Best Practices of Setting Decay In Account Scoring

​

Best Practices for Setting Decay In Account Scoring

In Account-Based Marketing (ABM) platforms, decay functions are employed in account scoring to ensure that engagement metrics reflect the most recent interactions. This approach prevents outdated activities from disproportionately influencing an account’s perceived engagement level, thereby maintaining the relevance and accuracy of the scores.

​

Why to Use Decay in Account Scoring:

Decay mechanisms adjust scores over time to prioritize recent interactions over older ones. For instance, if an account hasn’t engaged with your content or platform recently, its score would gradually decrease, signaling reduced current interest. This dynamic scoring ensures that sales and marketing efforts are directed toward accounts demonstrating active engagement.

​

🧠 Best Practices: How to Decide Decay Days

​

🔹 1. Base it on intent strength of the activity:

👉 You can apply no decay only for very high intent actions — like a demo request — if your strategy demands it.

​

🔹 2. Base it on Sales Follow-Up Timelines:

💡 Principle: If your SDRs typically follow up within 14 days of an MQL surfacing, you want to ensure the signals contributing to the score are recent enough to still be relevant. Example for Illustration ❌ Without proper decay:

• An account, let’s say had 50 impressions (1 pt./impression, clicks an ad (5 pts./click) and visits your site (5 pts./click). Hence, score = 50x1 + 1x5 + 1x5 = 60
• They go quiet for 45 days
• No new activity, but the score is still 60 → SDR reaches out… way too late

✅ With aligned decay:

• Same activity: ad click + visit = 60 pts
• But decay is set to: Impressions: 7-day decay, Ad click: 30-day decay, Site visit: 30-day decay
• After 30 days of inactivity, score drops to ~0 → Account no longer looks “Warm” in your system

🎯 Result: Sales is only spending time on accounts that are actively engaging, not ghosts from 6 weeks ago.

​

🔹 3. Use Activity Frequency to Adjust Decay

💡 Principle: The more frequent an activity type is (like impressions), the shorter its shelf life. Example for Illustration ✴️ Scenario A – Frequent Signal (Ad Impressions):

• An account sees your ad 15 times over 1 week → Each impression = 1 pt. → Total = 15 points
• If impressions decay in 7–14 days, this score quickly fades unless deeper engagement follows.

✴️ Scenario B – Rare Signal (Pricing Page Visit):

• A target account suddenly visits your pricing page (25 points) with 90-day decay
• This rare but high-value visit stays visible longer, signaling possible buying intent.

✅ Summary Visual:

​

🎯 Strategy Tip:

If you’re unsure, run a retro analysis (short for retrospective analysis): Description: Retro analysis  is when you look backward at historical data to understand what worked (or didn’t), so you can make better decisions moving forward.

• Validate if your current scoring & decay logic is working
• Identify patterns in what signals actually led to closed-won deals
• Optimize decay windows, score weights, or prioritization rules

​

🔍 Example: Retro Analysis for Decay Decisions

Let’s say you want to determine the ideal decay period for website visit. Step 1: Collect Closed-Won Deals Pull a list of accounts that became closed-won opportunities in the last 6 months. Step 2: Extract Web Activity For each account:

• Look at the last visit to a high-intent page (e.g. pricing, demo, case study)
• Record the date of the visit
• Record the date the opportunity was created

Step 3: Calculate Time Gap Subtract the date of the page visit from the date the opportunity was created. Step 4: Analyze Patterns Let’s say:

• 80% of accounts that visited pricing or case study pages converted within 20–40 days
• A few took longer (~60 days), but those were outliers

Step 5: Adjust Decay Logic Based on Findings You might decide:

• 90-day decay is too generous — those visits stay “hot” in your system long after they’re relevant.
• Instead, set decay to 30–45 days to match the actual buyer journey timing.

​

🔄 Account Engagement ≠ Sales Readiness

​

🚦How to Think About It:

✔️ Account Engagement = How engaged the account is ✔️ Journey Stage = Where the account is in its decision-making path (Awareness, Consideration, Opportunity, etc.) 🔍 Example to Illustrate:

• Let’s say Account A is very engaged, but mainly with upper-funnel content. They’re in the Awareness stage. Even though their score puts them in the Hot bracket, they’re not sales-ready yet.
• On the other hand, Account B shows fewer signals, but they’re high-intent signals. This account is deeper into the decision-making process, making them more likely to convert. Even though they’re Warm, they should be prioritized over A.

​

✅ Recommended Practice:

• Use Account Engagement to identify who is engaging the most.
• Use Journey Stage + Signal Quality to decide how and when to engage.
• Prioritize high-scoring signals, not just high scores.

​

🔍 Pro Tip:

When in doubt, ask:

• What got them to this score?
• Are the signals high intent or just high volume?
• Where are they in their journey?

A Warm account in Consideration can be a stronger Sales opportunity than a Hot account still discovering your brand.

​

✅ Best Practices to Assign Scores Toward Warm/Hot:

As account stages are based on percentile, the scores can’t be defined but you can follow these core principles for scoring.

• Score based on intent strength, not just activity volume.
• Cap weak signals: Don’t let impressions alone push someone into Warm. Cap their influence to ~10% of total score.
• Use decay to ensure scores reflect recent behavior (you already have this!).
• Allow score accumulation across multiple signal types (ads + web visits + 3rd-party intent). Give extra weight to diverse activity types. Use weights under the Overall Score tab.
• Benchmark against real-world historical Hot accounts to reverse-engineer thresholds.