The average marketing team pays for more tools than it uses. Overlapping features, forgotten subscriptions, a free trial that quietly became a paid plan two years ago. Nobody decided the stack should be this big. It just grew, because nobody is responsible for the whole thing.
A stack audit cuts the cost and, more importantly, the chaos. [Gartner has consistently found that marketers use only about a third of their martech stack's capabilities](https://www.gartner.com/en/newsroom/press-releases/2023-11-13-gartner-survey-finds-marketers-utilize-just-33-percent-of-martech-stack-capabilities). You are almost certainly paying for capability you do not touch. The audit finds it.
Why stacks bloat
Tools accumulate the way junk accumulates in a garage. Each individual addition made sense at the time. A project needed a one-off tool. A new hire brought their favorite app. Someone signed up for a trial to test something and forgot to cancel.
The problem is that nobody owns the stack as a whole. Individual tools have champions; the total collection has no owner. So it grows in one direction only, because adding a tool is a small decision anyone can make and removing one is a decision nobody is responsible for making. Bloat is the default state of a martech stack left unmanaged.
Run the audit: list everything
Start by listing every tool you pay for, with four columns: cost, owner, what it does, and who actually uses it. Pull this from your billing, not from memory, because the tools you forgot about are exactly the ones the audit is supposed to catch.
The list itself is often the revelation. Teams routinely find subscriptions they did not know were still active, tools two people are paying for separately, and platforms whose annual cost nobody had added up. Seeing it all in one place is half the value.
Look for the four patterns
Once you have the list, look for four things.
Tools nobody uses. The forgotten subscriptions. These are the easy cuts.
Tools that overlap with each other. Two tools doing the same job because different people picked different favorites.
Tools that overlap with features you already own. This is the big one. HubSpot, for instance, includes features that SMBs often pay for separately elsewhere: email, landing pages, forms, basic reporting. If you own the platform, paying for standalone tools that duplicate it is pure waste.
Tools whose cost is out of line with their value. The expensive tool used for one small thing. Maybe it stays, maybe a cheaper option replaces it, but it gets examined.
Decide what to cut
For each tool, three questions. Does it have an owner? Does it produce a measurable result? Would anyone notice if it disappeared?
A tool with no owner, no measurable result, and that nobody would miss is a clear cut. A tool that overlaps with something you already own is a consolidation candidate. Be honest about actual usage, not about what you intended to use the tool for when you bought it. The intention does not show up on the invoice.
Check dependencies before you cancel
Here is the discipline that keeps the audit from breaking things. Before you cancel a tool, check what depends on it. Integrations that route data through it. Automations that trigger from it. Data that lives only there and would be lost.
Export anything you need to keep. Map the dependencies before pulling the plug. The audit is about removing waste, not about breaking a working system because you cut a tool that quietly did something important. The diligence here matters as much as the decision to cut.
Should you consolidate?
The biggest version of pruning is consolidation: collapsing several tools into one platform. For most SMBs this wins. Fewer tools means fewer integration headaches, fewer logins, lower per-seat costs, and data in one place instead of scattered.
The trade-off is vendor dependence and sometimes accepting a good-enough feature instead of a best-in-class standalone. Weigh it per case, but for most small teams the simplicity and cost savings of consolidation beat the marginal feature loss. The HubSpot versus Salesforce comparison covers the core platform decision that usually anchors a consolidation.
How to run a stack audit in a morning: the three-tab spreadsheet method
The reason most teams never do a stack audit is that "reviewing every tool" sounds like a week-long project. It is not. A morning is enough if you follow a simple structure.
Open a spreadsheet with three tabs. Tab one is the inventory. Pull every subscription from your company credit card statement and any billing inboxes you have access to. List the tool name, the monthly cost (convert annual plans to a monthly number), who is the named owner, and how many people actually use it. Do not rely on memory. Pull from the billing. You will find tools you forgot were still running.
Tab two is the decision. For each tool in the inventory, assign one of four labels: Keep, Cut, Consolidate, or Test. Keep means it is actively used and there is no cheaper replacement. Cut means nobody uses it or it duplicates something else. Consolidate means the function is covered by a platform you already own — if you have HubSpot, standalone email tools often fall here. Test means you are not sure and you need 30 days of data before deciding.
Tab three is the action list. For every Cut tool, the action is cancel at next renewal or cancel now if the cost justifies the admin. For every Consolidate tool, the action is migrate the function to the existing platform and then cancel. For every Test tool, set a calendar reminder 30 days out to review usage data before you decide.
With this structure, the audit itself takes two to three hours. The decisions are clear. The action list follows directly. Most SMBs that have never done this find at least two or three tools in the Cut category on the first pass, which often adds up to several hundred dollars a month in savings.
What to do with tools you are not sure about: the 30-day test protocol
The hardest category in any audit is the tool nobody is sure about. Someone thinks it is still being used. Nobody knows what it does. The original person who signed up for it left the company months ago.
Do not cut these blindly and do not keep them by default. Put them in a 30-day test protocol instead. Here is how it works.
First, restrict access to everyone except the person running the audit. Wait two weeks. If nobody notices the restriction and nobody asks why they cannot get in, that is a clear signal nobody is actively using it. If someone reaches out asking why access is blocked, you have found the actual user and you can make a real decision with real information.
Second, check the tool's own activity log or last-login data if it provides one. Most SaaS platforms show the last login date per user. A tool where nobody has logged in for 90 days is effectively unused, regardless of whether the subscription is still active.
Third, check integrations explicitly. A tool can appear unused at the front end while still passing data to something important in the background. In HubSpot, Zapier, or Make (formerly Integromat), search active workflows for the tool name. If it appears in a live automation, map where that data goes before you touch anything. Cancelling a tool that feeds a live workflow without migrating the data first breaks something and creates cleanup work that costs more than the subscription.
After 30 days, you will have enough information to make a real decision: keep it, cut it, or migrate the one function it is quietly performing to a tool you already pay for. The protocol turns an uncertain maybe into a confident yes or no without the risk of breaking something you did not know existed.
What this means for you
Your martech stack is bigger than it needs to be, because nobody decided it should be this size. It accumulated. An audit reverses that.
List everything from the billing, not from memory. Find the unused, the overlapping, and the duplicated. Cut what has no owner and no result. Check dependencies before you cancel. The money saved is the headline. The simpler, clearer stack is the benefit that lasts.