Marketing-qualified accounts: How an MQA is defined, scored and handed to sales

A marketing qualified account (MQA) is a target account that fits your ideal customer profile (ICP), has enough engaged members of its buying group, and shows enough buying signals inside a set time window to justify sales outreach. The unit is the company, not the person. 6sense community guidance puts a buying committee at 4 to 8 people, so set the contact count and time window from your own closed-won history.

What is a marketing-qualified account?

An MQA is a target account that fits your ICP, has enough engaged members of its buying group, and has shown enough buying signals inside a set time window to justify sales outreach. Those are the three parts: fit, people and time. Remove any one and you are back to a list of interesting companies.

Clay's glossary describes an MQA as a company "ready for sales outreach based on the combined engagement signals from multiple contacts within that organization." The unit is the account, not the person. That is the main break from the marketing-qualified lead (MQL), which scores one individual. Clay's suggested build is to aggregate signals "into an account-level score that combines both profile fit and overall engagement."

How platform definitions fit under it

Vendors ship their own versions with their own names. The 6sense Qualified Account (6QA) is one. 6sense says 6QAs "are best used as a replacement for SQLs," meaning sales-qualified leads, though they can stand in for MQLs as well. Its default rule set maps onto the neutral definition like this.

Each row comes from the 6sense default definition. The buying-group count does not appear as its own line there. In our reading, it sits inside the predictive buying stage score. If you build your own definition, you have to write that count down yourself. Access also has a condition: 6sense states that an organization "must have the Predictive add-on package to have access to 6QA features." Demandbase describes using predictive insights "to identify marketing-qualified accounts most likely to generate pipeline."

Writing your own version

A platform-neutral definition fits on one line. As an example only, not a benchmark:

MQA = account in ICP tier A or B
      AND at least 3 engaged buying-group contacts
      AND qualifying signals within the last 30 days
      AND no open opportunity

Start from your ideal customer profile template. Set the contact count and the time window from your own closed-won history, not from the example above. Be careful about setting the bar too low. Clay names setting "the qualification threshold too low" as the most common mistake, because it sends sales accounts with weak intent. Later sections cover scoring the components and the handoff to sales.

How is an MQA different from an MQL?

The short version: an MQL counts a person, and an MQA counts a company. Clay defines the MQL as "an individual person deemed ready for sales." It says the lead model may suit mid-market companies where purchasing power is concentrated with fewer individuals, while the account model aligns better with enterprise B2B sales where buying committees are common.

The table below is our template for writing the difference into a service-level agreement. Copy it and replace each cell with your own rule.

Clay lists the MQA success measures as the MQA-to-opportunity conversion rate, pipeline velocity for qualified accounts, and ultimately, the revenue generated from MQA-sourced deals. The MQL side of the table is our own analogue.

Two rows that break most rollouts

The data row is where projects stall. RevOps Co-op states that qualified accounts "will only work with an automated method to auto-match leads to accounts." Without matching, a new lead lands on a sales development representative while the account owner is already in a deal.

The success row changes how pipeline reads. The same post notes that aggregating at the account avoids over-reporting a single account with seven people engaging simultaneously with your brand. In our reading, that is the main reporting gain for a team that owns a pipeline number.

Which stage the MQA replaces

Not every vendor maps the MQA onto the MQL. 6sense describes the analogous object for most organizations as their MQLs or sales-qualified leads (SQLs), and, as noted in the previous section, prefers the replacement of the SQL stage. Its qualified accounts should be prioritized by your organization's SDRs/BDRs, meaning sales and business development representatives. Decide which stage you are replacing before you rebuild funnel reports, or conversion rates will compare different things.

Keeping both is common. Clay notes that many teams use a hybrid model, and RevOps Co-op adds that sales will still want to know if a committee member has filled out a demo request or performed some other high-value activity. It also advises against the account model for a transactional sales process. For the person-level side, see lead scoring.

When an account model fits and when it does not

Moving from leads to accounts changes routing, reporting and the credit conversation at the same time. It pays off under a short list of conditions, and it is worth confirming them in your own data before rebuilding anything.

RevOps Co-op says the account model "makes the most sense for organizations with an enterprise sales motion or a long sales cycle due to the nature of their product." Those organizations, it adds, rely on marketing to generate dozens of touch points over a sale cycle and have a lot of conflict over who "sources" an opportunity. A third sign is arguments between sales teams or inside sales and full-cycle sellers because leads are assigned to someone other than an account owner.

