Startup Usage Data: Spotting Accounts Ready to Upgrade
Some of your customers are already outgrowing their plan, and the proof is sitting in your usage data. They invite more teammates, hit limits, try locked features and log in every day. This guide shows you how to turn that raw activity into a short weekly list of accounts that are ready for a bigger plan, and how to reach out in a way that feels helpful rather than salesy.
Key takeaways
- Pick 4 to 6 usage signals that genuinely predict an upgrade, such as seats added, limits reached and locked features tried.
- Combine them into a simple points score you can calculate in a spreadsheet, then refine it once you see real results.
- Review a short list of high-scoring accounts every week and assign each one a clear owner and next step.
- Lead with the customer's own progress and a specific benefit, not a generic "time to upgrade" message.
- Track which signals led to real upgrades so your score gets sharper every quarter.
Late January is a good moment for this work. Many customers have just set their goals for the year, budgets are fresh, and the teams that will grow fastest in 2028 are already showing it in their first few weeks of activity. If you can spot them now, you can help them move to the right plan before they hit a wall and get frustrated.
This is not about squeezing more money from people who are not ready. Done well, upgrade outreach is a service: you notice a customer bumping into a limit and show them a better option. Done badly, it feels like surveillance.
A quick note before we start: how you collect and use customer data is shaped by your privacy notice, your contracts and laws such as UK GDPR and EU GDPR. This article is general information, not legal advice, so check your own obligations with a qualified professional if you are unsure.
Why usage data beats guesswork
Most small startups decide who to approach about an upgrade by gut feel. A founder remembers a friendly customer, or a salesperson works through the newest signups in order. That approach misses the quiet power users who never ask for help, and it wastes time on accounts that are nowhere near ready.
Usage data changes the question from "who do we like?" to "who is behaving like a customer who needs more?" It lets you:
- Find accounts early. You can reach out when a team is growing, not after it has already hit a hard limit and opened a support ticket.
- Focus a small team. If you only have a few hours a week for expansion work, a ranked list tells you where to spend them.
- Personalize the message. Knowing that an account added three people this month gives you something real to talk about.
You do not need a data team. A clean export, a spreadsheet and a weekly habit will get you most of the way.
The usage signals that usually matter
Every product is different, so treat this list as a menu. Your goal is to pick the handful of signals that line up with the reason people actually move up a plan in your business.
Growth signals
These show that the account itself is getting bigger.
- New users or seats added in the last 30 days.
- More projects, workspaces, locations or records created than usual.
- Invitations sent to people in other departments or with different job titles.
Limit signals
These show the account is pressing against the edges of its current plan.
- Usage at or above a set share of a plan limit, such as storage, contacts or monthly actions.
- Repeated warnings or blocked actions because a limit was reached.
- Workarounds, such as several free accounts from the same company domain.
Intent signals
These show curiosity about what the next plan offers.
- Clicks on locked features or "available on higher plans" prompts.
- Visits to the pricing page or plan comparison page while logged in.
- Support questions about features, security reviews or admin controls that sit on a bigger plan.
Health signals
These show the account is engaged enough to benefit from more.
- Active users as a share of total seats.
- Logins or key actions per week, compared with the same account last month.
- Use of your core feature, the one that delivers the main value of the product.
Pro tip
Look back at the last 10 to 20 accounts that upgraded on their own. Pull their activity for the 60 days before the upgrade and note what they had in common. Those shared behaviors are your strongest starting signals, and they come from your own customers rather than a generic list.
How to build a simple upgrade readiness score
A readiness score turns several signals into one number you can sort by. Keep it simple at first. A score you understand and trust beats a clever model nobody can explain.
- Choose 4 to 6 signals. Pick at least one from growth, limits and intent, plus one health signal. More than six makes the score hard to reason about.
- Set a threshold for each. Decide what counts as a hit. For example, "added 2 or more seats in 30 days" or "clicked a locked feature at least twice this month."
- Assign points. Give strong signals more points. A limit reached might be worth more than a single pricing page visit.
- Add a health gate. If active usage is low, cap the score or flag it for a check-in instead of an upgrade conversation.
- Calculate it weekly. Export the data, total the points and sort accounts from highest to lowest.
- Set a cutoff. Choose a score above which an account goes on this week's outreach list. Start with a cutoff that gives your team a manageable number of accounts.
- Review results every quarter. Compare the scores of accounts that upgraded with those that did not, and adjust points and thresholds.
If you already send product events to a tool such as Segment or HubSpot, you may be able to calculate some of this automatically. Check current features, pricing and privacy settings before you connect anything, and start with a manual version so you know exactly what the numbers mean.
An upgrade readiness score template
Copy this into a spreadsheet, one row per account. The points and accounts below are made up for illustration only.
| Account | Seats added in 30 days (3 pts) | Plan limit reached (4 pts) | Locked feature clicks (2 pts) | Pricing page visits (1 pt) | Active users healthy? (gate) | Total score | Next step |
|---|---|---|---|---|---|---|---|
| Example Co A | Yes, 4 seats | Yes, storage | Yes, 3 clicks | Yes | Yes | 10 | Account owner emails this week |
| Example Co B | No | Yes, contacts | No | Yes | Yes | 5 | Send limit guide, review next week |
| Example Co C | Yes, 2 seats | No | Yes, 2 clicks | No | No, low logins | 5 (capped) | Health check-in, no upgrade pitch |
Add a column for the result after 30 days, such as "upgraded," "not now" or "no reply." That one column is what turns the score from a guess into something you can improve.
