There is no line item on your marketing budget labeled "bad data." No invoice arrives each quarter itemizing the cost of a contact whose job title changed eight months ago, the lead that got routed to the wrong sales rep because the company size field was blank, or the campaign that underperformed because 30% of the list had decayed into irrelevance. The cost is real — IBM Research puts it at $15 million per organization per year on average, and $3.1 trillion annually across the US economy — but it hides inside metrics you already accept as normal: low email open rates, disappointing campaign ROI, sales sequences that go nowhere.

34%

of B2B contact data decays every year — meaning a list built 18 months ago may have over 50% of its records materially inaccurate

This post is about making that invisible cost visible, and giving you a framework for auditing your own HubSpot database before the next campaign cycle begins.

The 34% Annual Decay Problem: A Field-by-Field Breakdown

People change jobs. Companies restructure. Phone numbers get reassigned. The B2B data you captured last January is already partially obsolete today, and the rate of that decay is faster than most RevOps teams account for in their planning.

The industry benchmark is 34% annual decay across a B2B contact database. But that aggregate number obscures significant variation by field type. Understanding which fields decay fastest tells you where to focus your enrichment and validation efforts.

Job titles decay at 65.8% per year. This is the fastest-decaying field in any B2B database and the most consequential for personalization and lead routing. Nearly two-thirds of the job titles in a database captured a year ago are now inaccurate. If your lead scoring model weights job title — and most do — you are scoring on stale signals.

Direct phone numbers decay at 42.9% per year. Mobile numbers are somewhat more stable than desk phones, which have largely disappeared from knowledge worker environments post-2020. If you are running outbound sequences that include calls, almost half your phone data is wrong within 12 months.

Email addresses decay at 37.3% per year. Corporate email aliases tied to employment are the primary driver. When someone leaves a company, their email typically deactivates within 30-90 days. You will not always know — many email servers do not bounce on the first send. They soft-bounce, sit in limbo, and quietly damage your sender reputation over time.

Company firmographics (size, revenue, industry) decay at approximately 20-25% per year. Slower, but not static. Acquisitions, pivots, and hypergrowth change these fields meaningfully. A company you classified as a 50-person SMB two years ago may now be a 500-person enterprise mid-market account.

Physical addresses and LinkedIn URLs decay more slowly — roughly 15-18% annually — but are often wrong to begin with because they were never captured accurately in the first place.

The practical implication: if you run a campaign to a contact list you built 18 months ago without re-enriching it, you are talking to a population that is substantially different from the one you targeted. For a list of 10,000 contacts, that could mean 5,000+ records with at least one material inaccuracy.

The 1-10-100 Rule Applied to HubSpot

The 1-10-100 rule is a data quality principle from the quality management discipline that translates directly to CRM operations:

$1 to verify and correct a record at the point of capture
$10 to clean and fix a record after it has entered your system
$100 to do nothing and absorb the downstream cost of bad data

This rule gets compressed in HubSpot environments because the CRM is not a data warehouse sitting at the end of a process — it is the live operational system that drives campaigns, sequences, routing, and attribution in real time. Bad data does not sit in a corner; it actively participates in every revenue motion you run.

Consider what $1, $10, and $100 look like concretely:

At the $1 level, you are talking about validation at form submission — checking email syntax with a regex, using a form enrichment tool to auto-populate company and job title at the moment of conversion, setting required fields that prevent records from being created with critical data missing. This is the cheapest intervention and the highest-leverage one.

At the $10 level, you are running enrichment workflows after the fact — pulling in a provider to fill missing fields, running a deduplication pass, flagging contacts for manual review. This is where most teams actually operate, because most teams underinvest in the $1 layer.

At the $100 level, you are absorbing the costs that do not feel like data costs: the sales rep who spent 20 minutes researching a prospect before a call and still pitched the wrong persona, the email sequence with a 1.2% reply rate because half the recipients had already left the company, the marketing campaign that missed its MQL target because the segment you targeted had drifted from your ICP over the past year.

The insidious thing about the $100 category is that it rarely surfaces as a data quality problem. It surfaces as a "we need better messaging" problem, a "sales isn't following up fast enough" problem, or a "our ICP needs to change" problem. The root cause goes unexamined because it is invisible.

Where Bad HubSpot Data Shows Up in Your Revenue Engine

Email Campaigns and Deliverability

Every send to a decayed email address is a risk. Hard bounces directly harm your sender domain reputation. Soft bounces and spam trap hits do the same over time, more quietly. If your bounce rate creeps above 2%, most email providers begin throttling your deliverability — which means your clean records also start missing the inbox. Bad data on 20% of your list can degrade campaign performance for 100% of your sends.

Lead Routing and Assignment

Most HubSpot routing rules use firmographic criteria: territory by region, company size, or industry vertical. If those fields are blank or inaccurate, the routing logic fails silently. The lead either lands in a default queue where it ages, or it goes to the wrong rep entirely. Speed-to-lead matters enormously here — contacts contacted within five minutes of conversion are 21 times more likely to convert than those contacted after an hour. A routing failure caused by a missing "State" field can cost you that window entirely.

