“Drowning in data” is an apt summarization of the past few years for most modern marketers.
2015 began with a bang – or more accurately an explosion – of more marketing channels, media, and metrics. With many businesses increasing their marketing budget as a result of either the economy slowly nursing itself back to health – and owners realizing the value of good content marketing – the pressure is on to find out what those investment dollars are actually buying. In this article Ultra Graphics will throw you a life raft, teach you how to rise above the ever-expanding sea of data, what to track and when, and how to interpret those numbers to spot trends and learn if your campaign was successful or not.
Check out our introduction to in part #1 of the article here
Step 1: Set goals and decide what you want to track.
Before getting too far into the campaign, you need to stop and take some time to set goals that would make you consider the campaign successful. It can be as simple as “sell X product,” or more complex as “sell X more product than last year, gain 10% email shares, and 25 facebook likes.” To decide what you want to track, use your goals as a springboard and imagine what numbers you would need in order to reach those goals. Of course you would need sales numbers, but in addition you might want to track email shares, coupons redeemed (for a coupon based campaign), total sales per recipient, total sales per recipient for the past year, landing page visits, campaign specific phone calls, and of course verbal in-person reports. Keep in mind, however, that the return is not always a dollar value. The more complicated (but more fun) conclusions to come to are when a marketer is trying to figure out if the campaign was worth it even if there was no revenue whatsoever.
Step 2: Determine the initial cost.
For all portions of the campaign, determine the cost to setup, design, produce, and promote the campaign. Already you can start compiling some data and making some assumptions and predictions that will support your goals. If you have a list of recipients, you can then calculate the cost per recipient. Remember, marketing campaigns are almost always long term, and there might be changes to both the cost and the return. You won’t know the full benefit of all the pieces until all the data is in, and that can be a very long time. We also learn that if we don’t gather the right data, we will have no idea if the return coming in is in response to the campaign – which leads us to step 3…
Step 3: Use the right tools to gather the data.
What can be the most intimidating step is knowing where and how to collect all the necessary data. When taken as a whole, it seems daunting when a marketer imagines getting incessantly and hopelessly overcome by incoming numbers and metrics as if they’re facing down a twenty foot wave. The trick is that when collecting, organization should be the priority with the different streams of data. Depending on the channel/media there are many separate tools out there to track visits, shares, likes, views, and participation. Google analytics can track page views and events such as form entries and external links. Facebook insights tracks likes, shares, link clicks and other interactions. Personalized landing page services such as purlem.com have their own reporting built in. Email marketing services such as Mailchimp have robust reporting. There are services available that allow you to create marketing phone numbers that come with call tracking. The methods to collect are aplenty. Much of the reporting might also rely on the human element, asking questions like “how did you hear about us? What brought you in today?”, etc… and don’t be afraid to ask questions after the campaign in the form of surveys, or feedback forms.
Step 4: Pontificate.
Marketing ROI: an example
Multitouch campaign for product that includes a direct mail piece, multiple emails, and PURL (a personalized online landing page).
To Track Each Piece:
Make sure to have a call to action on the mail piece – in this example visiting the PURL would be the action but you could also add a trackable phone number. Use an email service like Mailchimp that tracks opens and clicks as well as landing page visits with Google Analytics. Use a PURL service like purlem that tracks visits and interactions such as filling out a survey.
To Determine ROI:
Determine the cost of each piece, and don’t forget to also account for the value of your time! Determine product revenue as a result of the campaign (This is the tricky step, you have to be able to know why and when your customers bought!)
(Revenue – Cost) / Revenue = ROI
This is the big one – now you have piles and piles of data. How do you make sense of it? The challenge is to try to make grand conclusions based on tiny details. Start with asking yourself the simple questions, such as “did this campaign bring more sales?” With the data in hand, how do you answer that question? If your campaign had a postcard with a coupon, you can add up the coupons redeemed, their value, and the value of the total sale. Then you can calculate the revenue minus the coupon value. If you gathered year over year data for your recipients, you can calculate the difference they spent now vs. last year, and see if the campaign influenced their buying habits. If you can go back two, three, five years, then you can get a better idea of the trend. If the sales numbers for your recipients stays level in the past but jump during the timeframe of your campaign, you can assume it had a positive effect. If you were diligent about collecting in person reports from your recipients, you would have anecdotal evidence that the campaign influenced them. All the little details and correlations are the data gold that will lead you to make educated guesses about the effect your campaign had on the customer.
Another way to think about the campaign is how it affected a single person, not an entire group. Using the data you gathered, can you follow a single person on your recipient list from initial contact all the way to the sale? Say the target customer received a postcard with a personalized landing page (with a survey), an email reminder, and a coupon. You could track when the customer visited the landing page, when they submitted a survey, if they opened the email, and if they used the coupon. If you are diligent about collecting data during and after the campaign, you could learn why the customer chose to participate, and what brought them to the decision to make a sale through a feedback form, or the simple question “what brought you in today?” Depending on the answer, you might be able to figure out what worked in the campaign and what didn’t. Maybe the email was unnecessary, or the coupon offer was too little for the effort, etc…
Now take that concept of tracking a single person and multiply it by your whole list. Sprinkle in data that you collected such as sex, age, income, interests, etc… (this is where it’s important to decide what you want to track at the front end). Maybe your campaign worked great on males age 26-38, and completely bombed for all females. With what now must be a MASSIVE spreadsheet, you can sort data, make calculations, and learn. The failures are the most valuable data – because the next time a similar campaign is created, maybe the target demographic is a little more dialed in, or the copy and imagery is adjusted to appeal to a wider audience.
If you’ve finished reading and are staring dumbfounded at this sentence, congratulations you’re a marketer. Marketers have to be part data analyst, part soothsayer, and part tech geek.
The recurring theme is fourfold:
- Know what you want and what you expect to get before starting.
- Data is king, gold, the boss, and whatever other analogy you can think up.
- The return is not always going to be about profits or revenue.
- Most campaigns are a ton of work, and are almost always long term.