Behavioral Scoring Model

The RFIS Scoring Model

RFIS is BIGDBM's behavioral signal scoring framework. It measures how recently, how often, and how intensely a consumer has shown intent, combining them into a single Strength score that surfaces the highest-quality in-market audiences.

R Recency
·
F Frequency
·
I Intensity
S Strength

Each behavioral signal in BIGDBM's data is tagged with three individual metrics: Recency, Frequency, and Intensity, combined into a single Strength score. Higher positive Strength values indicate consumers who are actively in-market right now. Negative values indicate suppression or disqualification signals.

ℹ️ All three input signals (R, F, and I) are measured over a rolling 24-hour window refreshed daily. Recency is expressed in whole days, where today = 1 (never zero, to prevent division-by-zero in the Strength formula).
Association Types

RFIS scores are not computed in isolation; they are calculated for specific pairings between identity signals. Each association type tells you something different: intent category interest, household linkage, device linkage, or the full cross-device graph. Every pairing gets its own independent R, F, I, and S values.

Identifier key
HEM
Hashed Email
SHA-256 or MD5 of an email address
The anchor identity in BIGDBM's graph. Every association is rooted to a HEM. a privacy-safe, irreversible hash of a consumer's email that can be matched against advertiser CRM data without exposing the raw address.
IAB
IAB Category
Interactive Advertising Bureau taxonomy
A standardised intent/interest category from the IAB Content Taxonomy (e.g. Automotive > Sedans, Home & Garden > Gardening). When a HEM is observed in the context of an IAB category, an RFIS score is computed for that HEM–IAB pair.
IP
IP Address
IPv4 / IPv6 network address
The household or network identifier. An IP address links a consumer to a physical location or organisation. RFIS scores on HEM–IP associations indicate how consistently and recently a given identity has been observed at a specific network.
MAID
Mobile Ad ID
Apple IDFA / Google GAID
A device-level identifier for mobile attribution and targeting. MAIDs are reset-able and opt-out-able by the user. RFIS scores on HEM–MAID associations track how fresh and reliable the link between a person's email and mobile device is.
The four pairings
HEM IAB Intent Signal

Links a consumer identity to a specific intent category. This is the primary association for building in-market audiences. An RFIS score is computed for each unique HEM–IAB pair observed, so a single HEM can have dozens of independent RFIS scores across different IAB categories (e.g. a high Strength score in Automotive > Trucks and a separate one in Financial Services > Mortgages).

📧 Consumer identity 🏷️ IAB category interest 📊 Per-category RFIS score 🎯 Audience segmentation
HEM IP Household Signal

Links a consumer to a household or network location. The RFIS score here indicates how reliably and how recently a given HEM has been observed at a particular IP address, useful for household graph construction, co-op targeting, and IP-based audience extension. A high Strength score means the consumer was consistently observed at that IP address recently; a decaying score suggests the association may no longer be current.

🏠 Household linkage 📡 Network presence freshness 🔗 IP-based audience extension
HEM MAID Device Signal

Links a consumer's email identity to a mobile device. The RFIS score reflects how recently and frequently the HEM and MAID were co-observed, indicating how confident we are that this device still belongs to this person. A strong, recent association is suitable for mobile campaign targeting and cross-device attribution. An aged association (high Recency) may indicate the consumer has upgraded devices or reset their advertising ID.

📱 Cross-device resolution 🔑 IDFA / GAID linkage confidence 📲 Mobile campaign targeting
HEM MAID IP Full Graph Signal

The richest association: a three-way linkage between a consumer's email, their mobile device, and a household IP address. When all three are co-observed, the RFIS score reflects the strength of the complete identity cluster: person + device + location. This is the highest-confidence signal for omnichannel targeting, frequency capping across channels, and measurement. A high Strength score here means all three identifiers were recently and repeatedly seen together, producing a highly reliable identity match.

🔗 Full identity cluster 📺 Omnichannel targeting 📏 Cross-channel frequency cap 📐 Attribution & measurement ⭐ Highest confidence match
ℹ️ Each association type produces its own independent set of RFIS values. A single HEM may have a Strength of +7.0 on HEM↔IAB (actively researching a category), a Strength of +1.2 on HEM↔IP (household link is ageing), and a Strength of +5.0 on HEM↔MAID↔IP (device and household were co-observed recently). These scores are used independently depending on the targeting use case.
R
Recency
How long ago was the signal last observed?

Recency is the number of days since the consumer signal was last observed. It answers the question: "How fresh is this intent?" A signal seen today has a Recency of 1; a signal seen a week ago has a Recency of 7.

