Whenever marketing employs any campaign tactic, the ultimate goal should always be revenue.
Those of us in B2B with long sales cycles simply can’t wait until deals start closing to determine whether our tactic is worth the money we’re spending on it. Generally, a couple of quarters go by before we start seeing returns, and we’re usually under pressure from the CFO to shut off our expensive campaigns before then if we can’t prove the tactic is working.
Marketers need multiple checkpoints throughout the buyer journey to ensure we’re on track for positive ROI. Generally speaking, those checkpoints are early in-app indicators, engagement, and attribution.
Think of it like steering a ship. If you only relied on GPS, you may still high center yourself on a reef if you don’t also check a nautical chart and take a peek out a window every once in a while. One measurement on its own really isn’t enough.
This is true with any campaign, but chat tools add an extra layer of complexity. Out-of-the-box it’s unusual to see where people came from before they engaged with the chatbox. And knowing where those engagements come from can make the difference between accelerating ideal fit chat engagements and being at the fickle whim of random prospects who stumble upon your page.
So, in addition to early in-app indicators, engagement, pipeline attribution, and revenue attribution measurements, we also need to understand Channel statistics if we want a chance to influence the volume of chat engagements.
Basic Chat Best Practices
While we’re primarily focused on performance optimization, here are a few must-do’s:
Minimize the number of prompts for more information (similar to web forms, less is more)
Put some personality into your bot
Make sure meeting requests are logged as Campaign Members
Do everything in your power to capture UTM sources and associate them with the appropriate campaign member
Don’t set it and forget it – constantly improve your playbooks!
Speaking of UTM parameters…
Setting Up UTM Capture
I’ve had chat vendors act surprised when I ask if they capture UTM parameters and pass those on to Salesforce in the Campaign Member created during a chat engagement, but I know I’m not the first or last person who will ask for this essential piece of data. Unfortunately, many popular providers haven’t tackled this data collection point.
Pro Tip: UTM parameters are a must-have for marketers who spend money on digital advertising, whether those ads are on a social media platform or a search ad. It’s a far more elegant way to capture channel impact than creating a landing page for every advertisement and building campaigns to correspond to each page. (For more on using UTMs, check out our guide here or our article on campaign best practices).
Some chat providers pass UTM information to the Lead or Contact record. We recommend creating a Flow in your CRM to push this information to the Campaign Member when appropriate to ensure your tracking is aligned correctly.
Now that we’ve discussed a frequent gap in chat reporting let’s talk optimization!
In-app analytics is a good place to start optimizing any campaign in the first few days or weeks. You can see conversation rates, email capture rates, and meeting rates. The trick is finding a reasonable industry benchmark. For example, home page prompt conversation rates may be as low as 1-2%, but on landing pages with a stronger CTA, you may aim for as high as 5%.
While these benchmarks help determine whether your playbook’s initial “Howdy!” message that entices a conversation is compelling, it still won’t tell you:
If the right people are engaging with chat
If meetings are converting into sales
Where they clicked before engaging with your chatbot
That said, we all need to start somewhere. This article gives an in-depth look at average performance for home page engagement and the levers that can be pulled to get more engagement. Your tone, message, and CTA can all influence whether or not someone takes you up on that demo.
Engagement scoring is the next logical step in our optimization journey. Scoring models allocate points to key interactions and often apply a time decay so that salespeople are focusing on the most engaged people today. For an excellent podcast episode on the topic of scoring, click here.
The real value here is layering in demographic and firmographic data as a useful way to see whether the right people and companies are engaging with our carefully crafted playbooks. Early on in your chat playbook’s lifecycle, this should give you an indication of whether or not you’re heading toward the ultimate goal of more revenue.
It’s also our first peek at where these chatty people are coming from.
As our chat playbooks age, we can expect more and more meetings to translate into pipeline. Depending on the average amount of time from first sales engagement to opportunity creation, that could be as little as two weeks after you launch the report to two months before you begin seeing pipeline attribution.
Ideally, that attribution keeps moving up and to the right as more and more people engage with your chat playbook. Particularly if you follow our motto of ABI:
Always Be Improving
To analyze the pipeline associated with chat, two models come to mind. Our Middle Touch model assigns all attribution to the touch that happened immediately before an opportunity was created. While we can expect most chat interactions (particularly if the primary CTA is a demo request) to occur immediately before opportunity creation, we recommend using a multi-touch model to determine the true effectiveness.
We also recommend pivoting the data by Channel, which is derived from the UTM capture we outlined earlier. Again, knowing where our chat interactions come from is the best way to understand how to influence the volume and quality of those interactions. In our case, we’re working on improving our SEO and paid advertising using the ABI model and mirror nearly the same optimization reports we use for chat.
We’ve done our homework, improving our message and delivery using early indicators. Our chat tool has been in place for more than four months, so it’s time for the moment of truth.
Did we achieve more chat-influenced revenue (using a multi-touch attribution model) than the cost of our tool? In other words, what’s our ROI?
Using these reports, we can see an estimate of the total contribution to revenue our chat interactions have produced and which channels those chatty people have come through. We like looking at channel sourcing at different stages (engagement -> pipeline -> revenue) because some channels are more effective at driving initial engagement while others are more effective at translating into revenue.
Additional Attribution Resources
For a deep dive on attribution, download our attribution buying guide.
Check out an expert discussion on how to connect campaigns to pipeline so you can confidently answer questions about the revenue impact of marketing campaigns.
With these lifecycle-appropriate benchmarks in place, your optimization ability is only as limited as your time. We wish you the best of luck on your data journey and encourage you to engage with the chatbox on this page if you have any questions.