The same piece names the friction. The model can be problematic for companies that have both direct-to-consumer and business-to-business selling models, and for teams fixated on speed to lead, the matching step may be a hard sell. Before any of this, Finance, sales and marketing have to agree on what an account is, because Finance may define an "account" differently than sales thinks of them.

If two or more signals show up clearly, we think the case for accounts is strong. If none do, an MQL model with good fit criteria is likely enough.

Reading the ABM numbers

Account-based marketing (ABM) figures often travel with this decision. Salesforce's ABM guide states that B2B companies with ABM programs report a 38% higher sales win rate and 91% larger deal sizes, leading to 24% faster revenue growth. Those are self-reported comparisons published in a vendor guide, not a benchmark you can hold your own win rate against. For that, use the win rate benchmarks.

The return on investment (ROI) claims vary from page to page. The same guide says 87% of B2B marketers agree ROI is higher with ABM, and elsewhere that more than 79% of marketers report higher ROI from ABM than any other marketing effort. A separate Salesforce post cites ITSMA (Information Technology Services Marketing Association) research that 84% of businesses that use ABM find their ROI is higher. In our reading, these measure what marketers believe, not measured returns.

Salesforce also frames ABM as something that should support and complement traditional lead generation, not replace it. That fits the hybrid approach covered earlier.

How many engaged contacts make an MQA?

A working answer is the size of one buying committee. 6sense's community guidance puts that committee at 4, 8 folks depending on the size of the company and complexity of the deal. The roles it lists run from a decision maker who "gives the final yes" to a gatekeeper who "controls access or budget", For more on committee size, see our buying committee benchmarks.

Count per opportunity, not per account

One account can hold several buying groups. Forrester gives the case where a target account for a software company may contain three opportunities, one for each of its software products, each with its own buying group. A raw contact count across the whole account mixes those groups together.

Here is an example, not a benchmark. For example, of 30 engaged contacts at one account, a share of 0.2 maps to the buying group for the product in question, so 6 contacts (30 × 0.2) count toward that opportunity. In our reading, that is a real group inside the four-to-eight range. The other contacts may belong to different groups and should be scored against their own opportunities.

Forrester also asks teams to treat second and third leads from the same organization not as duplicates but as reinforcing signals of buying intentions. Dedupe rules that merge them hide the count you need.

Keep signals separate from contact engagement

Contact engagement tells you who is involved. Signals tell you whether the group is in market. Forrester splits them into visits to your own website (first-party signals); partner event participation (second-party signals); and visits to relevant product review sites (third-party signals). Triggering events may include a company relocation or expansion or a change in regulations. Score these as their own component; our intent data guide covers the sources.

An example scoring model

The points below are illustrative. Replace them with weights drawn from your own closed-won history.

EXAMPLE ACCOUNT SCORE (per opportunity, illustrative only)

Fit (ICP tier)
  Tier 1 = 30 | Tier 2 = 20 | Tier 3 = 10

Engagement (buying group for this product only)
  Each engaged contact mapped to the group = 5 (cap at 8 contacts)
  Each distinct role covered = 5
  (decision maker, champion, influencer, user, gatekeeper)

Signals
  First-party (own site visits) = 10
  Second-party (partner event)  = 5
  Third-party (review site)     = 10
  Trigger event                 = 10

Recency decay
  Activity 0-30 days old   = full points
  Activity 31-60 days old  = half points
  Activity over 60 days    = zero

Threshold
  MQA when score >= 80 AND roles covered >= 3

Example: Tier 1 (30) + 6 contacts (30) + 4 roles (20)
         + first-party (10), all within 30 days = 90 -> MQA

The role floor matters as much as the total. Six engaged users with no decision maker can clear a points threshold and still stall in sales.

How to set the MQA threshold from historical conversion

Set the threshold from your own closed history. A number borrowed from a peer or a platform default describes their funnel, not yours. The method has four steps.

