What is an account engagement score? It is one number that sums up activity from everyone at a company, not just one lead. It weighs how many roles are engaging, how serious that activity is, research happening off your site, and how well the account fits. As a result, leadership sees which companies are moving toward a deal.
Definition: An account engagement score is a weighted roll-up of activity across all known contacts at one company, used to judge how close the whole buying group is to a purchase.
Updated November 2026
What an MQL Count Can’t Tell You
For example, picture a quarterly review. Marketing reports 1,200 MQLs, up 30%. Then the CRO asks which companies are closer to buying. Yet nobody can answer.
That gap exists because MQLs count people, while groups make buying decisions. As a result, it creates three blind spots:
- Volume hides concentration. Fifty MQLs from one company can look like fifty opportunities.
- Silent buyers go uncounted. A CFO who reads your pricing page but never fills in a form adds nothing to the MQL count.
- The count says nothing about readiness. An MQL shows interest, not progress toward a decision.
The research backs this up. Forrester’s waterfall benchmarks put conversion from inquiry to closed deal in a lead-centric process at less than 1%. In other words, fewer than one win for every hundred interested people. For the difference between the two lead stages, see our guide to MQL vs SQL.
The Four Inputs of an Account Engagement Score
The score combines four weighted inputs. Together, they describe the whole buying group rather than its loudest member.
| Input | What it measures | Example signals | Suggested weight |
| Breadth | How many roles are engaging | Visits or replies from finance, IT, and operations | 35% |
| Depth | How serious the activity is | Pricing page views, demo requests, integration docs | 30% |
| Intent | Research happening off your site | Third-party topic surges in your category | 20% |
| Fit | How closely the account matches your ideal customer profile | Industry, size, tech stack | 15% |
Breadth carries the most weight on purpose, because deals with one engaged person are fragile. That is why single-threaded deals fail so often. You can adjust the weights for your market, but agree with them with sales first.
Two Accounts, Two Different Stories
For example, consider two accounts. Account A produced 14 MQLs, but 12 came from one marketing coordinator. Account B produced only 3 MQLs, yet six people across four departments engaged, including the CFO viewing pricing.
| Measure | Account A | Account B |
| MQLs | 14 | 3 |
| Breadth (out of 35) | 10 | 30 |
| Depth (out of 30) | 8 | 24 |
| Intent (out of 20) | 0 | 16 |
| Fit (out of 15) | 15 | 12 |
| Total score (out of 100) | 33 | 82 |
So by MQL count, Account A looks like the priority. In reality, Account B is far closer to buying.
Account Engagement Score Point Values, Decay, and Caps
Weights tell you what matters. Point values, on the other hand, tell you how to score each action. Here is a simple starting model:
| Signal | Points | Counts toward | Decay |
| New role engages (first visit or reply from a new department). | 6 per role | Breadth | Halve after 30 days |
| Pricing page view | 8 | Depth | Half after 14 days |
| Demo request or meeting booked | 12 | Depth | Halve after 30 days |
| Integration or security docs viewed | 6 | Depth | Half after 21 days |
| Blog or newsletter visit | 1 | Depth | Drop after 14 days |
| Third-party topic surge | 10 | Intent | Drop when the surge ends. |
| Strong fit with a closed-won pattern | Up to 15 | Fit | No decay |
Two rules keep the score honest. First, cap points per contact, for example, at 15. That stops one busy reader from inflating the whole account. Second, decay every activity signal. Clearly, a pricing visit from yesterday should count far more than one from last month.
Not sure which signals deserve the most points? Our list of buying signals ranked by strength is a useful starting point.
Build Your First Version in a Spreadsheet
You do not need a new platform to start. In fact, most teams can build a working first version in an afternoon.
- Export activity by account. Pull the last 90 days of web visits, email replies, meetings, and form fills from your CRM and marketing tools, grouped by company domain.
- Tag each contact with a role. Use job titles to sort contacts into finance, IT, security, operations, and end users. As a result, breadth becomes countable.