Common mistake
Treating raw activity as readiness. An account with huge usage but one exhausted admin who is about to leave is not ready to spend more. Always pair growth and limit signals with a health check, and read the account notes before you reach out.
Running a weekly upgrade review
A score is only useful if someone acts on it. The simplest way to make that happen is a 30-minute weekly review with whoever handles sales, customer success and, ideally, one person from product.
- Start with the top of the list. Look at the 5 to 15 highest-scoring accounts. Fewer is fine for an early team.
- Sanity check each one. Read recent support tickets, notes and billing history. Remove anyone with an open complaint, a payment problem or a recent downgrade.
- Pick the right next step. Not every account needs a sales email. Some need a help article, a quick call or simply a note to check again next week.
- Assign an owner and a date. Every account on the list leaves the meeting with one name and one action.
- Review last week's outcomes. Spend five minutes on what happened with the accounts you contacted before.
If you are a solo founder, run the same review with yourself on Friday afternoon. Put it in your calendar so it survives busy weeks.
Reaching out without being pushy
The best upgrade messages feel like good customer service. They start with what the customer is trying to achieve, point to something specific you noticed, and offer a choice.
A few principles help:
- Lead with their progress. "Your team has grown to 12 people this month" is a compliment. "You have exceeded your plan" is a warning.
- Be transparent about what you noticed. Mention the general pattern, not a creepy level of detail. Customers are fine with you knowing seat counts and limits. They may not enjoy hearing which buttons they clicked at midnight.
- Name one clear benefit. Tie the higher plan to a problem they already have, such as admin controls for a growing team or higher limits for a busy season.
- Offer a low-pressure next step. A short call, a comparison of plans or a trial of the feature works better than a hard close.
- Respect "not now." Note the reply and set a reminder for a later date instead of following up every few days.
Product prompts can help too. Save personal emails for larger accounts or for cases where the right plan is not obvious.
Hypothetical example
Imagine Priya runs a small scheduling software startup with four staff. Her weekly score flags a physiotherapy clinic that added three new practitioners in January, reached its booking limit twice and clicked the locked multi-location feature. The clinic's notes show friendly support chats and no billing issues. Priya writes: "Congratulations on growing the team this month. I noticed you hit the booking limit a couple of times, and it looks like you might be opening a second site. Our next plan raises the limit and adds multi-location calendars. Happy to walk you through it in 15 minutes, or I can send a quick comparison if that is easier." The clinic asks for the comparison and upgrades the following week. The details are made up, but the pattern is common.
Keeping the data honest and the customer's trust intact
Upgrade scoring only works if the data is reliable and customers feel respected. A few habits keep both in good shape.
Check your data quality
Make sure events are tracked consistently. If a release changed how an action is logged, scores may jump for every account, so keep a short change log of tracking changes.
Use data in line with your privacy notice
Your privacy notice should explain, in plain language, that you use product usage data to improve the service and to tell customers about relevant plans or features. In the UK and EU, check that you have a suitable lawful basis for this use and that marketing emails follow consent rules where they apply. In the US, commercial emails need a clear way to unsubscribe and a valid physical postal address under CAN-SPAM.
Close the loop with product
If many accounts hit the same limit long before they are ready to pay more, share that pattern with whoever owns pricing.
Your upgrade readiness checklist
- Pull activity for recent self-serve upgrades and list their shared behaviors
- Choose 4 to 6 signals across growth, limits, intent and health
- Set thresholds and points for each signal in a spreadsheet
- Add a health gate so low-engagement accounts get a check-in instead
- Book a 30-minute weekly upgrade review with a clear owner
- Write two or three message templates that lead with customer progress
- Record the outcome for every account you contact
- Confirm your privacy notice and email practices cover this use
- Review scores and outcomes every quarter and adjust the points
Frequently asked questions
How many accounts should be on the weekly list?
Only as many as your team can contact properly that week. For a solo founder that might be three to five accounts. For a small sales or success team it might be 10 to 20. A short list handled well beats a long list nobody gets through.
What if we do not track product events yet?
Start with what you already have: seat counts, plan limits, billing data and support tickets. Even those few fields can produce a useful first score. Add event tracking for your core feature and locked feature clicks next, since they are often the most telling signals.
How do we know if the score is working?
Compare upgrade rates for accounts above your cutoff with those below it over a few months. If high scorers upgrade noticeably more often, the score is useful. If not, revisit your signals, starting with the accounts that upgraded but scored low.
The bottom line
Your customers are already telling you when they need more from your product. By picking a handful of meaningful signals, scoring accounts in a simple spreadsheet and reviewing the top of the list every week, you can reach the right people at the right moment with a message that genuinely helps them.
Set aside an hour this week to study your last few upgrades, choose your signals and build the first version of your score. Then run your first weekly review, record what happens, and let real results shape the next version.
