Using AI to Identify Upsell and Expansion Opportunities

What Is Expansion Revenue?

Before diving into the how, it's worth being clear on the what.

Expansion revenue is revenue generated from existing customers beyond their initial contract. It typically comes in three forms:

• Upsell: The customer upgrades to a higher tier, plan, or package (e.g., moving from a Basic to Pro subscription)
• Cross-sell: The customer buys an additional product or service (e.g., adding a marketing module to a CRM subscription)
• Seat expansion: In per-seat SaaS models, the customer adds more users

Collectively, this is sometimes called Net Revenue Retention (NRR) or Net Dollar Retention (NDR) — the percentage of revenue retained from existing customers after accounting for churn, downgrades, and expansion. Businesses with NRR above 100% are growing from their existing base alone, before a single new customer is acquired.

For SMBs, expansion revenue is often the most profitable revenue type: no acquisition cost, no lengthy sales cycle, no cold outreach.

Your Best Sales Leads Are Already Customers

If you're only looking for growth in new logo acquisition, you're working harder than you need to.

The data on this is consistent: selling to an existing customer has a 60–70% success rate. Selling to a new prospect? Around 5–20%. And yet most SMBs spend the bulk of their sales and marketing budget chasing cold leads — while their warmest, highest-intent prospects sit in the customer database, going unnoticed.

The reason this happens isn't laziness. It's bandwidth. Without a dedicated account management team, it's almost impossible to manually monitor hundreds of customer accounts for expansion signals. Things get missed. Upsell moments pass. Customers who were ready to buy more quietly upgrade with a competitor, or just don't bother.

AI changes this. With the right setup, you can automatically monitor your entire customer base, surface the accounts most likely to expand, and trigger the right outreach at the right time — all without hiring an account team.

The Signals That Indicate Expansion Readiness

AI identifies expansion opportunities by monitoring signals that humans can't track at scale. Here's what to watch for:

Usage Signals

The most reliable expansion signal is product usage — specifically, when a customer is bumping up against the limits of their current plan or heavily adopting features associated with higher tiers.

• Usage caps approaching: A customer who's used 90% of their monthly API calls or storage is a natural candidate for an upgrade conversation
• High feature adoption: A customer who's using eight out of ten core features is more likely to need advanced functionality than one who's only using two
• Frequent logins and high session time: Active users are getting value — and engaged customers buy more
• Power user identification: If one or two people within a customer account are using the product intensively, they're potential expansion champions

Team and Org Signals

• New hires in the customer's team: A growing team often means more seats or broader use cases
• New office locations or market expansion: Companies scaling their operations often need to scale their software too
• Leadership changes: A new decision-maker sometimes means a fresh look at tools — and an opening for an expansion conversation

Engagement Signals

• Asking about features they don't have access to: This is a direct expansion signal — the customer is actively wanting more
• High NPS scores: Promoters are more likely to expand than passives or detractors
• Positive support interactions: Customers who are happy with your support team are warm for commercial conversations

Commercial Signals

• Approaching renewal: Renewal conversations are natural moments to discuss expanding the relationship
• Contract anniversary: A year-in review is an organic touchpoint to discuss growth
• Recent successful outcomes: If a customer just hit a milestone using your product, they're in the ideal mindset for an expansion discussion

How AI Monitors and Scores These Signals

Rather than manually reviewing each account, AI tools aggregate these signals and score accounts by expansion likelihood. Here's how the main platforms approach it:

HubSpot (CRM + Sales Hub)

HubSpot's CRM tracks contact and company-level activity — email opens, website visits, product usage (via integrations), and deal history. With HubSpot's AI-powered predictive lead scoring, you can score existing customers using the same engine used for prospects, adapting it to expansion criteria rather than acquisition signals.

Set up a custom property for "expansion score" and create a workflow that updates it based on usage data (pulled via HubSpot's API or a tool like Zapier) and engagement metrics.

Cost: Sales Hub Starter from ~£18/month

Breadcrumbs.io

Breadcrumbs is specifically built for revenue acceleration and scores both prospects and existing customers. It connects to your CRM, product analytics, and marketing data to build a unified expansion score. One of its strengths is that it allows you to define what "expansion-ready" looks like for your specific business model — useful if your upsell triggers are non-standard.