📅
If the signal was observed today → Recency = 1
The minimum value is always 1, never 0, to keep the Strength formula mathematically stable.
🗓️
If the signal was last observed N days ago → Recency = N
For example, a signal last seen 5 days ago → Recency = 5.
📉 As Recency grows, the Strength score decays. A consumer seen 30 days ago carries 30× less weight than one seen today, assuming identical Frequency and Intensity.
Seen today1
3 days ago3
7 days ago7
30 days ago30
F
Frequency
Is this a positive or negative signal?

Frequency is a directional multiplier that determines whether the signal represents positive intent (the consumer is in-market) or a negative / suppression signal (the consumer should be excluded). It is always either +1 or -1.

Frequency = +1 (Positive signal)
The consumer demonstrated positive intent: browsing, clicking, researching, or engaging with category-relevant content within the last day. Their Strength score will be positive.
🚫
Frequency = −1 (Negative / suppression signal)
The consumer expressed disinterest, opted out, or triggered a suppression event. Their Strength score will be negative, flagging them for exclusion.
💡 Frequency acts as the sign gate of the Strength formula. Multiplying by +1 preserves the score's direction; multiplying by -1 flips it. This is what makes Strength capable of expressing both in-market and suppression signals in a single number.
Positive intent+1
Suppression / negative−1
I
Intensity
How many times was the signal observed in the last day?

Intensity is the raw count of signal appearances within the last 24 hours, measured at 30-minute intervals. It captures how deeply engaged or how persistently active a consumer is within a category.

⏱️
30-minute sampling window
Signals are counted in discrete 30-minute slots over the last 24 hours. A consumer can register at most 48 appearances in a single day (one per half-hour slot). In practice, most active signals fall in the 1–10 range.
📊
Higher Intensity = stronger raw signal
A consumer who triggered the same behavioral signal 8 times today is far more engaged than one who triggered it once. Intensity captures that depth before Recency discounts it.
🎯 Intensity is the numerator of the Strength formula. It is then divided by Recency, so a high-intensity signal that's weeks old still decays appropriately.
Light touch1
Moderate4
High engagement10
Maximum (48 slots)48
Strength Score

Strength is the final computed score: a single number that combines all three signals into a prioritizable, sortable value that answers: "How likely is this consumer to convert right now, and in which direction?"

Formula
Strength = Intensity Recency × Frequency
Where Intensity = appearances in last 24 h  ·  Recency = days since last seen (min 1)  ·  Frequency = +1 or −1
S
Strength
The composite intent score. Positive means in-market, negative means suppressed
🟢
Positive Strength (e.g. +8.0)
Consumer is actively in-market. The higher the value, the more recent, intense, and positive the behavioral signal. Prioritize these consumers for outreach.
🟡
Low positive Strength (e.g. +0.1 → +1.0)
Signal exists but is fading; the consumer was in-market several days or weeks ago. May still be relevant depending on sales cycle length.
🔴
Negative Strength (e.g. −2.0)
Suppression signal. This consumer should be excluded from targeting. The magnitude indicates how strongly the suppression signal was observed.
Strong buy signal+8.0
Moderate intent+2.5
Fading signal+0.14
Suppression−3.0
Worked Examples

The following scenarios show how different combinations of R, F, and I produce different Strength scores and what they mean for targeting decisions.

Scenario R
Recency
F
Frequency
I
Intensity
Strength = I÷R×F Signal
🚗 Hot auto shopper
Visited dealer site 8× today
1 +1 8 +8.00
Max priority
🏡 Active home buyer
Browsed listings 5× today
1 +1 5 +5.00
High
📱 Mild tech interest
1 product page view today
1 +1 1 +1.00
Low
⏳ Fading auto intent
8 visits but 3 weeks ago
21 +1 8 +0.38
Fading
📅 Distant signal
Researched a month ago
30 +1 4 +0.13
Very low
🚫 Recent opt-out
Suppression event today, 3×
1 −1 3 −3.00
Suppress
🔕 Strong suppression
Multiple opt-outs today
1 −1 8 −8.00
Hard exclude
🎯 Recommended thresholds for targeting: Use Strength ≥ +1.0 for broad in-market audiences, ≥ +3.0 for high-intent campaigns, and exclude any consumer with Strength < 0 as a suppression candidate. Adjust based on your vertical's typical purchase cycle length.
📐 Quick mental model: Think of Strength as "how many fresh signal appearances per day, adjusted for how long ago they happened." A score of +5 today means the same behavioural weight as 5 same-day appearances, regardless of whether they happened in one burst or across multiple days.