1. Pull the last year of opportunities. Export every opportunity created in that period, plus the accounts in your target list that never produced one. You need both groups, or you cannot compute a rate.
2. Count engagement before creation. For each account, count engaged contacts and signals in a fixed lookback window before the opportunity date. For example, use 30 days for shorter cycles and a longer window for enterprise deals. For accounts with no opportunity, use a rolling window of the same length.
3. Bucket and compute the rate. Group accounts by engaged-contact count and divide opportunities by accounts in each bucket. Look for the bucket where the account-to-opportunity rate steps up and stays up. That step is your candidate threshold.
4. Size it against sales development rep (SDR) capacity. Multiply qualifying accounts per month by the bucket's rate, then check that number against what your SDRs can actually work.

bucket = engaged_contacts_in_window (0, 1, 2, 3, 4+)
rate(bucket) = opportunities(bucket) / accounts(bucket)
threshold = lowest bucket where rate rises and holds
expected_opps_per_month = qualifying_accounts_per_month * rate(threshold)

As an example, not a benchmark: suppose the rate steps up at 3 engaged contacts in a 30-day window, and accounts at that level convert at 0.2. In this example, if 30 accounts cross the line in a month, 30 accounts at a 0.2 rate yield 6 opportunities. If your SDRs can work all 30 accounts with proper research, keep the line. If they cannot, raise it a step and recompute.

Write the rule so reps can read it. 6sense, for instance, has users name each qualification reason, with an example reason of "“>3 G2/ TrustRadius/ 6sense Keywords in last 30 days”". Add exit and re-entry rules as well. Its default definition requires that "No opportunities were created or lost in the last 90 days", and accounts usually leave the qualified state once an opportunity is created. Winback accounts may requalify a year after closed-lost.

Loosening and tightening over time

The vendor's own guidance is to loosen or tighten qualification rules over time to balance high-intent targeted accounts against medium-intent, large-pipeline ones. In our reading, the SDR queue tells you which way to move.

Plan for lag when you change a rule. In 6sense, a new configuration takes approximately 24 hours to take effect. Record the change date so the before and after rates stay comparable. The account scoring model template has columns for this log.

What sales does with an MQA, and how the SLA changes

The lead-era service-level agreement promised one thing: a named person gets a call within a set time. An account-level SLA has to promise more, because the unit is now a group of people at one company. In our reading, each clause maps to one of the demand program objectives Forrester describes, which gives both teams a shared reason for every step.

Route to the owner, then enrich

Routing comes first. RevOps Co-op states that "Qualified accounts will only work with an automated method to auto-match leads to accounts." Every new contact should land with the account owner, not in a separate lead queue. Records that fail to match still need a home, since the same guide says "You will still need a process in place to review and follow up on leads."

Next, map the buying committee and fill in the gaps before the handoff. 6sense advises teams to "enrich those records with contact data. Then, hand them off (automatically if possible)" into the customer relationship management (CRM) system. Speed matters here: "The longer you wait, the colder the lead gets."

The cadence and the recycle rule

The same 6sense post describes a three-touch email sequence for the SDR or business development representative (BDR):

1. Email 1: Value & Relevance
2. Email 2: Social Proof & Insight
3. Email 3: Direct Ask & Urgency

If there is no reply after the third email, the account goes back to paid social or the regular nurture program. That is your recycle rule, written down.

Forrester defines the three objectives this way: activate covers new logo demand and "reactivate previously disqualified demand"; validate means to "Capture and engage additional buying group members"; accelerate means to "Increase the velocity of the opportunity and buying group." It adds that "the opportunity and all buying group members must move together from activate to validate to accelerate." Its authority play covers the moment "when buying group members change," which is why the last row exists.

Time limits

Write follow-up limits per step, not per person. For example, and not as a benchmark: owner accepts the account within a set number of business hours, all 3 touches go out within 30 days, then the recycle rule fires. Check your own response data against speed-to-lead benchmarks before you fix the numbers.

MQA SLA clause template
Trigger: account crosses MQA threshold
Routing: matched to account owner within [X] business hours
Enrichment: buying committee mapped and enriched before handoff
Outreach: [N]-touch sequence completed within [Y] days
Recycle: no reply after final touch -> paid social or nurture
Re-entry: account requalifies on new engagement
Owner of each step: [name the team]

The full version lives in our marketing-sales SLA template.

How to report MQAs

An MQA report answers two questions: did marketing produce qualified accounts, and did sales act on them. Six measures cover both. Define each one once, in writing, so marketing and sales read the same number.

