Overlapping SaaS tools: find and consolidate them
Two applications overlap when they do the same job for the same people. To find them, group the inventory by job, then compare the seat lists of each pair by email address. The people who hold a paid seat in both are the first candidates. Their number, times the price per seat of the tool to drop, times twelve, is an upper bound and not yet a saving. Ask them which tool they use, decide which one stays, migrate, and end or reduce the other contract at its notice date.
Two applications overlap when they do the same job for the same people. You find them in two passes: group everything the organisation pays for by job, then compare the seat lists of every pair that shares a job. What the comparison gives you is a list of people who hold a paid seat in both, and they are where consolidation starts.
What counts as an overlap
| Case | Overlap? |
|---|---|
| The same people hold paid seats in two tools for the same job | Yes. The clearest case |
| Two teams each use their own tool for the same job | Yes, but someone has to migrate. A larger decision |
| A suite already paid for includes a feature, and a separate tool is bought for it | Yes, if the included feature is good enough for those who use it |
| Two tools in the same category that serve different work | Not necessarily. The category says where to look; the people who hold both say whether it is a duplicate |
| One paid tool and one free plan | Not a cost question. Possibly a data question |
Step 1: group the inventory by job
Write next to every paid application the job it does, in two or three words. Categories in which organisations commonly end up with two tools:
- video meetings
- team chat
- project tracking
- documents and wiki
- design
- customer relationship management
- customer support
- monitoring
- AI coding assistants
- AI chat assistants
The inventory has to be complete for this to work, and the second tool is exactly the one that tends to be missing from it: bought by one team, paid by card. Build the list from invoices and card statements, as described in spend analysis basics, and not from what IT administers.
Step 2: compare the seat lists
For each pair that shares a job, take the billable seats of both applications and match them by email address. Count three groups: people only in the first, people only in the second, people in both.
Check two things before trusting the count. People can have different addresses in the two tools, such as an alias or an old domain. And external guests appear in seat lists without being employees.
Step 3: compute the upper bound
Decide provisionally which tool would stay. The larger contract is the usual starting assumption. Then:
Upper bound per year = people with a paid seat in both × price per seat per month of the tool to drop × 12.
A worked example with invented figures: tool A has 120 paid seats, tool B has 40 at 15.00 a month. 25 people hold a seat in both. 25 × 15.00 × 12 = 4,500.00 a year.
This is an upper bound. It assumes that every one of the 25 can give up tool B, which nobody has confirmed. Label it that way wherever it is shown, and keep it out of any total of savings.
Step 4: ask the people who have both
Ask each of them one question: which of the two do you use, and what for? The answers fall into three groups.
| Answer | What follows |
|---|---|
| I only use one of them | The other seat can be released. This is an ordinary unused seat |
| I use both for the same thing, depending on who invites me | A candidate for consolidation once a decision is taken |
| I need the second one for something the first cannot do | Write the reason down. It is a requirement for the decision |
How to word the question and follow up is covered in asking employees about their licences.
Step 5: decide which tool stays
| Criterion | Question |
|---|---|
| Use | How many people use each, and how many only that one? |
| Requirements | What did people say the other tool cannot do? |
| Cost | What does a seat cost in each, on the tiers actually needed? |
| Connections | What is each tool wired into: single sign-on, calendar, ticketing, reporting? |
| Data | What has to move, and is there an export and an import? |
| Contract | Which notice date comes first, and can seats be reduced before the end? |
| Outside parties | Do customers or partners require one of them? |
The decision belongs to the owners of both applications together. A decision taken by procurement alone is reversed by the first team that loses a workflow.
Step 6: migrate, then end the contract
- Set the date from the contract: the notice date of the tool to drop. See preparing a SaaS renewal.
- Move the people who have both first. They already know the tool that stays.
- Move the data that is needed and archive the rest.
- Give notice or reduce the seats, in the form the contract requires.
- Record what was removed. Seats removed, times price, times twelve is the saving; the upper bound from step 3 was the question.
If full consolidation is not worth a migration, the partial result still stands: everyone who said they use only one tool gives up the other seat.
What goes wrong
- The upper bound is reported as a saving. It shrinks with the first answers.
- The cheaper tool is kept by default. If it lacks what a team needs, that team buys a third tool.
- Migration cost is left out. Moving years of project history is work, and it belongs in the comparison.
- The contract date is missed. The migration finishes one month after the renewal, and both tools are paid for another term.
- Only administered tools are compared. The duplicate nobody in IT knows about is the one to look for.
How Cureman does it
Cureman keeps a list of applications that do the same job, by category. When an organisation pays for two of them, it takes the one with the larger yearly cost as the one to keep, matches the billable seats of the other against it by email address and shows per application the paid seats, the yearly cost and how many people also hold a seat in the first. The figure is computed as people with both, times the price per seat of the smaller tool, times twelve. It is labelled as an upper bound, with the assumption that everyone with both can give up the smaller one, and it has low confidence. The finding recommends asking those holders which tool they actually use, and a question to them goes out only after a person approved it. See the page about licence optimisation.
Questions and answers
What are overlapping SaaS tools?
- Two or more paid applications that do the same job in the same organisation, for example two video meeting tools or two project trackers. The overlap that costs money is the group of people who hold a paid seat in both. Two teams that each use a different tool are a consolidation question too, but a larger one, since somebody has to move.
How do you find duplicate SaaS applications?
- List every paid application with the job it does, and look for jobs with more than one entry. For each pair, compare the two seat lists by email address and count the people who appear in both. Applications paid by card or expense claim are often missing from the inventory, so the list should start from invoices and card statements.
How is the saving from consolidating two tools calculated?
- The number of people with a paid seat in both tools, times the monthly price per seat of the tool to be dropped, times twelve. That is an upper bound, because it assumes everyone with both can give one up. The saving that counts is the seats actually removed at the contract's renewal, less what the migration costs.
Which of two overlapping tools should be kept?
- The one that more people rely on for work that the other cannot do, not automatically the cheaper one. Compare who uses which and for what, what it costs per seat, what each is connected to, how hard the data is to move, and whose contract can be ended first. The users' answers decide more than the feature lists.
When is it right to keep two tools for the same job?
- When they serve different needs that one tool does not cover, for example an engineering team and a marketing team with different workflows, or when customers or partners require one of them. Then the question becomes who needs a seat in both. Most people need one.