Automation mini-guide part 4: Improving marketing and sales automations
I would like to share with you some insights on marketing and sales automation, and the change it can bring to our sales process and the way we interact with prospects and customers. Automation is a broad term that can be applied to many different domains, in this Mini-Guide we focus specifically on workflow automation and sequences in the execution of marketing and sales. In this fourth part, we will elaborate on how to measure and improve the performance of your marketing and sales automations. You can find all parts of this Automation Mini-Guide here:
- Part 1: An introduction to marketing and sales automation
- Part 2: How to start with marketing and sales automation
- Part 3: Examples of marketing and sales automation
- Part 4: How to measure and improve workflow performance
Configure and test, optimise and test again
Once you have mapped out your sales process, worked out all the steps and conditions, know what should trigger or stop a workflow, and created all the content for the automation, you can start creating the workflow in your marketing automation tool. We strongly recommend to keep a workflow small in the beginning and to only include the core of the process you want to automate. This way, you avoid starting with a huge workflow with many different segmentations and branches, but from which you can derive little insight due to a lack of data in each branch. Instead, we think it is better to create a small workflow and collect enough data to then extend that workflow with more personalisations based on the different segmentations you want to include in your marketing or sales process.
When you design a workflow for the first time, you should test it and see if it works before you start using it. If there are no errors found in the workflow, you can turn it on and monitor it closely for the first few days to make sure everything is running smoothly. After a while, you can make changes to the workflow to improve it. Usually after a week, or sometimes a month, you will have collected enough data to assess how the workflow is performing. Is the workflow doing what it is supposed to do? Are expectations about lead behaviour being met? Is the conversion rate of the workflow satisfactory? Based on the data, you can make further adjustments and expand the workflow. If you make changes, it is important to keep testing them and get more data by using them. Make sure you retest each adjustment before implementing it and keep monitoring the data to optimise further.
There are several things you can do to test your workflows and gain new insights into their performance. Below, we discuss three different actions you can take to analyse, extend and improve your workflows in a data-driven way.
How do you prove what works and what doesn’t?
If you want to learn proactively based on data that you can generate in your own sales process, you need to focus on concrete components that you can control yourself and thus gain targeted insights based on expectations or knowledge that you already have.
There are three ways in which you can proactively test hypotheses and collect targeted data: split testing (A/B testing), workflow goals, and the attribution of your marketing workflows to actual sales results.
Split testing
A/B testing is a proven way to improve your marketing and sales workflows. By running two or more strategies simultaneously and comparing them against each other, you can determine which one is the most effective. Split testing is a scientific approach that allows marketers to compare two or more versions of, for example, the same webpage, app, email or, in this case, workflows. It allows you to measure the effectiveness of different automations and compare the results. The process is simple: based on the expectations you have, your hypotheses, you create multiple variations of your content and publish them simultaneously. Then, you analyse which variant performed better in terms of conversions (i.e. leads, sales or whatever your goal may be) and make it the default option for all your leads. For example, in the case of a marketing email, you can create two versions and send both versions to a small portion, say 5% of your contact database. The best performing version is then sent to the other 90%. That way, you optimise your touchpoints in a data-driven way.
A/B testing is a great way to improve your workflows because it gives you valuable insights into what works and what doesn’t. In the same way as in the example of e-mail marketing above, you can also test different variants of automations. A first way is also to test different variants of messaging to compare the differences. In addition, you can test actual changes in sequences of workflows, e.g. variants with more or less touch points, different content formats, and so on. It is important to work in a structured way and not to randomly make as many variations as possible. Sufficient data should be available to measure each change and to compare it with the original situation. Above all, work on the basis of concrete expectations from previous experiences in your sales process.
Besides focusing on proven best practices, it is important to test everything you create to make sure it works for a specific audience and in a specific situation. By planning A/B tests for different automated emails, you can learn a lot about the behaviour of your contacts in response to your messaging and delivery. It’s great that we can now follow up with leads and customers automatically, but if the automated emails don’t match the situation and expectations of the recipient, you’re only doing more harm than good. Split testing is particularly important to ensure that automations don’t just magnify inefficiencies.
Setting workflow goals
How you can implement this partly depends on the functionality of the automation tool you have. For example, within HubSpot this is a functionality you can configure in the workflow itself, but if you don’t have this capability available in your tech stack you can also analyse it manually.
You can set a workflow goal to set the objective of your workflow and measure the success of the workflow towards that specific goal. By tracking the progress of contacts through the workflow, you can learn how automations affect their progress and timings through the engagement funnel. What you are looking for is to find out what the conversion rate is of this particular workflow to the goal you have in mind with the automation. This is easiest to analyse if, in addition to the conversion rate of the workflow, you also have data on this part of the sales process that you used to do manually.
As an example, you have automated a sequence to follow up on leads with whom you have had an initial discovery or demo meeting. The goal is to set up a second meeting where other people from the decision unit are also present. If you did this manually before and now automated, you can compare the conversion rates and time-to-conversion of the workflow compared to the manual way. Based on these insights, you can see if you need to make further adjustments to improve your workflow. Then you can also analyse these goals for each of the different A/B tests you have created. If you have created several A/B tests, you can then analyse these goals for each of the variants.
Attribution
Marketing attribution refers to measuring the impact that your marketing efforts have on sales. Marketing attribution in B2B sales is the ability to measure the impact of marketing on revenue. It is a critical component of B2B sales because it helps businesses determine which marketing initiatives are most effective at driving revenue and thus prevents focusing too much attention on vanity metrics.
While split testing and workflow objectives can provide data and insights in a more immediate way, attribution often takes longer – especially since in B2B sales, sales cycles are typically quite long. Advanced automation tools provide the ability to attribute the influence of touchpoints on contact conversion, allowing you to better understand which content and channels perform best in a contact’s path to conversion.
There are different attribution models, and it depends on the type of tactics you use and the part of the sales process you analyse which attribution model is most relevant. If you’re lucky with your tech stack, you won’t have to break your head over this, but your tool will calculate it for you. In HubSpot, for example, you can assign your workflow to a marketing campaign, allowing you to see different attribution models and thus the impact of your marketing initiatives on opportunities generated and ultimately the impact on revenue.
Automation Mini-Guide
You can find all parts of this Automation Mini-Guide here:
- Part 1: An introduction to marketing and sales automation
- Part 2: How to start with marketing and sales automation
- Part 3: Examples of marketing and sales automation
- Part 4: How to measure and improve workflow performance
