and the other commenters point out. Thanks for your comments.

    Jim

  • by Bianca Sun Mar 31, 2013 via web

    I really like this article and enjoyed reading through the comments. Good quality information and discussion. I do personally really like the quick temperature gauge as you put it because it's simple and fast, although my habit is to use the variation of the formula that would mean the breakeven ratio is zero. This is just habit. I quite like your point about your simpler ratio providing the ratio of 100% based on the breakeven scenario.

    As a Mar Comms Mgr myself I find it super interesting to look at the various alternatives and perspectives. I particularly like Cindy's point which I interpret to really be about "time". The time horizon over which you could calculate ROI can be challenging and I would suggest trying various time horizons, as that can help to identify trends and fluctuations.

    The only thing I would add to the discussion is that as the marketer for your business you can also argue for a focus on "return" to also mean other metrics, not just revenue and profits. While that is the end goal so to speak, we know that consumer behaviour is a funnel, a series of events (awareness, choice, evaluation, purchase, post purchase evaluation etc) so having some "milestone" metrics I think is also a key value in identifying.

MarketingProfs uses single
sign-on with Facebook, Twitter, Google and others to make subscribing and signing in easier for you. That's it, and nothing more! Rest assured that MarketingProfs: Your data is secure with MarketingProfs SocialSafe!