- Apply the point table. Add points for each signal, then apply decay based on the activity date.
- Add intent and fit. Next, bring in topic surges from your intent provider and a simple fit score from your ideal customer profile.
- Check against reality. Finally, compare the top 20 scores with your open opportunities. If the lists barely overlap, adjust the weights before you share anything with sales.
Then, once the model holds up for a month, move it into your CRM or ABM platform so it updates on its own.
Turning the Account Engagement Score Into Action
In practice, a score is only useful when each band has an owner and a next step. That way, handovers carry account context instead of single leads.
| Score | Stage | Owner | What happens |
| 0 to 39 | Aware | Marketing | Nurture with relevant content and ads. |
| 40 to 59 | Engaged | SDR | Research the account and map the buying committee. |
| 60 to 79 | Active | SDR and account executive | Reach out to several roles, not just the most active contact. |
| 80 to 100 | Ready | Account executive | Create an opportunity and attach every known stakeholder. |
A Weekly Account Engagement Dashboard
Above all, keep the weekly view short. Five columns are enough for a pipeline meeting:
| Column | What it shows |
| Account and tier | Who you are looking at |
| Score this week vs last week. | Whether the account is warming or cooling |
| Roles engaged | How many departments are active, by name? |
| Top signal this week | The single strongest action, with its date |
| Owner and next step | Who acts, and what do they do by Friday? |
Also, sort the list by change in score, not by total score. Accounts that jumped this week usually deserve attention first.
Changing What You Report to Leadership
After all, leadership does not need more metrics. Instead, it needs metrics tied to revenue.
| Instead of reporting | Report this. |
| MQLs generated | Accounts reaching Active or Ready |
| MQL-to-SQL conversion | Active-to-opportunity conversion |
| Cost per MQL | Cost per engaged account |
| Leads by source | Pipeline value from engaged accounts |
These numbers also make it easier to report ABM ROI to finance because they connect spend to accounts and pipeline.
Switching Without Breaking Things
However, targets, tools, and compensation are often built around MQLs, so a hard switch can backfire. A phased approach, based on Forrester’s crawl, walk, and run approach, works better:
- Run both for one quarter. Keep MQL reporting while you calculate account scores in parallel.
- Test which predicts better. At quarter end, check which metric better predicted the opportunities that were created.
- Shift targets gradually. Move marketing goals toward engaged accounts, and align definitions with sales.
According to Forrester, leaving MQLs behind is “no longer a question of if, but when.”
Where an Account Engagement Score Goes Wrong
- No time decay. Old activity keeps accounts looking warm long after interest has faded.
- Counting the wrong visitors. Filter out customers, job seekers, partners, and competitors.
- One person inflating the score. Cap points per contact.
- Building it without sales. If sales does not trust the score, nobody acts on it.
That said, MQLs still have a place. They work well for individual follow-up and for judging content performance. They just should not be the headline metric.
How ColedaB2B Helps
As a partner, we run account-based experience programs that reach the whole buying group, not one contact. Our own network of seven B2B publishers adds first-party engagement signals from your target accounts, which feed straight into breadth and intent scores.
Want to see which of your accounts are actually engaging? Share your target list. Then we will show which accounts are active across our network and how many roles are involved in each.
Get a target account engagement plan
FAQs:
It is a single number that combines activity from everyone at one company. It reflects how many roles are engaging, how serious the activity is, research happening off your site, and how well the account fits your ideal customer profile.
An MQL measures one person’s interest. By contrast, the score measures the whole buying group’s activity, so it shows which companies are moving toward a decision rather than which individuals downloaded something.
Assign weights to breadth, depth, intent, and fit; give each signal a point value; then add decay and per-contact caps. A common starting split is 35, 30, 20, and 15, adjusted to your market.
It depends on your model. In a 100-point system, many teams treat 60 and above as ready for sales outreach and 80 and above as ready for an opportunity.
Not straight away. Run both for a quarter, compare which one better predicts opportunities, and then shift targets gradually toward engaged accounts.