You’ve just settled into your seat at the conference table with a hot, fresh cup of coffee in your favorite mug and a stack of media performance reports that list every KPI imaginable. You’re ready to answer any question C-Suite throws at you about what your campaigns achieved last year, so you can get approval to do it again this year (with a larger budget, of course). You’re feeling great.
“What was the ROI of last year’s most successful campaign?”
For many of you, the answer is already on the tip of your tongue. You can clearly relate all your media & marketing inputs to real business outcomes without rambling on about clicks, likes, or any of the other vanity metrics that you know the C-Suite doesn’t care about. You can prove the ROI of your campaigns in your sleep. You rock.
If that sounds like you, let’s talk about how we can work on increasing the ROI you’re already tracking. We’ll skip the marketing attribution section of our pitch and get straight to work on finding a better way to tell your brand’s story.
For the rest of you, read on.
The phrase “Marketing Analytics & Attribution” may send you into a panic. It may invoke a long-forgotten disdain for the statistics classes you were forced to endure years ago, or remind you of how many hours you’ve spent troubleshooting Excel formulas.
While machine learning and statistical modeling can take any data-driven marketing strategy to the next level, there are much simpler ways to understand how your marketing efforts are impacting business outcomes, without ever pulling out a calculator.
Deciding which KPIs will define the success of a campaign sounds straightforward, but depending on how your business is organized, this can get complicated fast. Measuring the wrong KPIs can ruin a marketing team’s reputation with the C-Suite, so if your business has a complex sales cycle, sells through a distribution or dealer network or is otherwise organized in “silos”, you have to get this right from the start.
In today’s hyper-connected marketplace, it’s never been easier for marketers to be involved in touchpoints at every step of the customer journey. However, just because you can be involved during each step doesn’t mean you should take all the credit (or blame) for what happens at the bottom of the funnel. It’s critical to understand where the handoff from marketing to sales should take place when defining KPIs for your campaigns, otherwise you might be setting yourself up for failure.
For example, a company selling pet supplies online has a low-touch sales cycle and can reasonably expect their digital marketing efforts to usher buyers all the way from Awareness to Conversion. Proving ROI in this scenario is simple, since each order is related to a digital campaign.
Conversely, a heavy equipment manufacturer can expect B2B buyers to begin their research online, but these buyers aren’t likely to make the decision to purchase without engaging with the sales team first. Marketing teams that support high-touch sales cycles like this are likely to track KPIs such as white paper downloads, store visits or phone calls, since their campaigns will have a limited impact on a prospect’s decision to buy once the lead is passed to the sales team.
In order to estimate the ROI for upper-to-mid funnel campaigns, you should assign a monetary value to each KPI based on historical sales data, close rates and input from the sales team. Otherwise, your results won’t be related to business outcomes and the C-Suite won’t see the value of your work.
If you need a more exact measurement of success for these upper-to-mid funnel campaigns, you’ll want to integrate your marketing activities with your CRM database. One of the easiest ways to do this is by using UTM codes for all digital campaigns.
UTMs, CRMs, & ROI
Without UTM codes, you’re going to have a hard time measuring the success of any digital campaign. You’ve probably seen a long string of UTM codes at the end of a URL in your browser’s address bar. While these codes may look like gibberish, they provide incredibly valuable insights about your marketing efforts when configured correctly.
UTMs are passed along to your web analytics platform, allowing you to see which campaigns, channels, and ad versions are driving the most valuable behaviors on your website. These codes can also be passed along to your CRM’s database so that when a prospect converts to a customer, you can attribute the revenue generated from that customer to your marketing efforts. Over time, this will also allow you to optimize future campaigns based on the lifetime value (LTV) of your customers.
There are 5 basic UTM parameters:
- Campaign: Often, this is simply the name of the campaign. However, you can add product lines, campaign objectives, or any other categories (separated by “-“ or “_”) that make sense for your business. Just be consistent with these categories when creating UTMs in the future.
- Examples: wintersale, productname-lead-whitepapername
- Medium: The type of media being used.
- Examples: email, paidsocial, organicsocial
- Source: The specific origin of the traffic.
- Examples: facebook, linkedin, google
- Content: The content field can be used in several different ways. For banner ads, it may be used to differentiate creative versions. For emails with multiple links to the same page, it can be used to determine which CTA is driving the most conversions.
- Banner Ad Examples: summersale-v1, 25off-v2
- Email Examples: buynow, learnmore
- Term: This field can be used to track keyword performance for paid search campaigns, however if you use Google auto-tagging then you can safely ignore this field.
Here is an example URL for a campaign named Super Charge Your Social Media, running on the Google Display Network:
By using this URL, we can filter reports in our web analytics platform to see exactly how many leads came from the Super Charge Your Social Media display campaign. These UTMs also allow us to see how traffic from the campaign behaves compared to other campaigns and channels. Also, by passing this information along to our CRM we can see exactly how much revenue this campaign brings in over time. Pretty cool, right?
With a little up-front strategic planning, UTMs make tracking ROI across multiple campaigns simple. Just keep in mind that the most common pitfall when using UTMs is inconsistency, so make sure to follow best practices when establishing a protocol for using them.
What if my customers don’t buy online?
While it is certainly easier to attribute sales to marketing tactics when your customer journey is mostly digital, it’s still possible to track offline conversions.
For high-touch sales cycles, phone calls are inevitable. By using unique call tracking numbers for different audiences, mediums, and campaigns, you can uncover which tactics are the most effective at driving calls that convert. Like UTM codes, call tracking activity can be integrated with your CRM database for direct attribution.
If your product or service requires your audience to visit a physical location (think equipment demos, restaurant visits or in-person meetings) then store visit conversions can be a great tool for measuring success. By relating ad impressions & clicks to physical store visits, you can get a pretty good idea of how much foot traffic your digital campaigns are generating for your sales team.
By restricting a promotion to a single channel, you can attribute 100% of revenue from that promotion to the chosen channel. While this can limit the reach of your campaigns in the long term, it can quickly shed light on which channels are the most receptive to promotions in the short term.
Media Mix Modeling
Since we promised you wouldn’t need a calculator for this article, we won’t go into the specifics of how OLS Time Series Multiple Regression can help you predict the future and optimize your media mix for maximum ROI. Instead, just imagine being able to confidently tell the C-Suite, “If we reallocate 20% of our TV budget from DFW to Houston, we’ll increase sales by $1.1 million,” or “There will be heavy thunderstorms in Denver this weekend, we should dial back our social media campaigns by 50% since we know anything more will be wasted media spend.”
If you really want to get into the details of everything Media Mix Modeling can do that , just tell us what you want to predict, and we’ll show you how it works.