Sequences and Personalization

Personalized sequences that reference job title, company, or industry are unusable against decayed data. A VP-level message sent to someone who is now an individual contributor after a layoff is not just irrelevant — it can actively damage the relationship and the brand. Personalization built on bad data produces an uncanny valley effect that experienced B2B buyers notice immediately.

Lead Scoring and Prioritization

If your lead scoring model uses job title, seniority level, company size, or industry — and virtually all do — stale data produces phantom scores. Records score high based on who they were, not who they are. Your sales team works down a prioritized list of "high-intent leads" that is actually a mix of real opportunities and outdated signals.

Attribution and Revenue Reporting

Revenue attribution models in HubSpot depend on accurate contact-to-company associations, clean deal records, and reliable first/last touch data. Duplicates fracture attribution. Decayed company associations create orphaned contacts. A contact that exists as two separate records — one with a conversion event and one with the deal attribution — will permanently misreport pipeline influence. You will make channel investment decisions based on a distorted picture.

How to Audit Your Own Database: A Practical Approach

You do not need a specialized tool to get a baseline picture of your data quality. Here is a practical audit you can run inside HubSpot today.

Step 1: Baseline your completeness on critical fields

Go to Contacts → All Contacts and filter to your total active contact count. Then filter by each of the following fields being unknown/empty:

Email address
Job title
Company name
Industry
Number of employees
Country/region
Phone number

Export the counts. Calculate the percentage of your database missing each field. This is your completeness baseline.

For most databases, you will find:

20-40% missing job title
30-60% missing phone
15-30% missing industry
40-70% missing company size

If your numbers are lower than this, either your data capture is unusually rigorous, or your database skews heavily toward recently captured contacts.

Step 2: Identify email bounce history

In HubSpot, filter contacts by "Email hard bounced is true." These contacts should be suppressed immediately and flagged for investigation. Then filter by "Email status is invalid." This is your acute deliverability risk population.

Step 3: Find your duplicate exposure

Use HubSpot's built-in Duplicate Management tool (Contacts → Actions → Manage Duplicates). HubSpot will surface pairs it identifies as likely duplicates based on email or name similarity. The number of pairs it surfaces is almost always an undercount — HubSpot uses exact-match logic, so "Jonathan Smith" and "Jon Smith" at the same company will not be flagged. But the tool gives you an order-of-magnitude sense of the problem.

Step 4: Assess recency distribution

Filter contacts by "Last activity date" and segment into cohorts: active in last 30 days, 31-90 days, 91-180 days, 181-365 days, over 365 days. The contacts in the 181+ day buckets are your highest decay risk. If a significant portion of your database falls in the "over 365 days" category, you should assume substantial field-level decay across the full population.

Step 5: Spot-check a random sample

Pull a random 50-record sample from your database. Manually verify five fields per record: job title, company, email deliverability (LinkedIn lookup or Hunter.io search), phone (format validation at minimum), and company size. Calculate accuracy rate. This is your ground truth — it will often be lower than the completeness metrics suggest, because completeness only measures whether a field has a value, not whether that value is currently correct.

What Good HubSpot Data Hygiene Actually Looks Like

Good data hygiene is not a project. It is a system. The teams with the best CRM data quality share a few structural characteristics:

They validate at the point of capture. Forms have required fields for the minimum viable record. Email syntax is validated before submission. Progressive profiling captures incremental firmographic data across multiple touch points rather than demanding it all at once.

They enrich on conversion, not retroactively. When a new contact enters the database, an enrichment workflow fires within minutes to fill standard firmographic fields. This produces a more complete record from the start and reduces the cleanup burden downstream.

They run scheduled deduplication passes. Monthly or quarterly dedup audits catch the accumulation of duplicates that arrive through list imports, form fills, and manual entry.

They track data freshness explicitly. They know when a record was last enriched, and they treat records older than 12 months as candidates for re-enrichment before any outbound motion.

They have defined "campaign ready" criteria. Before a contact is eligible for a campaign send or a sales sequence, it must meet a minimum completeness standard: email present and valid, job title present, company name present. This prevents the worst-case scenario — sending a personalized sequence to a record that has no reliable fields to personalize on.

The Next Step: Know Your Database Grade

The first move in any data quality improvement effort is establishing your baseline. You cannot prioritize remediation without knowing where your worst exposures are.

The most efficient starting point is a structured audit of your HubSpot database — not just a completeness check, but a field-level assessment that tells you where decay, invalidity, and enrichment gaps are concentrated.

Get your free database health scan: Connect your HubSpot via OAuth and receive an A-F grade distribution across your contact database in 60 seconds — no data is extracted or stored. Claim your free health scan with MarketingSoda Refine™

The Real Cost of Bad HubSpot Data: A RevOps Breakdown