“I do precision guesswork, based on unreliable data, provided by those with questionable knowledge to meet or exceed formidable expectations.” -Sound familiar? In the today’s B2B sales world, winning a new deal doesn’t happen without engaging conversations. It…
A How-To for Marketers: Building a Clean Marketing Database
Databases • April 12, 2013
[Feeling hungry for more information about strategically using data to get after your target accounts? Join us for a webinar on April 24 and learn the secret sauce of account-based pursuit. Register now!]
After reading DemandGen’s recent post about marketers’ focus on cleaning up their data to achieve better ROI, it got me thinking about the processes we use here at OppSource and what best practices we recommend to our clients to help control the health of marketing databases. Although most marketers already have a database in place, there’s always an opportunity to create a new or supplementary database (pursuing a new target segment, for example) and there are some key things you need to plan before you purchase or create a database.
Before purchasing/creating a marketing database:
- Define your Total Addressable Market (TAM). You really can’t be too exhaustive in defining exactly the companies and contacts to which you want to market. Examples of company attributes: Minimum/Maximum Revenue, Minimum/Maximum Number of Employees, Geography, SIC code, etc. On the contact side, be sure to define titles, roles, and departments as well.
- Do a little digging on third-party sites (D&B, government sites, etc.) and get a feel for just how many companies and contacts meet your TAM. If it feels too small, your constraints may be too narrow. If it’s too large, you might want to reconsider who your ideal targets include.
- Contact multiple third party list source providers. Provide them with your very detailed Total Addressable Market definition and ask them to tell you how many companies/contacts they have to meet your needs. If the number of companies is larger than your number in Step 2, this should be an alert that the data provider may not have the highest quality data.
- Provide the list providers with any suppression rules and suppression lists. Make sure you don’t end up buying data you don’t need or won’t use.
- Ask for samples ( 20 to 50 records) from all list providers. Take the time to review the samples and confirm that the contacts and companies provided fit your Total Addressable Market and that the contacts/companies are not included in the suppression list. Marketers often get trigger-happy to complete data projects and end up buying lists without really knowing what they’re buying. The last thing you want is for a salesperson to get something passed to them that doesn’t match what you discussed. If the samples look good, move forward with to Step 6. If the list provider is unable to provide samples that fit into your TAM or included contacts/companies that should have been suppressed, move on to another provider.
- Don’t pull that trigger too soon – make sure you take the time to understand what the list providers’ processes are for replacing bounced contacts.
Once you’re confident that you have providers that can get you the data you need at the right price for your company, here are some helpful hints for the purchasing process:
- Purchase all of the data that fits into your TAM and is not supposed to be suppressed from the cheapest list provider first.
- Next, give the second cheapest list provider a new suppression list that includes the purchased data from Step 1 and any other contacts/companies that were on the original suppression list.
- Repeat this process until you run out of list providers.
You have now purchased a clean database. Prior to importing any of this data into you system, take some time and review the data to confirm you’ve purchased the right companies and contacts one last time. This can be done rather simply by using Excel. If you run into any issues with the data, contact the provider(s) immediately to rectify the situation. It’s much easier to do this on the front end rather than if someone complains months down the road.
When you do import the data, don’t forget to properly add any tags/attributes/categories you’re using throughout your programs so that it’s consistent with any existing databases.