arrow Back to all Plays

reading time icon 4 minutes read

Intent-Based Audiences


Find perfect-fit prospects who need your product (but just don't know it)

  • step one


    Identify measurable events in the market that correspond to product needs.

  • step two


    Construct your audience(s) by combining multiple data sources in one place.

  • step three


    Use audience size to predict channel performance.

  • step four


    Launch, test, and iterate.

What it is


Remember that time you saw an ad for Taco Bell, and all of a sudden, you found yourself craving a Crunchwrap Supreme? You didn’t even know you were hungry, but before you knew it, you were in line ordering.


Traditionally, that type of targeting has only been possible in the B2C world. (See Target predicting women were pregnant). But this guide shows B2B marketers how to identify both in-market buyers and those buyers that should be in-market and haven’t even realized it yet. (Let’s call them the soon-to-be.)

It starts with identifying specific data attributes and events that correlate to or predict a pain that your product solves, and using those to build tight audience segments that you can activate across channels.

We’ll show you how to use one or more data vendors to identify high-intent segments and reach them across inbound and outbound channels. The result? You’re using your limited resources – both budget and manpower – on the right prospects.



Who it's for


Anyone who:

  • Has a persona whose likelihood to buy increases when a specific event occurs 

  • Has a very concrete understanding of their ICP


Expect to see improvements in the top-of-funnel conversion rate.


Relevant channels

  • Most Relevant: Outbound Email Sequences

  • Potentially Relevant: Target on LinkedIn, Facebook, Instagram, Youtube, GDN, and Gmail (Export data to Twitter, Reddit, and Quora)


How it works

  • step one


  • step two



  • step three


  • step four




1. Identify the best intent signals for your ICP


First, determine which events would allow you to identify a prospect that is in-market. Some commonly-used examples include:

  • Searching for keywords related to your product (aka keyword intent)

  • Visiting your website (aka website traffic deanonymization)

  • Reading reviews about your product (aka visiting sites like G2, TrustRadius, etc.)


These types of attributes are typically used by account-based marketing (ABM) platforms like 6sense, Demandbase, Bombora, G2crowd, etc. 


The constraint of this strategy is that it’s a limited pool at any given time. So the next step is to predict the soon-to-be buyers. They’re the key to consistent top-of-funnel volume.  


  • Think about the combination of attributes that indicates a need for your product. You can reverse engineer these by analyzing your current customers and triangulating common patterns. (E.g. Three months after a funding event and hiring a new head of finance, a customer went in-market for a spend management tool). 

  • There is a universe of publicly available attributes out there if you know where to look for them beyond basic firmographics like company size and industry. 

  • Funding (Crunchbase, Pitchbook, Harmonic)

  • Website traffic volume & sources (SimilarWeb, SEMRush, DataforSEO)

  • Tech stack (Builtwith, Aberdeen, HG Insights, PurpleSonar)

  • Social following (Harmonic,, Phantombuster)

  • Open job postings (Rocks & Gold, Sourcestack, Predictleads, Jobspikr)

And if you’re a product-led growth company, you should have a mountain of 1st party data to mine.

When evaluating data, think about both breadth and depth. Just because a data point exists inside a database, doesn’t mean that it’s available for a large number of companies. 

Expect the process to be iterative – expect a certain amount of experimentation to identify the right signals that matter for your ICP. 

The more attributes you use to filter your audience, the smaller your audience. For example, not a lot of companies have a funding event AND a new hire for a specific role, AND tech installation of a specific technology all at once. We recommend starting with just one or two highly correlated predictive attributes first and layering them on top of your ICP criteria. Then iterate.

(Of course, if you were to use a targeting platform like Primer, this process becomes much easier. See how it works here.)


2. Build a custom audience and push into CRM and ad platforms


Once you’ve identified the best predictors of buying intent, you need to collect the data from various data vendors. Then you’ll merge it into a spreadsheet. Once you’ve got a list of companies, you’ll need to figure out what channels you want to target them on. Channel selection often depends on the size of your audience. Typically, audiences of less than 10K people won’t do as well on social platforms.


We’ve seen customers leverage intent-based audiences for outbound outreach, on paid social, or both. Our native integrations make it easy to import these custom audiences to LinkedIn, Facebook, Instagram, Youtube, GDN, and Gmail as well as Salesforce, Hubspot, Outreach, or a CSV.


3. Launch with custom messaging


The benefit of this strategy is that you can be extremely segmented and personalized in your ad and email copy. You can reference the intent data you’ve gathered to show them how well you understand the pain that they’re in. 


Here’s an example of an outreach email Primer has sent to prospects which yielded a 30% positive response rate.



That said, we’ve heard from customers that more generic messaging can also be very successful. You don’t need to show them everything you know about their company. You can win by simply picking the right time to explain how your product fits its pain point.

We encourage you to experiment to find the right segments and strategies for your audience.


3A. Validate with an A/B test (optional)


How do you know if all this extra work of pulling in additional data is worth it? If you want to measure the incremental lift of targeting an intent-based audience versus targeting your generic ICP, set up an A/B test: 

  1. Split your aggregate ICP into a control group and a test group. 

  2. Control group: Receives your usual sales and marketing outreach.

  3. Test group: Filter leads through the intent data and only apply sales and marketing outreach to them. 

  4. Compare the positive response rates of each group – we’ve seen intent-based audiences respond 5-7x as often as leads in the general ICP.


4. Measure results with audience cohort tracking


If you’re only using outbound as part of your strategy with your signal-based audience, then your KPI will likely just be positive response rates. However, if you are also using paid social, we recommend using audience cohort tracking


When you target your audience natively on ad platforms, the common way to measure results is via last-click conversion via UTM parameters. But very few B2B purchases are made from one click to the ad. With this audience cohort tracking, attribution is dead simple because you determine upfront who will be receiving your ads. This means you can measure the percentage of leads who converted after being served a lead – whether they clicked on the ad or just viewed it.


Lastly, while it’s true that new companies and people will qualify into your signal-based audience over time, we have noticed that there’s usually an initial surge of leads and it then tapers out. Only so many new companies meet the criteria of your signal. Signal-based audiences are by definition a highly qualified segment of your ICP. Other marketing activities are required to push more prospects in your ICP through the funnel.

About us

Primer unlocks hyper-precise targeting for B2B marketers. Get better leads and spend less. Sign up