I feel very strongly that if you do not have the time to test a lifecycle (or money to pay someone else to test it for you), you shouldn’t be building it at all. The revenue cycle model and associated program are one of the most complex pieces to get right in Marketo. Having a poorly working lifecycle process is worse than not having one at all! You don’t want to ruin your relationship with sales by routing leads backwards from SQL to MQL.
So, assuming you are going to test, here are some suggestions of things that you should do before, during, and after the testing process to ensure you do a thorough job.
Before Testing
- Make sure the CRM administrators are aware that you are going to be doing testing and this will involve creating fake records for leads, contacts, accounts, and opportunities. If they aren’t comfortable with this and want you testing on Sandbox, find that out before you build everything in Production.
- Develop a list of scenarios to test each campaign in your lead lifecycle and have your CRM administrator and Marketo power user review it if possible.
- Create scenarios to test all potential sources of new leads - list imports, form fill-outs, manual creation in Marketo, creation in CRM as a lead, creation in CRM as a contact, and creation via the API if applicable.
- Develop scenarios to test things that do and do not create a score, if you are using scoring to move leads forward in lifecycle. The same applies to status changes.
- Pay careful attention to long wait steps. Consider what would happen if the lead changed in some way while in the wait step and make sure you’ve accounted for this properly.
- Think about activities that might trigger multiple campaigns and somehow trigger a lead to move backwards in lifecycle. Score changes and status changes are big causes for these problems.
- Don’t forget scenarios to test leads created in partitions as well as those moving into other partitions if you are using workspaces and partitions.
- Make sure your revenue model is using the manual stage change trigger for all the transition rules.
- Limit all lifecycle campaigns to only apply to a test list.
- I recommend making a smart list with “Email Address is”, which allows you to add test records to your list as needed. Then place the filter for Member of Smart List in all of your lifecycle campaigns.
- Pay careful attention to Advanced Filters that will need to be adjusted as you add another filter.
- Include way more potential test records than you anticipate needing and use a standard number format to vary them. (e.g. testlead1@abc.com, testlead2@abc.com, etc.)
- Update all alert and escalation emails so that they are going to your fellow testers, not other marketing or sales people. Check both the Lead Owner dropdown and the email address section of the send alert to make sure you’ve removed everyone who shouldn’t get it.
- Update the CRM sync step to sync to a different owner (if necessary).
- Update all long wait steps to something reasonable for testing (on the order of 5 - 30 minutes instead of days, weeks, or months), but giving yourself sufficient time to make positive or negative changes to statuses or stages (or receive updates from CRM) so you can test your scenarios.
- Create a tab in your lifecycle testing document for recording the changes you make to each specific campaign and specify exactly what changes you’ll need to make after you are done. Trust me, you won’t remember by the end of this.
During Testing
- When possible, test in person with someone with CRM administrative access and the primary Marketo administrator. A standard acquisition lifecycle takes approximately two full days to test thoroughly for an inexperienced tester. If testing in person is not possible (hours, travel, etc.), schedule 2-3 hour test sessions daily until complete. An hour is not sufficient time to make progress due to set up time and sync speed.
- Record the email address of the test records you are using for each scenario.
- Detail out each step that you’ve taken in the results column. This will be especially helpful if you are testing multiple scenarios at a time (because you are waiting on something) or if you are troubleshooting because something didn’t work as intended.
- Don’t test too many things at one time. I would recommend testing no more than 2 - 3 things at once. Not only do you want to avoid making mistakes because you’re distracted, but your co-testers will get lost and confused if you are jumping around too much.
- If you are still waiting on some results or some of the test scenarios do not apply, mark them as Waiting or Ignore or some other status so you can clearly see what you need to return to later, especially when testing over multiple days.
After Testing
- Remove the test list from all of your smart campaigns.
- Update all of your wait steps back to their original timeframe.
- Adjust all of your alerts and assignments back to the original owners.
- If not going live right away, deactivate campaigns.
.png)