The Economic Cost Curve
No identity resolution system operating at internet scale will be 100% perfect. The practical question is: What is the expected economic outcome of activating a record with some risk of inaccuracy versus not activating a record at all?
In many channels, the marginal cost of activation is low enough that occasional record-level inaccuracies do not undermine overall ROI. In higher-cost offline tactics, the cost of being wrong increases sharply and should be managed with stricter gating and validation.
The best way to think about this is to treat identity data like any other performance input: confidence-weighted, cost-sensitive, and governed by unit economics.
The Two Real Risks
When people talk about “data quality,” they are typically reacting to one of two risks:
1) The Cost of Being Wrong
You spend money targeting the wrong person, wrong household, or wrong company.
2) The Cost of Not Acting
You fail to market to a real in-market visitor because you rejected an imperfect record, and you lose the conversion opportunity.
Most teams focus only on the first risk. In practice, the second risk is often larger, especially in lower-cost digital channels.
Inaccuracy Dimensions and Impact Weighting
A record can be imperfect in multiple ways, and each has a different economic impact. There's a few dimensional considerations.
A) Identity Linkage Is Wrong
The record refers to the wrong individual.
Impact: potentially high, depending on channel cost and targeting precision.
B) Identity Linkage Is Right, but Attributes Are Stale
The person is correct, but employer, title, income band, household composition, etc. are outdated.
Impact: often low in broad marketing; higher in narrow B2B targeting.
Email or address is deliverable, but enrichment fields are incomplete or slightly off.
Impact: usually minimal in email and retargeting; moderate in direct mail; higher in highly personalized outreach.
A mature activation strategy distinguishes between these scenarios and applies the right controls.
The Core Framework: Cost per Attempt vs Cost of Error
A productive way to think about this is through a cost-benefit analysis.
What does it cost to make an attempt in this channel?
What is the downside if the record is imperfect?
What is the upside if the record is correct?
What additional validation should be required at this cost level?
As channel cost rises, the required confidence threshold should rise with it. For example, if you plan to send an expensive personalized direct mail piece or kit, it is generally advisable to have several prior marketing touchpoints to validate that the record merits the implied unit cost of that outreach.
Channel-by-Channel Tradeoffs
1) Low-Cost Digital Channels (Generally Forgiving)
These channels are economically tolerant of some record-level imperfection because the cost to test is low.
Cost per attempt: fractions of a cent to a few cents
Downside of being wrong: low (bounce, unsubscribe, missed impression)
Upside of being right: high (conversion, pipeline, repeat purchase)
Recommended approach for Email:
Measure outcomes at cohort level, not record-by-record
Expect occasional anomalies without invalidating the strategy
Digital Retargeting / Programmatic Ads
Cost per attempt: low relative to offline tactics
Downside of being wrong: low (some wasted impressions)
Upside of being right: strong incremental lift
Recommended approach for Digital Retargeting / Programmatic Ads:
Broader activation can be reasonable if unit economics work
Tighten gating for expensive audiences or constrained budgets
Optimize toward conversion and incrementality, not “perfect identity”
2) Medium-Cost Channels (Require Gating)
These channels carry meaningful marginal costs, so confidence controls matter more.
Sales Outreach (SDR Sequences, Manual Personalization, Direct Outbound)
Cost per attempt: moderate to high (labor-driven)
Downside of being wrong: wasted effort, brand friction, internal skepticism
Upside of being right: meetings and pipeline
Recommended approach for Sales Outreach:
Gate on intent (Hot/Warm) and engagement thresholds
Treat identity as a qualifier, not a guarantee
Combine identity (“who”) with behavior (“why now”)
Iniate after digital touchpoints have potentially returned the user to the site for one or more additional sessions
3) Higher-Cost Channels (Strict Controls + Secondary Validation)
As costs increase, activation should require stronger evidence.
High-Cost Paid Social (ex. LinkedIn)
Cost per attempt: materially higher than broad programmatic
Downside of being wrong: elevated CAC
Upside of being right: strong response in narrow segments
Recommended approach of High-Cost Paid Social:
Use intent gating and frequency thresholds
Run smaller test cohorts before scaling
Prioritize behavior-qualified audiences
Direct Mail (Postcards, Letters)
Cost per attempt: moderate to high depending on format
Downside of being wrong: wasted print/postage, potential brand awkwardness
Upside of being right: high response and memorability in certain verticals
Recommended approach for Direct Mail:
Do not send mail solely because a record exists
Gate on intent and/or prior digital engagement
Ensure address verification runs before fulfillment (if you run Direct Mail with Untitled, this happens automatically).
Operational hygiene reduces the effective cost of being wrong.
4) Very High-Cost Fulfillment (Certainty Required)
Premium Interest Kits / High-Cost Sends
Cost per attempt: very high
Downside of being wrong: unacceptable waste
Upside of being right: meaningful revenue impact
Recommended approach for Premium Interest Kits / High-Cost Sends:
Require multiple independent qualification signals
Consider staged funnels (digital → mail → premium send)
Apply secondary validation before triggering fulfillment
At this tier, identity resolution is one component of a broader qualification framework you should have layered into your marketing operations anyways.
Confidence Tiers for Activation
A disciplined model is to define activation tiers:
Tier 1: Broad Activation
Allows modest uncertainty
Measured in aggregate ROI
Tier 2: Qualified Activation
Requires intent and engagement thresholds
Measured in pipeline efficiency
Tier 3: Verified Activation
High-cost offline channels
Requires secondary validation and strong behavioral proof
Measured in strict unit economics
This prevents the common mistake of applying the same risk tolerance across all channels.
What “Good” Looks Like in Practice
A defensible activation posture includes:
Using buyer intent as a confidence layer
Matching channel cost to confidence threshold
Validating where cost justifies it
Measuring outcomes at cohort level
Iterating based on economic results
Identity resolution should be governed by performance discipline over anecdotal record checks.
The Big Question
The economic question is not “Is this record perfect?”
It is: Given the cost of this channel, is the expected upside of targeting a real visitor worth the small risk that some records will be imperfect?
For low-cost digital activation, the answer is often yes. For higher-cost offline activation, the answer should be yes, but with proper phasing, gating and validation.
