43% of Your MQL Lift Came From One List Import on March 17. Nobody Checked.

Emily Goh · 2026-05-07

The number looked great. The story was wrong.

March was a good month for marketing. MQLs came in at 1,840 -- up 43% from February's 1,285.

Budget meeting was on the 28th. Marketing presented the number. Leadership was pleased. The SDR team got two more inbound leads per rep per day. The pipeline forecast got updated.

Nobody asked where the 555 extra MQLs came from.

The answer: a list import on March 17. A trade show from Q4 of last year. 620 contacts. Almost all of them loaded as new leads, scored against your MQL threshold, and auto-qualified.

No one had checked the list for recency, role relevance, or whether those contacts were already in the CRM. 68% of the MQL "lift" was a single batch upload of 7-month-old event contacts.

By May QBR, the pipeline from those MQLs had converted at 1.2% instead of the expected 8%. Marketing's pipeline contribution number was down. SDR productivity metrics looked bad. Someone asked if we needed to revisit the MQL definition.

The real problem was one list, validated by nobody, uploaded in the middle of the month.

Why list imports break your funnel metrics

List imports are the most common source of MQL inflation -- and the hardest to catch because they look like organic growth.

When a contact is imported into your CRM and scores above your MQL threshold, your system treats them the same as a lead who filled out a demo request form. Same MQL status. Same routing. Same reporting.

But they are not the same.

An inbound demo request is a signal. Someone decided on their own, on a specific day, to raise their hand.

A contact from a 7-month-old trade show list is a database entry. They may not remember meeting you. They may have changed roles. They may have already evaluated and rejected you.

The funnel metrics mix both in the same bucket. Your MQL count goes up. Your MQL conversion rate goes down. Your SDRs work a harder list with lower quality. The board sees pipeline underperformance and asks what's wrong with sales.

What's wrong is upstream. In the data.

What it looks like on a Tuesday

It's a Tuesday in March. The event team just sent over the attendee list from a conference you sponsored in October.

620 contacts. Company names. Titles. Business emails. Some with LinkedIn URLs.

The marketing coordinator uploads it into HubSpot. The import runs. 580 contacts created or updated. The scoring model runs. 310 hit the MQL threshold.

Now you have 310 new MQLs in your queue.

How many of those contacts are already in your CRM? You don't know. The import tool checked for exact email matches, but one of your existing contacts uses [email protected] and the list has [email protected]. Two records now.

How many of those contacts have the right title? You don't know. The list has "VP" in the title field but it's a mix of VP of Sales, VP of Engineering, and VP of Office Administration.

How many of those contacts are at companies in your ICP? You don't know. The conference attracted a broad audience. 40% of the companies are too small or in the wrong vertical.

You don't know any of this. The 310 MQLs are already in the queue.

The budget problem this creates

Your marketing budget is allocated based on MQL volume and MQL conversion rate.

If your MQL count spikes 43% but your conversion rate falls because the new MQLs are low quality, you have three bad outcomes at your next budget review:

1. Pipeline contribution looks low. Marketing generated $X in pipeline. Against the inflated MQL count, that's a lower revenue-per-MQL than expected. Finance asks if marketing spend is efficient.

2. SDR productivity looks bad. SDRs worked the list. Response rates were low. Connect rates were low. Demos held were low. Sales blames marketing. Marketing blames sales. The real issue is the data quality of the contacts.

3. Wrong channel gets the credit. The import was tagged as "Events" in your source tracking. Events now looks like your highest volume channel. You shift budget toward events. But the quality is wrong, so the next event import produces the same result.

The budget decision was based on bad data. The bad data came from a list nobody validated.

What validation actually looks like

Before any list import touches your CRM, three questions need answers:

How many of these contacts already exist? Not just by exact email match -- by domain + name match, which catches the people who use a different email format or have updated their address. If 30% already exist, you're not adding 620 records; you're adding 434 and updating 186. Your MQL lift is actually smaller.

How many of these contacts meet your ICP? Title, company size, industry, geography. Run the list against your ideal customer profile criteria before it enters the scoring model. Filter out the records that would never convert before they inflate your numbers.

How fresh is this data? A list from 7 months ago has a 22.5% annual decay rate on B2B contact data, per ZoomInfo's benchmark. That means roughly 13% of the contacts have changed roles or companies since the event. Before you import, you're importing inaccurate information for 1 in 8 records.

None of this is complicated. It is a 30-minute review before the upload. But it is never done because the process for "check the list before importing" doesn't exist in most marketing ops workflows.

Building that step -- and making it non-optional -- is the difference between MQL numbers you can defend and MQL numbers you spend the next QBR explaining.

If the problem is data quality creating inflated MQL metrics, read about how to catch it: Understanding the Truth About Your Leads and Funnel Performance, Unmasking the Multi-Touch Attribution Trap, and How Dirty Data Is Costing You More Than You Think.

Gohvio is free to try -- gohvio.io