Making the most of your Gong credits

Field Notes · Andrew O'Driscoll, Founder, RevOps Sherpas · Jul 2026 · 8 min read

A field note on reading your consumption and getting full value from Gong's AI.

In June 2026, Gong introduced credit-based pricing for its AI usage. It's a packaging change more than anything, and not a surprising one.

Gong has always run on AI. What's changed is how much of it is running. The AI work under the hood has gotten more capable and more intensive. Smart trackers doing AI work in the background, MCP letting Gong exchange information with other systems, and API calls now doing AI-intensive work alongside the data querying and ingestion they've always handled. That's more capability than many teams first put to use, and capability like that costs something to run. Metering it in credits is where the whole industry has landed. Consumption-based pricing for AI is the norm now, not a Gong quirk.

Gong signaled when smart trackers arrived that their usage would eventually be metered, so the model everyone's working through now is one they saw coming. Your AI usage draws on a defined allocation of credits, and going beyond that allocation is where a decision about buying more comes in.

The pool itself is straightforward. Every paid core seat adds 2,000 credits a year into one shared company balance that resets each contract year, and you can buy more on top that last through the term. Gong spends whatever expires soonest first, and notes the allocations can change as packaging evolves.

Your balance drawing down over the term, with each credit bucket and its expiry from the purchase-history report. The balance reconciles to those buckets within 1%.

What's new is the need to understand your own consumption. When AI work draws on a shared annual pool, "how much are we using, and where" becomes the thing that decides whether you get full value from Gong's more capable AI, or run short of it before the year is out.

Use the whole pool

Your credits are included with your subscription, so the simple goal is to actually use them. Not run dry months before your reset, and not coast to the end of the year with a big unused balance sitting there. Both are value you're entitled to and didn't get. What you want is to use the allocation deliberately across the year and land near the reset having gotten what it's there to give you.

And a high burn rate isn't automatically a problem. Heavy consumption points one of two ways. Either your trackers are defined too broadly and you're spending credits analyzing calls and emails they never needed to look at, which is waste you can tighten. Or you're running real value-add AI work that drives an outcome your business cares about, in which case running past your allocation is a decent proxy for exactly that: you're getting real value from Gong's AI, enough that you want more of it. Same number, very different stories, and the only way to tell them apart is to look. When it's the second one, buying more credits can be the right call.

Gong gives you the data

Here's the thing worth saying plainly: Gong already hands you everything you need. The usage, balance, and purchase-history exports carry the atomic-level detail, right down to the calls and emails behind your consumption. Nothing here is hidden.

Which means you have a choice. You can pull those CSVs, build the pivots, and assemble your own view every time you want to check where things stand. You can drop them into your AI tool of choice and have it work through the numbers. Or you can use a UI built to do exactly this, on the same data, in a few seconds.

But reporting is only half of it. The part that's harder to build yourself is the modeling: changing your assumptions and watching what happens next, a what-if you can actually steer rather than a static report. That's the thing a pivot table won't do for you, and it's coming up.

So we built a read, and a model

We put together a free, unofficial Gong Credits Analyzer. Same numbers anyone can pull from their own instance, read back in a way you can act on, and model against before you commit to anything.

It runs entirely on your own Gong exports, loaded into the tool. Nothing is pulled from anywhere else, and nothing you load leaves your browser.

The first thing it gives you is a straight answer to the question everyone's asking. At your current pace, do you run short before your reset, or land with credits to spare, and by how much. It puts a number and a date on it.

Model your usage up and the sample runs short, out November 4, about seven weeks before the reset.

Ease it back and the same data flips the other way, landing with credits to spare. Same question, both answers, depending on your pace.

Underneath that headline, it uses Gong's own published rates from their help center, applied to the calls and emails in your report. It ties back to the totals Gong already gives you, so these are Gong's numbers, broken down by where they came from.

Where every credit went, broken out by feature type and reconciled to the totals Gong already reports. In this sample, all of it traces to smart trackers.

Model it, don't guess. This is the part a report can't do for you. It keys off your smart trackers. Turn individual trackers on or off. Adjust the call and email volume a tracker is working against, since a tracker can be set to watch calls, emails, or both. Make a broad assumption across all your trackers at once, or tweak them one at a time. Every change updates your run-out date in real time, in credits rather than dollars, so the math stays about the thing you manage. You're not being told what to do. You're seeing what happens when you change your own inputs.

Model it tracker by tracker. Turn individual trackers on or off, adjust the call and email volume each one works against, and watch the runway update as you go.

The tool is honest about the projection. It's a flat-pace read off a snapshot of your usage, not a month-by-month forecast that pretends to know your seasonality, and it says so plainly.

One note on scope. Today the Analyzer models the credits your smart trackers consume. Support for seeing MCP and API consumption is on the way.

Why we bothered

This isn't a product. It's a thing we kept needing while helping teams sort out their credit picture, so we made it and put it out for anyone to use. Run it on your own numbers, see what surprises you, and make an actual decision about what to keep and what to lean into.

Run the free Gong Credits Analyzer at gongcredits.revopsherpas.com. It runs in your browser and nothing you load is uploaded.

Tags: Gong, Credits, AI, Smart Trackers, Pricing