How incident.io cut account duplicates by 70% with DataAgent

The challenge: when bad data becomes a business blocker

Most teams don’t have a data problem. They have a reality problem.

Their CRM has records (contacts, accounts, activity history) all sitting in Salesforce, looking like a system of truth. The problem is that the system stopped reflecting reality a long time ago. People changed jobs, duplicates accumulated, and contacts sat unmatched.

For modern GTM teams leaning into AI and automation, the gap between what the CRM says and what's actually true is getting more expensive by the day.

For Khaled AlSaleh, RevOps Leader at Incident.io, this was the thing creating friction in GTM execution every day.

The root problem: poor data burdens the entire team

Incident.io is a fast-growing company, and Khaled was responsible for keeping the GTM system clean enough for the team to act on as a team of one. No dedicated data function, or engineering support. Just a RevOps leader trying to hold a system together.

The problems weren’t complicated. They were the kind every RevOps leader recognizes immediately: account duplicates accumulating faster than they could be resolved, contacts going stale as people changed jobs, and a growing pool of 19,000 contacts sitting in an “Unknown Company” bucket that couldn’t be matched to any account.

That last problem had a sharp downstream edge. Contacts without a matched account don’t just sit idle, they get routed to the wrong reps and flagged as dead leads. When Khaled eventually started working through that pool, he found contacts at significant target companies that the team had been missing entirely.

Khaled was determined to build a trustworthy system his team could actually execute from.

The solution: DataAgent’s automated approach

With DataAgent, teams move from manually hunting down bad data to a system that catches it automatically so sellers, automation, and AI workflows can move without questioning the data underneath.

How it works

DataAgent is an execution layer powered by Common Room’s identity resolution engine. The same engine understands who a buyer actually is across fragmented records, job changes, and multiple data sources.

Because Common Room already understands identity at the person and account level, DataAgent doesn’t need admins to configure rules or logic before surfacing issues. It already knows where the system has drifted from reality, and the outputs are specific enough to act on immediately.

This means teams get three things with DataAgent from day one:

• Confidence in who they’re reaching out to: Flags contacts with outdated CRM records so reps don’t waste their time on people who’ve already moved on.
• A single, trusted view of each buyer: Identifies overlap when the same person exists across multiple records, so routing, attribution, and AI workflows operate on one clean signal.
• Accounts that reflect the real world: Automatically surfaces duplicate accounts so forecasting, territory planning, and segmentation are built on an accurate foundation.

For Khaled, this underlying identity layer was what separated DataAgent from every tool he’d evaluated before.

“The power is in the richness of your data and in the context that you have. The organization structure gives us the power to see whether the person is actually associated with the account, and whether two accounts are actually part of the same organization. If you think about any other platform, that would be quite difficult to achieve.” — Khaled AlSaleh, Incident.io

That context — identity resolved, relationships mapped, signals interpreted across systems — is what makes DataAgent’s outputs trustworthy enough to act on. Other tools ask teams to define the problem before they can find it. DataAgent finds it first.

“It’s something that seemed almost unachievable. You can debate whether or not other tools can do this, but achieving what you want often involves a lot of fiddling and experimenting. Here we get data we can trust, and what’s more, it alerts us when a contact is out of sync, and we can just update it.” — Khaled AlSaleh, Incident.io

The impact: less time fixing, more time executing

Measurable improvements at Incident.io

The results were immediate and operational. Since adopting DataAgent, Incident.io saw concrete movement in the metrics that matter most:

• Account duplicates dropped from ~3% to ~0.8% across 140,000+ accounts, and continue to trend lower as new duplicates are caught on a rolling basis.
• 3,700 of 19,000 “Unknown Company” contacts were matched to real accounts, recovering leads from significant target companies whose signals had been silently lost in routing.
• Cleanup time collapsed from days or weeks to minutes — with no dedicated headcount required to sustain it.

“When we started the exercise, about 3% of our accounts were duplicates. Today it’s about 0.8. That is material. What’s buried in this statement is that we can now size the problem. In the past we relied on intuition, whilst now I have a strong view of how many duplicate accounts and contacts we have, and how many outdated contacts need to be updated. ” — Khaled AlSaleh, Incident.io

Eliminating the need for dedicated headcount

One of the most consequential outcomes for Incident.io was avoiding a hire that would have otherwise been necessary to manage data quality at scale. Khaled was explicit about what he didn’t want: a full-time platform manager.

What he needed was a system that did the work, one that someone could log into, action in minutes, and trust was running correctly in the background.

“What I don’t want to do is hire someone full time to manage a platform. What I don’t mind doing is having someone log into Common Room and clean up the data. We’re talking about minutes of work to achieve this. If you think about the impact it’s going to have downstream and the speed at which we can move, it’s tremendous.” — Khaled AlSaleh, Incident.io

Days of work, down to minutes

When asked to quantify the time savings, Khaled put it plainly:

“I wouldn’t say minutes versus hours. I’d say minutes versus days — or weeks.” — Khaled AlSaleh, Incident.io

That delta isn’t just about RevOps efficiency. It’s about whether contact-level cleanup happens at all. For most teams, it simply doesn't happen because the cost and effort make it impractical.

DataAgent changes that math entirely.

A stronger foundation for modern go-to-market

As businesses lean further into AI and automation, reliable data stops being a nice-to-have. It’s the foundation every workflow depends on. When the system is accurate, teams can prioritize better, segment more precisely, and execute with confidence.

When it isn’t, AI workflows don’t just underperform, they amplify the problem, routing bad signals and compounding stale assumptions at scale.

Khaled described the ability to launch plays and take action on contacts and accounts with greater confidence — not because the tool promised better outcomes, but because the data underneath was finally reliable.

“I can definitely do it with greater confidence.” — Khaled AlSaleh, on launching plays and executing on accounts and contacts

CRM data has always degraded over time. Teams have always worked around it. For a long time, that was just part of operating in GTM.

But the way teams execute has changed. When AI, automation, and modern workflows all depend on accurate data, “working around it” stops being an option. The system is either accurate enough to act on, or it isn’t.

DataAgent is built for that shift, not as a one-time fix, but as a way to keep the system reliable over time. Either your data reflects reality, or it doesn't. That gap is what determines whether GTM actually works.