These days, artificial intelligence (AI) and machine learning have become the hottest topic in marketing.
Several martech companies recently announced new AI solutions at Dreamforce. Salesforce recently debuted its Einstein AI Solution for CRM, claiming that AI “will be more disruptive and powerful than any previous shift in technology.” Others such as Oracle and Microsoft have also made announcements recently about their own AI initiatives in the past few weeks. Demandbase, the account-based marketing technology provider, recently announced a new AI driven business graph that now powers their solutions, claiming that the technology “replaces cold calling.”
All these innovations promise to simplify datasets for marketers and sales people to make more intelligent decisions based on data and behavioral triggers. They boldly claim a myiad of benefits such as better filtering of leads, ad targeting, better customer interactions, deeper insights into website interactions and significant lift in productivity for salespeople and marketers.
But is AI actually going to deliver on its promise for B2B marketers?
As Mark Smith, President of Kitewheel puts it in his LinkedIn article, what marketers call “AI” today is just a new name for analytics. This iteration is just “automated analytics.”
As we know, all analytics, whether it’s just descriptive analytics or predictive analytics, is only as useful as the underlying data. As it turns out, AI faces the same problem as traditional marketing automation – data accuracy.
“Data is AI’ s Achilles’ heel,” says Doug Bewsher, CEO of Leadspace in a recent interview with Demand Gen Report. “No matter how smart the algorithms, their effectiveness is crippled if you use with them with the same old basic, static contact data in your CRM or purchased from traditional data vendors.”
Scott Brinker, the publisher of chiefmartech.com, also voiced skepticism about AI’s usefulness in the marketing context. He called out that for machine learning to work well for marketers, the data needs to be accurate to be “relevant to the future.” If marketers want to use AI-based technology to help them identify their best leads or best accounts, they need to feed it accurate, real-time buying signals and data.
While it’s natural to get excited about these new AI-driven offerings, this is actually a good time to think about the health of your prospect database and take some steps to improve data accuracy and quality, and to add more real-time data to your marketing automation system. At the end of the day, AI is simply the latest iteration of analytics geared towards addressing age-old marketing challenges like these:
- How do we segment our leads to increase response rate?
- How do we best follow up with the leads we’ve generated? Which ones do we need to continue to nurture? Which ones are sales ready?
- How can we target the right leads at the right time with the right message?
Here are some immediate steps you can take to improve the quality of data in your marketing automation system to better segment your leads, prioritize your leads and pass more leads to sales.
1.Capture Social Data and Use it to Understand Buyer Interests and Intent
Most marketers are already capturing a lot of data – email engagements, website clicks, paid ads – that help us infer buyer interest and intent. But most of this data is fairly indrect. Social media data is a bit different though. Social media is where b2b buyers are self-publishing content, expressing their interests and pain points. Social media data is arguably the most contextualized type of buyer data that marketers can access.
Interactions on social media – where the average person spends about two hours every day – provide a window into the topics that potential cutomers are talking about, the influencers they follow, and whether they are doing research on products in your market.
To fully take advantage of this real-time, contextual data, you’ll need to get this data into your marketing automation system.
First, be sure to capture people’s interactions with your organic social media content. Many social media management platforms (i.e. Hootsuite, Oktopost, Sprout Social) can be integrated with popular marketing automation and CRM systems. By integrating your social media management platform with your marketing automation system, you’ll be able to create custom fields on your lead records that tell you which social network each lead came from, what link(s) they clicked on from your post(s), and timestamped records of how leads continue to engage with you on social and on your website over time. With this data stored in your marketing automation system, you’ll be able to put leads into buckets for nurturing or remarketing based on their social media engagement. To learn more about how to set up this integration, check with your social media management platform provider.
To get even more insights on buying interests and intent, consider adding a social demand generation solution to the mix. With technology like Socedo, you can identify your target prospects on Twitter, based on their personas and real-time buying signals such as the topics they tweet about and the brands and influencers they follow. From there, you can set up automated nurtures to reach out to prospects and promote your content in the context of their social media activity (i.e. send a link to an e-book on the same topic the lead just Tweeted about). You can also sync new leads’ contact info along with their recent tweets into your marketing automation system. Once you have social data on your lead records, you can easily set up lead scoring and email nurture programs based on these data points.
Based on our experience on the Socedo Marketing team, we’ve found that when people talk about certain topics and events/conferences on social media, they tend to be good leads for us. For example, we’ve found that when people mention keywords like #demandgen and #leadgen, they often want to learn more about our brand. Because we know the topics leads recently mentioned, we are able to set up campaigns that send leads emails offering related content (e-books, webinars) aligned to their interest.
2. Monitor your existing leads’ online behavior to identify warm leads and create a better customer experience
By monitoring your existing leads’ interactions with your brand across all channels, you can detect buying interest, reach out immediately to those who have shown an interest, personalize your communication, and accelerate certain segments of leads through your funnel.
Chances are, you’re already monitoring leads’ engagement with your ads, emails, and website pages in your marketing automation system. These are all signals that indicate buying interest and can be incorporated into your lead scoring model. Social media activity represents another opportunity to understand your leads and personalize your messaging.
With a social lead acceleration solution, you can track the leads in your database, match their email addresses to their Twitter profiles, and then monitor their interactions with your brand, competitors and relevant keywords on social media in real-time. With real-time social activity data appended to your lead records, your marketing team can:
- Identify socially engaged leads in your database
- Segment your database by based on what prospects are engaging with on social media
- Add social behavioral data to your lead scoring model
- Deliver a higher volume of sales-ready leads to your sales team.
The value of this approach lies in this: adding timestamped behavioral activities on individual lead records in your marketing automation system. You can use this timestamped data to trigger real-time marketing actions tailored to that persona’s recent behavior.
For example, when a lead uses a particular hashtag on Twitter, trigger an email that acknowledges them tweeting about that topic and include a link to resource relevant to the context of that tweet. Or let’s say you’re sponsoring a big conference in your industry, you can track which of your leads are tweeting or mentioning the conference hashtag. Chances are, someone who tweets about the conference is going to attend the event. You can reach out these leads and invite them to meet you at your booth. Check out this blog post to see some real-time campaigns our marketing team has implemented.
Based on our own experience, we’ve found that emails triggered by real-time social media behavior have an average of 47% open rate and 5.7% click-rough rate. That’s more than double the open rate and 6 times the click through rate compared to our weekly nurture emails. These campaigns informed by real-time social media activity have full-funnel impact too. One of the keywords we’ve been tracking – #contentmarketing – has helped drive 63 Marketing Qualified Leads over the past 90 days and 3 new customers to-date.
3. Use third party data to enrich your dataset
While you can hope that your website visitors give you accurate information about themselves when they fill out a form to download your e-books or sign up for your webinars, the reality is that people routinely provide false information on these forms.
To combat data inaccuracy, use a variety of data sources to get a complete picture of your leads. At Socedo, our first-party data is captured through landing page form fills when people sign up for a webinar or downloads an e-book. We also fill our database with socially engaged leads sourced from our own product – which targets our audience on Twitter, finds their corporate email addresses and syncs these contacts into Marketo. Once lead data goes into Marketo, we use Clearbit – a data enrichment service – to append company size, address and technographic data to these contacts.
At the end of the day, before implementing any technology, you need to understand who is in your database and what they respond to. It takes experience and conversations to understand your target audience. Remember that human nature isn’t predictable enough for analytics to accurate account for all behavior.