How do you know whether your lead-generation program is working and delivering a good ROI for the company?
You may be doing some lead tracking to understand conversion rates and customer profitability, which is great. But the sales team will inevitably let Marketing know that (1) Marketing was just a small step in closing the sale so the sales team deserves the credit; (2) the sales team would have found and closed those leads anyway, so there is no incremental value; or (3) the leads are fine, but there is just never enough.
We need reliable measurements to both prove and improve our marketing effectiveness.
The first two parts of this article series outlined what is necessary to improve lead quality (covered in Part 1) and how to improve sales alignment for better lead transition, tracking, and sales support that increases the close rates of those leads (covered in Part 2). We'll now look at the measurements challenges for lead generation as we provide insight into the most-common questions lead-generation marketers raise with respect to measurements.
Lead-Generation Measurements
Measurements around lead generation are made easier by the fact that once a contact is generated, that contact's interactions and outcomes can be tracked (systems and operational issues aside). But there are also challenges unique to lead generation that require a closer look at what it takes to get reliable measurements.
Let's address the most-common measurement questions, starting with the basic and moving to the more advanced. As you look through these, remember that there are four primary categories of marketing measurement that we can use to assess the incremental impact of marketing: pre- and post-tracking, or trending; market testing; modeling; and surveys.
Lead tracking: How do you match lead contacts to buyer contacts when different people from the same company are involved in the buying process?
This match-back challenge can occur when the basic measurement technique is to track from the marketing tactic that generated the lead to the closed sale.
In some companies, the sales system does not contain enough detail to link the marketing contact, who may be a recommender or decision-maker, with the buyer, who may be from procurement or another person on the purchasing "committee." You have a contact who is engaged with Marketing and passed into the sales organization as a lead, but your only tracking option is to go to a purchase database and try to make a match.
If you choose to match by contact name, you will understate Marketing's impact since you know many of the buyers listed in the financial system do not represent the full set of decision-makers and recommenders you influenced. But if you match by just company name (as I have seen companies do), you will overstate your marketing impact, especially if you have a reasonable number of large clients with multiple buying groups.
A good compromise for those companies limited to some form of match-back is to match by company name and location. It is still subject to error but gets closer than either extreme. An alternative is to use periodic surveys of your marketing contacts to determine the sales close rate. The survey can also be used as a checkpoint or calibration of your match-back process.
Multitouch marketing assessment: Which marketing tactic gets credit for the lead?
Typically, a prospect is touched by multiple marketing contacts prior to qualifying as a lead that is put into the sales cycle. In most lead measurements, the last marketing contact gets credited for generating the lead. When most of your marketing efforts are response-oriented, this approach can work fine. Every marketing contact is generating a level of interest, and it took that last marketing contact to get a certain set of prospects to become leads.
This tracking approach, however, will undervalue marketing initiatives that are better at generating interest and boosting the response for other marketing initiatives, and it can also overvalue marketing touchpoints that pick up leads that would have otherwise come in through other contacts.
The solution is not to allocate credit to multiple contacts but to set up measurements that isolate the incremental impact of specific marketing initiatives. This can be done through market tests when assessing specific tactics or through modeling when assessing multiple tactics and the effects of their interaction. Modeling can establish correlations to identify which touchpoints are most influential in generating leads or closed sales.
(For a more detailed explanation of options and issues, see "Measuring the Impact of Multi-Touch Marketing.")
Structured measures: What incremental impact has marketing generated?
The best measure to show the true incremental impact of marketing is market testing using test and control groups.
Generally, I am not an advocate of complete "no-contact" control groups that require eliminating marketing contacts for the purpose of demonstrating the incremental value of marketing, because we are typically not trying to prove that no marketing is a viable alternative. Although there are some instances where complete no-contact control groups make sense, more often we will set up control groups that have "business as usual" marketing and sales contacts, with the exception of a select marketing initiative(s) that is(are) are withheld. Market testing is also very effective to build the case for adding new marketing initiatives.
For lead-generation marketing, there are four market-test measurements that are particularly worthwhile in helping to address your more critical marketing decisions:
- Test a marketing tactic or integrated set of tactics to determine the incremental lift on leads generated
For direct marketing, your approach is to withhold specific direct-mail or email communications from a random sample control group to determine the incremental lift in generating leads. If the control group nets the same proportion of qualified leads as the test group, it is likely that your other lead-generation activities are capturing the opportunities out there. Any lift in the test group over the control group is attributed to your marketing.
