Overview
Scoring models in Autobound help you combine firmographic, signal-based, and research data to rank contacts and companies based on how well they fit your ideal customer profile. This makes it easier to determine which leads deserve manual outreach (calls, LinkedIn steps) and which can move through automated sequences in tools like Outreach, Salesloft, or Clay.
Video:
Step 1: Bring in Contact and Company Data
Start by adding a list of contacts to your Autobound table. Include firmographic details such as title, company, website URL, location, and annual revenue. These columns serve as the foundation for your scoring prompt.
If you’ve already created scoring models elsewhere—such as in your CRM or marketing automation platform—bring those existing scores or attributes into Autobound as a column. This allows you to combine historical scoring logic with Autobound’s real-time insights and research for a more accurate and adaptive scoring model.
Step 2: Add a Research Column
Create a research column that performs deep research across both Autobound’s proprietary data network and public web sources.
This research combines data from:
• Real-time public web content such as company websites, press releases, and blogs
• Proprietary data sources not directly accessible through normal web search, including LinkedIn, G2, Reddit, Twitter, and other partner datasets
• Autobound’s Relevance Model, which runs in the background to rank and prioritize the most meaningful insights based on your industry, persona, and product focus
The result is a rich summary of why each contact or company is relevant, what recent signals or conversations matter most, and how your solution can provide value in that context.
These insights feed directly into your scoring prompt, helping the model understand not just what data exists, but why it matters.
Step 3: Create a Scoring Prompt Column
Add a scoring model column that references both your firmographic and research data. This column applies weighted logic to your criteria—such as company size, revenue, industry, or presence of buying signals—to calculate a total score for each record.
You can design the weighting logic manually or use a tool like ChatGPT to help define it. For example, you might give more weight to funding signals or certain technologies used by the company, while de-emphasizing size or geography.
The output typically includes:
• A total score (e.g., 0–100)
• A breakdown of the weighted factors
• A short explanation describing why the score was assigned
Step 4: Combine Scores with Other Data
You can make your scoring model more robust by including additional data sources such as MQL scores, CRM activity, or website engagement data.
Add these as columns in Autobound and reference them in your scoring prompt. This gives your scoring model a 360-degree view that combines firmographic, behavioral, and intent-based data into one unified score.
Step 5: Route Actions Based on Score
Once your scoring model is live, you can automatically route prospects into the right follow-up paths based on their score.
For example:
• Scores above 75 can be sent to manual call or LinkedIn sequences through Outreach or Salesloft.
• Scores between 50 and 75 can enter a semi-automated email sequence or be reviewed by SDRs.
• Scores below 50 can remain in an automated nurture flow or be routed back to marketing for further warming.
These conditions can be defined with Run Conditions or formulas in Autobound. You can also combine routing logic with exports to your CRM or sequencing tools, ensuring each contact automatically lands in the right workflow without manual sorting.
This score-based routing helps reps prioritize high-value opportunities while keeping all other leads engaged automatically.
Step 6: Sync Scores and Activity Back to Your CRM
Export scores and related activity data back into your CRM or outbound tools. This keeps your sales and marketing teams aligned and allows reps to filter, sort, or trigger follow-up cadences based on Autobound’s scoring logic.
Integrating this data into your CRM ensures every contact’s score, reasoning, and corresponding workflow step are visible and actionable.
Summary
By combining firmographics, proprietary research, and real-time signals, you can build a scoring model in Autobound that:
• Uses data from both public and private networks (LinkedIn, G2, Reddit, Twitter) enriched through Autobound’s Relevance Model
• Scores and explains each prospect’s priority based on your defined criteria
• Automatically routes contacts into the right workflows or sequencing tools
• Syncs back to your CRM for unified visibility and tracking