# How we audit our SaaS spend for waste

The SaaS spend audit agent we run on Kortix — reconciles recurring card and bank charges against our subscription register every week and flags duplicate tools, unused seats, price hikes, and shadow IT, recommending only — it never touches a subscription.

Canonical page: https://openchatly.com/use-cases/saas-spend-audit

SaaS spend sprawls quietly. A team trials a tool and keeps paying for it after
the trial ends. Two teams buy overlapping products without knowing it. A vendor
raises its price and the invoice just changes. A card gets charged for
something that was never logged anywhere. None of this shows up until someone
sits down with a spreadsheet and a list of bank transactions, which happens
rarely because it's tedious and the spreadsheet is usually stale by then.

We run a SaaS spend audit agent on Kortix that reconciles our card and bank
charges against our subscription register every week and posts what it finds
to Slack. It never touches a subscription — it only ever recommends. This is
how we keep our SaaS spend honest.

- **Team:** Kortix
- **Runs on:** Weekly cron
- **Connected systems:** Plaid · Google Sheets · Slack
- **Mode:** Read-only · recommend-only · never cancels

## The problem

The two records that would catch SaaS waste — what's actually being charged
and what's supposed to be — live in different places and drift apart. The bank
feed shows every recurring charge; the subscription register shows what
finance thinks it's paying for. A tool bought on a personal card during a
trial never makes it into the register. A price increase shows up as a bigger
number on the same line, easy to miss next to a hundred other charges.

A quarterly spend review catches some of this, but it's a point-in-time
snapshot done by hand, and three months is a long time for a duplicate tool or
a stale renewal to keep charging. Anything faster than quarterly means someone
doing the same manual comparison every week, which doesn't happen because
nobody has a spare afternoon every week.

## What we built

On Kortix, a weekly **cron** re-prompts a persistent agent **session**. It
reads recurring card and bank charges from Plaid, reads the subscription
register we keep in Google Sheets, and reconciles the two: duplicate or
overlapping tools, seats that look unused, price hikes since the last time it
checked, renewals coming up soon, and shadow IT — charges with no matching row
in the register at all. It posts what it finds to Slack, remembering what it
already flagged so the same unresolved item doesn't repeat every week.

## How it works

### Copy the setup prompt into a new session

Open your Kortix workspace, start a new session, paste this prompt and send it.
The agent writes the agent file, the subscription-audit skill and the weekly
trigger, and opens a change request for you to review. Prefer clicks? Choose
**Use this template** instead.

**Setup prompt**

```
Set up a weekly SaaS spend audit agent in this workspace.

1. Create the agent agents/saas-spend-audit.md. Its job is to reconcile recurring charges against the subscription register and recommend where to cut. It may use the Plaid, Google Sheets and Slack connectors. Plaid and Google Sheets are read-only.

2. Create the skill skills/subscription-audit/SKILL.md with these rules:
   - Pull card and bank transactions for the last 60 to 90 days from Plaid. A charge is recurring only when it repeats by merchant and amount.
   - Match each recurring charge to a row in the SaaS Subscriptions sheet by vendor and amount. A recurring charge with no row is shadow IT.
   - Flag duplicate or overlapping tools, unused seats, price rises above 5%, and renewals in the next 30 days.
   - Keep the last price and the last recommendation for each vendor in memory/saas-spend-audit-log.md. Report only new or changed findings.
   - Post one report to #finance per run, grouped by signal, with the evidence and a suggested action on each line.
   - Recommend only. Never cancel, pause, downgrade or change a subscription or a payment method, and never write to Plaid or the sheet.

3. Add a cron trigger weekly-spend-audit to kortix.yaml. It runs the saas-spend-audit agent every Monday at 09:00 with the prompt "Load the subscription-audit skill and run this week's SaaS spend audit."

4. Tell me which of these connectors are not connected yet.

5. Open a change request with all of it. Do not merge it.
```

### Connect your tools

Open **Connectors** in your workspace and connect your bank or card data, the
sheet that holds your subscription register, and Slack. Credentials are
brokered server-side and never enter the agent's machine.

- **Plaid** (Required):
Pulls recurring card and bank charges, read-only.
- **Google Sheets** (Required):
Reads the subscription register to reconcile charges against.
- **Slack** (Required):
Posts the weekly digest of flags and recommendations.