Mass-marketing tests require a geographic split of comparable markets or regions, matched through careful sales-trend analysis. The test markets receive the incremental marketing to determine the incremental lift in quality leads. - Test the incremental value of lead-generation marketing to the sales organization in terms of incremental sales conversions and average customer value
The same testing approach outlined above can be used to assess the impact on closed-sales rates or average customer value (in addition to lift in leads).
This comparison is very appropriate when the sales organization is trying to make the case that Marketing is not adding value to the sales process or that the leads generated would have been identified through the sales organization anyway. The analysis should determine whether the test group receiving marketing contacts experiences more leads, a higher net close rate, or higher average customer value per sale than those receiving no or less marketing contacts. This may require withholding marketing for the select accounts for a time, and it is justified when there is an established belief that less marketing is needed.
If you set up your tracking to monitor performance changes during the test, you may be able to detect a decline in performance early; you can then resume marketing instead of allowing the control group to hurt sales results.
Remember that measurement success is dependent on having statistically significant sample sizes to minimize the normal variance that might occur if you created the two groups and treated both the same. There are analytic processes that can help set up the experimental design. Also be sure to eliminate possible biases. For example, you are better off comparing a portion of each sales rep's accounts than splitting the sample by account rep, because rep performance may vary. - Test and assess new marketing initiatives
The addition of new marketing initiatives can easily fit into a market-test structure. If the marketing initiative is planned to launch nationally and there is the flexibility to do so, a measure of the initial impact can be obtained through a phased-in launch that occurs regionally.
Control groups are set up within the regions launching last. The primary objective of the measurement should be improving the new marketing initiatives (as opposed to continuing vs. canceling). So the analysis should look deeper than bottom-line results, for insights such as identifying that the marketing is working with specific target segments or where in the customer funnel the performance is stronger or weaker. - Test extending lead-generation programs beyond just lead volume to determine the potential for nurturing programs or sales-support marketing
As we addressed in the first two articles of this series, often there are opportunities to improve lead-generation effectiveness not by increasing lead quantities but by either increasing the conversion rates within the sales pipeline or by nurturing leads that were generated but not yet qualified. The ROI analysis can be used to show the profit potential of these initiatives. Market testing is used to develop effective programs on a small scale with minimal budget and then to build the case for additional funding.
Collaborative measures: How do we measure leads passed to external channel partners and resellers?
When marketing organizations are passing leads to external channel partners, there is often a huge gap in tracking that ultimately leads to the challenge identified in our first question, above, on accurately matching contacts to buyers.
There are three measurement approaches designed specifically for third-party relationships that are worth consideration:
- Set requirements for channel partners to report lead outcomes on a regular basis. Some companies are very successful with this approach, typically because they have good quality leads that the channel partner is willing to work for. In other cases, the channel partner views the customer relationship as his or her own and will not disclose such details. The goal is to find the win-win approach that improves marketing effectiveness and is beneficial to both parties.
- Set up a joint market test. This works well when testing new initiatives. If the channel partner understands that your company needs to demonstrate the value of the specific lead-generation investment to benefit both organizations, the channel partner may provide enough quality feedback for an assessment (maybe excluding customer details).
- Use survey research. Survey your leads periodically to measure their final purchase decision. Surveys are also a good way to get detailed information on your leads' funnel progression and leakage, and insight into their competitive considerations.
Recap: Assessing Lead-Gen Effectiveness
Briefly, these are how different methodologies can be applied to assess lead-generation-marketing effectiveness:
- Pre- and post-tracking is an easy, low-cost measurement that provides insight into relative effectiveness based on crediting just the final lead source. It is not entirely reliable but offers directional information that can improve performance.
- Surveys can help close data and tracking gaps. This is also a good methodology when assessing marketing programs where sales are closed through channel partners or resellers who do not report sell-through data back to your company.
- Market testing is very effective for isolating the impact of a specific set of tactics or the introduction of new campaigns. This also works well to identify the incremental lift of brand marketing on lead-generation effectiveness.
- Modeling is the best solution to assess multiple marketing channels. More-advanced modeling techniques can be used to assess the impact of nonmarketing factors on your marketing effectiveness, such as sales coverage, competitive activity, or market conditions.
* * *
In the next and final article in this series, we'll cover lead-generation metrics and how they are used to manage your overall performance.
