Know-How Exchange : Advertising/PR

Hi There,

I am not sure if I understand your question fully but I'll give it a try.

(i) There are a number of tools you can use to calculate the Return on Investment of a promotion by tracking that promotion through to sales. These methods will differ depending on what type of BTL marketing activity you have used.
- Website - Web Analytics programs offer an automatically calculated ROI for any pay per click adverts, links clicked through and other web marketing activities through to the sales of a product via your website. Examples of web analytics programs include:
- Web Trends
- Webside Story
- Omniture
- Click Trends
- Deep Metrix

- Email - Email Analytics programs (when appropriate code is placed on the shopping cart pages on your website) can track the effectiveness of your email campaigns through to sale and give you a conversion rate and ROI based on the number of emails sent out and clicks driven through to your site from the emails.
Examples of these include:
- EmailLabs
- ExactTarget
- Campaign Master
- Blue Hornet
- Cheetah Mail

- Direct Mail - A good CRM system should be enabled to have campaign tracking set up. You set up a name and forecast response rate for the campaign and make sure you have an appropriate call to action. Let's say for example the call to action was "Call 1800 QUICK and quote the number PROMO315 to receive your special promotional price on product XYZ. Communicate with your customer service and sales staff about the promotion and then when a lead or customer calls regarding the promotion a Sales Opportunity can be created in the CRM system with a sales value. As those sales opportunities are won (or lost) the system will calculate the ROI for the campaign.

If you do not have this type of CRM functionality, a simple spreadsheet can sometimes do the trick. Start with a list of prospects or inactive customer who have not purchased for over 12 months. As long as no concurrent campaigns are running, any sales over the next 6 months can tentatively be attributed to the campaign.

Tradeshow - Many stradeshow stands these days have "Stand Trackers". Electronic devices which scan pre registered customer smart tags and then store that information in a database for you, along with separate characteristics your sales team may apply to the lead (eg: hot prospect, or send follow up literature...etc). You can track responses to this in the same way as you track responses to a direct mail campaign with your CRM system.



(ii) To determine the life of a BTL you can try something like determining the average LIFE VALUE of a customer. Do this by separating out all currently inactive customers (customers who have dealt with your business before but no longer purchase from you). Then work out how long, on average they purchased from your company, and how much, on average, they purchased. You can then track the campaign for say, 3 to 6 months after the campaign has initially been launched, and then after that period, apply the customer life value estimate to all intial purchases by the customer. eg: If you found the average life of your customer to be 3 years and average purchase over that time to be $4000 then you can tentatively apply that value to customers who have purchased as a result of your campaign. Over time you may like to go back and check this, but it at least gives you an indication and also allows you to present good results to management staff members,

Hope this helps. Drop me an email if you need extra clarification,

Michelle Bourke

Hi,

There is no "generic" approach except to say "model". This, of course, presumes that the BTL activities are being conducted in parallel with the ATL activities, but not consistently across all markets/time.

For some products, the operationalization of the dependent variable is sales for others it's visits and for others it's inquiries. Figure out how to count what your client is trying to accomplish and you are half way home. The next question is trying to figure out the time frame for each case in your data set. The ATL activities are likely market-wide, but the BTL may be store specific. You may need to make each case a trade-area week in order to correctly account for localized BTL.

Once you have your cases figured out, you need to choose the form of your dependent variable. If you get granular with your geography, you'll probably find better results using logistic regression on the question of whether any sales/visits/inquiries happened at all during this week. If your BTL activities are big enough that you feel comfortable using a whole market, then you could use # of sales/visits/inquiries with an OLS regression.

When you've got your data set put together and run your regressions, your coefficients tell you whether or not the BTL activities had any incremental impact on the dependent variable (sales/visits/inquiries/etc).

I have taken these inputs from: https://forum.researchinfo.com/showthread.php?t=1285

Hope this will help.

Track all leads and enquiries back to their source by asking the enquirer or purchaser where they heard of you from. It should be an inherent part of the dialogue on the sales or order desk and should be a key metric for your CRM system.

Track all leads or enquiries through the sales process to determine if they metamorphose into quotations and then orders.

Ask all enquirers or purchasers for referrals so that you can extend the fruits of your BTL activities. Their willingness or lack of it is an indicator of the strewngth of commitment you have engendered.

Prior to BTL activities, get the sales people to predict their sales forecasts according to a uniform model (That’s one where everyone invents their figures to the same formula!)

Do the same utilising statistical techniques based on your historic sales data. Be sure to include the product or product group as a separate identifiable item in order to view any departure from trend. This may be ascribed to your BTL activities and the spells cast by the fairies at the bottom of the garden. Splitting out product groups will also show if there is cannibalisation of other product ranges due to sales activity being directed to a promoted product or finite customer spending power being used on the product they know most about this month.

What on earth does “the operationalization of the dependent variable is sales” mean? It was a term used by Greg Dawson when analysing the impact of technical maturity on financial IT systems and as far as I know he invented the term. It’s since been used to look at the “Thing that matters” in military matters and US Government economics where it sounds right up George W Bush’s vocabularistic (I know, that word doesn’t exist either) streak.. I guess here it means the parameter against which you judge success; In which case I agree!

Hears to you actualisationally enleveraging your advantagalisation gained from the meta-insightefulabilityness of our analiticasations.


Come back James Joyce: Ulysses is a push over in comparison to the language of marketing.

Regards


Steve Alker
Unimax Solutions