### Run it once with this prompt

Merge the change request, then start a session with the **saas-spend-audit**
agent and send this prompt. Check its matches against the register before you
let the schedule take over.

**Prompt**

```
Run this week's SaaS spend audit.

1. Read memory/saas-spend-audit-log.md for what you already flagged.
2. Pull recurring card and bank charges for the last 90 days from Plaid.
3. Match them to the SaaS Subscriptions sheet. Find duplicate tools, unused seats, price rises above 5%, renewals in the next 30 days, and charges with no row.
4. Post a ranked report to #finance with the evidence and a suggested action for each finding. Skip anything unchanged since last week.
5. Update the ledger. Do not cancel or change any subscription.
```

### Let the weekly schedule take over

The change request added this trigger to `kortix.yaml`. From now on the agent
runs every Monday at 09:00, each run on its own cloud computer. It reads its
ledger first, so an open item from last week does not come back as new. To
change the time, edit the cron line, or open **Triggers** in the workspace.

**kortix.yaml**

```yaml
triggers:
  - slug: weekly-spend-audit
    name: Weekly SaaS spend audit
    on:
      cron: "0 0 9 * * 1"
    run:
      agent: saas-spend-audit
      prompt: Load the subscription-audit skill and run this week's SaaS spend audit.
```

### Recommend, never cancel

Plaid and the register are read-only, and the agent has no write access to
billing. Every line in the report is a suggestion. A person decides what to
cancel, downgrade or renegotiate, and does it.

### Read the weekly report

Each run ends with one report in the channel. This is what a run looks like in
the session:

> **Weekly SaaS spend audit · Mon 09:00**
>
> Prompt: Load the subscription-audit skill and run this week's SaaS spend audit.
>
> Reconciled **47** recurring charges against the register. Found **3** new
> items: **2** charges with no register row, and a design tool whose price rose
> **12%** this month. An annual plan renews in **21** days, and the suggested
> action is to confirm its seat count before then.

> **The pattern**
> A weekly **cron** re-prompts a persistent **session** that reconciles Plaid
> against a Google Sheet register through read-only **connectors**. The audit
> rules live as a **skill**, a **ledger** tracks what's already been flagged,
> and the agent only ever recommends — a person owns every cancellation.

## Make it do more

Once it works, paste any of these into the same session.

**Draft register rows**

```
For each shadow-IT charge, write the row you would add to the SaaS Subscriptions sheet in the report, so a person can paste it in.
```

**Monthly spend by team**

```
On the first Monday of each month, add the total recurring spend per owning team from the register to the report.
```

**Renewal reminders**

```
30 days before each annual renewal, post a reminder to #finance with the current price, the seat count and the owning team.
```

## Guardrails

Reconciling spend means reading two financial records, so the boundaries are
explicit:

- **Isolation.** Every run happens in its own isolated sandbox. The session is granted access only to Plaid and the register it's scoped to, and only the Slack digest
  is written back out.
- **Scoped secrets.** The Plaid access token and Sheets credentials are
  encrypted in the Secrets Manager and injected into the sandbox at runtime,
  scoped to the agents you grant them to.
- **Recommend-only.** The agent has no write access to Plaid, no billing API,
  and no way to cancel, downgrade, or pause a subscription. Its entire output
  is a Slack message with a suggested action; a human does anything that
  changes a subscription.
- **No repeat noise.** A ledger persists across weeks, so an unresolved
  finding gets one entry, not a new alert every run — only new or changed
  waste is reported as new.
- **Everything is code.** The agent's persona, its audit rules, and its
  per-system permissions are files in the repo, versioned and changed through
  a reviewed **change request** rather than a dashboard setting.

## The outcome

- **Every week:** Charges reconciled against the register on schedule
- **Recommend-only:** Nothing is ever cancelled or changed automatically
- **5 waste signals:** Duplicates, unused seats, price hikes, renewals, shadow IT

The comparison that used to require someone opening a spreadsheet next to a
bank statement now runs every week without anyone starting it, and only
reports what's new or unresolved. The agent reads the charges and the
register; the finance team decides which recommendation to act on.