The measures and their formulas

The counting measures follow the definitions in the 6sense 6QA Trends report, where total qualified accounts means the total number of accounts that met the criteria for a 6sense qualified account (6QA) within the time period. New ones are the number of new 6QAs generated within the time period. The conversion and pipeline measures below are our own method and come from opportunity records in the CRM.

Total MQAs (period)       = count of accounts meeting MQA criteria at any point in the period
New MQAs (period)         = count of accounts that first met MQA criteria in the period
Worked share              = MQAs with at least one sales activity / Total MQAs
Unworked share            = MQAs with no sales activity / Total MQAs
MQA-to-opportunity rate   = MQAs that gained an opportunity in the window / New MQAs in the cohort
Days to first activity    = average of (first sales activity date - MQA date)
Pipeline per MQA          = pipeline value created from the cohort / New MQAs in the cohort

Sales activity means the same thing in every row. In the 6sense definition, worked is the percentage of 6QAs that had at least one sales activity, such as an email, call, or meeting. Time to first activity is the average time it takes for a sales representative to initiate the first sales activity (email, call, meeting) with a 6QA.

Measure conversion by cohort, not by period. An account that qualified late in the quarter has had little time to become an opportunity, so set a fixed window and report each month's cohort once that window has closed.

For example, and not as a benchmark: if a month produces 30 new MQAs and the cohort converts at 0.2, that month yields 6 opportunities. In this example, if the same 30 accounts show a worked share well below the whole, the rate understates what full follow-up would produce, in our reading.

Dashboard layout

Segmenting by GTM and by campaign follows how the platform frames it: you can filter this report to view trends for go-to-market (GTM) style segments, for segments used for campaigns. Time grouping options are day, week, month, quarter, and dimensions include time, CRM status, and reach score. A filter update logged on Aug 11 2026 added profile fit and ad campaign filters, which makes campaign cuts possible inside the report itself.

Rebuild the same panels in your own dashboard so the platform view and the CRM view reconcile. The demand gen dashboard template has the layout, the ABM metrics guide covers the downstream measures, and the pipeline velocity calculator turns conversion and days to first activity into a velocity figure.

The decision: MQA, MQL or opportunity-centric, and the steps to roll it out

The case against the MQA is not about scoring mechanics. It is about the unit being qualified. Forrester argued in 2022 that accounts are legal entities that do not make buying decisions. The buyer is a group of people inside the account working on one problem. A company that sells several products to the same account may have several such groups, each at a different stage. Forrester's view was that it's impossible to use one account stage (the MQA) to reflect the progression of multiple opportunities.

The alternative is the business-to-business Revenue Waterfall. Forrester describes it as a shift from managing individual leads to identifying, prioritizing, and advancing opportunities with connected buying groups. The tool it uses to find those opportunities is the opportunity map, a tool used to identify the potential opportunities that exist within an account.

The practical difference shows up at handoff. An account score that reads 15 people have engaged for a total of 642 engaged minutes does not tell a rep who to call. A count of interested people grouped by solution does.

We think the choice comes down to how many distinct buying decisions sit inside a typical target account. If it is one, an MQA loses little. If it is several, the account score blends them and sales misses the smaller ones.

Rollout checklist

1. Lead-to-account matching. Confirm every inbound contact resolves to an account record, and that unmatched contacts have an owner and a queue.
2. Stage definitions. Write one definition per stage and decide whether stages attach to the account or to an opportunity. In the Waterfall, the validate objective aligns with the engaged, prioritized, and qualified stages, and accelerate aligns with the pipeline stage.
3. Pilot segment. Pick one segment and one product line. Keep the rest of the business on the current model so you have a comparison.
4. Threshold test. Run the threshold against the pilot for a full sales cycle. For example, and not as a benchmark: a pilot of 30 accounts converting at 0.2 yields 6 opportunities. Compare that count with what the old model produced for the same segment.
5. SLA sign-off. Get sales leadership to sign the follow-up window, the disposition codes and the recycle rule before the pilot goes live, not after.

Review the pilot in your pipeline review meeting and decide on expansion from the conversion data, not from activity volume.

What this page does not cover

This page does not compare platforms, cover pricing or walk through setup inside any specific tool beyond the 6sense default rules it describes. It also leaves person-level lead scoring and intent data sources to separate guides. The service-level agreement template is an operating document, not legal advice, and the ABM figures quoted are not investment or financial advice.