There are many benefits to using a data-driven approach to lead selection, including higher conversion rates and improved ROI. In this second blog of our mini-series on using advanced data analytics in your acquisition strategy, we’ll discuss the data that will form the basis of your approach.
Read A Data-Driven Approach to Lead Selection: The Why, What and How for an overview on a data-driven approach to lead selection for acquisition campaigns.
What data do I need to maximise the value from my selection campaigns?
In order to develop the predictive models required to identify the leads you need to go after, you’ll need the following data:
Contact data of all prospects that were contacted (either by campaign or month). The more data you have, the more accurate the models will be. This can include:
- Dialler data for previous outbound campaigns. This will include who was contacted, when they were contacted and what the outcome of the call was.
- SMS data, including who was SMS’d, when and whether the SMS was delivered?
- Mailing data: A list of customers that were mailed and when.
- If multiple channel data is supplied, channel flags are required – this will allow for ideal channel selection.
You will also need response or sales data that includes who responded to the campaign offers and who took it up. Payment data will also be very important: monthly payment history in order to identify good and bad payers. This data should include cancellation dates, written-off flags, etc. Having a unique key across all of these datasets to be able to merge them together is of utmost importance. Often ID number is the best unique key to use and also allows you to enrich your data further using Bureau data or public domain data to help profile your leads.
How do I ensure selected lead information is a complaint?
Predictive models can be applied in two ways in order to provide you with a list of leads. The first would be to use your existing marketing lead database to refresh lead information like updating contact details, profiling them by quality, behavioural fit and the likelihood of responding to your offer. This allows your business to take your (already compliant) list and prioritise your marketing efforts to where it would be best spent.
The other possibility would be to take a POPI-compliant consumer lead list of South African leads, and finding the segments that best match your ideal customer profile. This would be a combination of the right behavioural fit and those with the highest likelihood of responding to your offer.
If I’m already applying selection criteria to my lead lists, how could I optimise my approach?
If you have historic campaign information, the first step would be to analyse the historic performance of leads, and to develop predictive models, exclusion criteria and a data-driven selection strategy. It’s always a good idea to implement new strategies on a sample size, and not to the entire environment, in order to measure the lift provided by these new strategies. This will allow you to motivate the investment in optimising your data-driven selection approach, but also enables you to understand the benefits and ensure you are making improvements and moving forward.
Remember to keep a small, random sample in your selection that you can monitor separately. This sample will represent the original, full database that you were considering for selection, prior to applying any smarts. This allows you to monitor the lift provided by the use of predictive models, but also helps build stronger historic data for future model re-development.
It’s as easy as that! If you want to find out more about using analytical models to identify quality leads, read more on Genius Leads. I will be writing another blog on some frequent questions my team and I get asked about this topic: how you would go about testing your models, how many channels are optimal and how you determine the success of your selections? Be sure to check back here soon for that post, or sign up for email alerts to receive it straight to your inbox.
