Question

Topic: Research/Metrics

Need Help Estimating Roi

Posted by lathans on 9/12/2007 at 1:15 PM ET 500 Points
I have been asked to outline a marketing plan for a company that has only ever done search engine PPC in the past. They gave me no budget to work with and requested a "recommendation" plan with associated budgeted costs. Company offers shipping services to everyone but target is small businesses. It has clients nationwide, but has never done any kind of local B2B marketing where there is great opportunity for clients specializing in small business. All traffic generated will be pointed to either the website or phone-in for quoting purposes. The company has just rebranded so included in this plan are printing costs for brochures, postcards, and business cards and company apparel. Already in the plan are Chamber of Commerce, print, YP, online banners, direct mail, sales leads, trade shows, press releases, and a sales incentive program. One last tidbit of info is that they have absolutely no tracking methods in place to measure any of this (except Google Analytics to see referer site traffic).

Here is my challenge: they want me to calculate ROI for each item listed. I can use traffic/circulation amounts where given for cost per impression where applicable, but I explained that everything in a plan has to work together to give an overall ROI, and that not all individual pieces will have a return $ amount. They want to be able to pick and choose what they want to do based on price. Any ideas about how I should approach this? Thank you!!
To continue reading this question and the solution, sign up ... it's free!
SUBSCRIBE
Or, SIGN IN

RESPONSES

  • Posted by koen.h.pauwels on 9/12/2007 at 2:17 PM Accepted
    I feel you - just got asked a similar question by a foreign furniture maker with a focus on small businesses: they face data overload with Google Analytics for site traffic, but have no idea about the effectiveness of their catalogues and print advertising...

    My suggestion is to start with a simple 'dashboard': what is the objective (e.g. monthly profits) and where do these come from (e.g. leads, lead conversion, profit per converted lead)? This will give you an idea where each marketing activity fits it: it generate leads and/or converts leads and certainly has a cost attached (which reduces profit per lead).

    Next you can include some synergies: the truth is probably in between management's desire to pick and choose based on highest individual ROI and your claim that 'everything' has to work together. For instance, press releases may be vital to generating leads, while sales incentives may be vital to converting them.

    Finally, come up with your best guesstimate of the impact of each marketing action on your objectives. Given the lack of tracking data, you can either
    (1) use an average estimate based on industry analysists or marketing academic literature
    (2) ask management to come up with their best estimate of the impact of e.g. direct mail on leads for (a) the current level, (b) a 50% reduced level, (c) a 50% increase and (d) the saturation level (what number of leads would the maximum feasible direct mail produce?). This allows you to come up with a direct mail-leads response curve, which you can later improve with actual data (you can always contact me for details).
  • Posted by Dawson on 9/13/2007 at 3:24 AM Member
    If I've read correctly, you need to account for setup (fixed) costs as well as variable. That could "throw a spanner in the works" unless you are careful.

    This organisation sounds like one where you could apply a "funnel" approach. i.e. determine which channels would be targeted at generating awareness of the proposition, which are used to follow up on the awareness to generate sales contacts and which are used for closing. This is a quite basic concept and is not necessarily suitable for more advanced marketing organisations but it is a good starting point (google AIDA model for more).

    At this stage, i would suggest that it is not possible to estimate ROI accurately (esp just using something like cost / impression which ignores the different nature of interactions in each channel). For me, one of the best methodologies is that recommended above by Keon. See a book "Marketing that Works" by Lodish for more on this - it's certainly the right book for your situation.

    A couple more points:
    1) Make sure that they are collecting data on all the activities that they finally execute so that they can improve still further next time
    2) Try to add some testing into the campaigns - make sure that they try slightly different approaches in different markets - execute the same strategy in all markets and they won't have a great idea what to do differently next time - I always recommend that my clients put around 10% of their budgets into so called experiments. Given that these experiments are unlikely to prove totally ineffective, the most that is at risk is a few % of their overall budget but the upsides in knowledge are huge.
BACK TO TOP

Post a Comment

Related Discussions

Community Info

Top 25 Experts
(Research/Metrics)