Cost: From ~$100/month

Salesforce Einstein (for Salesforce users)

If you're on Salesforce, Einstein AI can score accounts for expansion likelihood, recommend next actions, and surface at-risk accounts in the same view. It's more powerful but also more complex to configure — better suited to SMBs that have a Salesforce admin or operations resource.

Gong

Gong analyses recorded sales and CS calls using AI to detect when customers express interest in features they don't have, frustration with current limits, or intention to expand. It can automatically flag these moments so your team doesn't have to manually review every call.

Cost: Custom pricing, typically mid-market

Mixpanel or Amplitude (Product Analytics)

If your product has digital touchpoints (SaaS, app, portal), Mixpanel and Amplitude can track feature usage and cohort behaviour in detail. You can build dashboards that flag accounts hitting usage thresholds or exhibiting power-user behaviour patterns associated with expansion.

Cost: Mixpanel starts free; Amplitude from ~£400/month for growth plans

Building Your Expansion Playbook

Identifying expansion-ready accounts is only half the job. You need a playbook for what happens next.

Step 1: Define Your Expansion Triggers

Start by mapping your product or service to natural expansion moments. For each trigger, define:

• What signal fires it (e.g., 80% usage cap reached)
• Who gets notified (e.g., account owner or CS manager)
• What the outreach looks like (e.g., a personal email, a call, an in-app message)
• What's offered (e.g., a trial of the next tier, a commercial proposal)

Step 2: Segment Your Expansion Candidates

Not all expansion-ready accounts are the same. Segment by:

• Revenue potential: What's the realistic uplift from upgrading this account?
• Relationship warmth: Has your team had recent positive interactions?
• Decision-maker access: Do you have a direct line to someone who can approve an expansion?

High potential + warm relationship + decision-maker access = your top expansion priority.

Step 3: Personalise the Outreach

Generic upsell emails perform poorly. AI tools like HubSpot's email personalisation, Copy.ai, or even ChatGPT with CRM data can help you draft outreach that references the customer's specific usage, their outcomes, and what the next tier would unlock for them.

Example personalisation:

"Hi , I noticed your team has been using heavily over the past few weeks — you're approaching your monthly limit. I wanted to reach out before it became a blocker for you, and share what would unlock..."

This is far more effective than a standard "ready to upgrade?" message.

Step 4: Track and Optimise

Measure your expansion playbook like you'd measure a sales funnel:

• How many accounts enter the "expansion-ready" segment each month?
• What's the conversion rate from outreach to expansion?
• Which triggers have the highest conversion rate?
• What's the average uplift per expansion?

Use this data to refine your scoring model over time. The accounts you convert (and don't convert) tell you which signals are most predictive.

A Practical Example: How This Might Work at a 50-Person SaaS Company

 Here is an example I created so you can see how this works. Imagine you run a project management SaaS with 300 customers. You have two CS people and no dedicated account management team.

Without AI:

• CS team manually reviews accounts when they have time (rarely)
• Renewal conversations happen reactively, often too late
• Upsell conversations are ad hoc and inconsistent

With AI:

• Mixpanel flags 12 accounts approaching usage limits this month
• HubSpot's expansion score highlights 8 accounts with high NPS and recent product activity
• Gong flags 3 accounts where customers asked about features on the next tier
• The CS team has a prioritised list of 23 accounts to contact — with context and a clear playbook for each

Instead of guessing where to focus, your team has a clear, data-driven list. And the accounts that aren't on the list? They're being monitored automatically. Nothing slips through.

Getting Started This Week

You don't need to implement everything at once. Here's a minimal starting point:

1. Audit your product data: What usage metrics do you currently track? If none, set up Mixpanel or integrate your product with your CRM.
2. Define three expansion triggers: Pick the signals most relevant to your business model. Start simple.
3. Build a basic segment in your CRM: Create a smart list of accounts that meet your expansion criteria.
4. Draft one personalised outreach template: Test it with your top five expansion candidates.
5. Review results in 30 days: Which accounts responded? Which converted? Use that to refine your triggers.

The Bottom Line

Every business has expansion revenue sitting in its customer base, waiting to be unlocked. AI doesn't create these opportunities — it makes them visible and actionable, at a scale and speed that would otherwise require a full account team.

For SMBs competing with larger players, that's a meaningful advantage.

Related reading:

• AI-Powered Customer Health Scoring for Small Teams →
• How to Use AI to Reduce Customer Churn →
• Automating Customer Onboarding with AI →