Who Scratchpad is for
Scratchpad is for revenue teams that run on Salesforce and are tired of fighting it. If your reps avoid updating opportunities because the native UI is slow and clicky, and your managers spend Mondays chasing stale close dates and empty next-step fields, Scratchpad is built for exactly that pain. It sits on top of Salesforce as a fast workspace, so reps update deals in seconds and the data flows straight back into the CRM. RevOps leaders, frontline managers, and AEs who live in the pipeline get the most out of it.
It is a poor fit if you are not on Salesforce. Scratchpad is built around the Salesforce object model and inherits its permissions and rules, so there is no useful version of it for a HubSpot or Pipedrive shop. If you want CRM hygiene and forecasting on a different stack, look at a broader RevOps platform like Clari instead. Scratchpad earns its place only when Salesforce is already the system of record and the friction of using it is the problem.
A fast Salesforce workspace reps will actually open
The core idea is simple. Salesforce is slow to update, so reps put it off, and pipeline data rots. Scratchpad gives them a clean, spreadsheet-like surface plus Kanban boards where editing a field or moving a deal takes a second and syncs back instantly. Because the change lands in Salesforce through the API, managers see accurate data without nagging anyone. That speed is the whole reason reps adopt it, and adoption is where most CRM-hygiene projects die.
On top of the workspace, Scratchpad layers call recording and an AI notetaker. It records calls, transcribes them, and produces deal and account summaries with full history and context. Sellers capture notes against the right opportunity during or after a call instead of reconstructing them later. The hygiene monitor flags deals missing next steps, close dates, or amounts, which gives managers a running list of what needs attention before a forecast call rather than a quarter-end scramble.
Scratchpad Agents and AI field updates
The newer layer is agents. Scratchpad Agents listen to calls, read emails and notes, and then propose or directly apply structured updates to Salesforce fields: next steps, qualification criteria like MEDDIC fields, close dates, and amounts. Done well, this removes the most tedious part of CRM work, the manual data entry that reps hate and skip. AI credits meter this usage, and the higher plans pool credits across the team and lift the monthly allowance, so heavy AI use is a budgeting consideration rather than a flat feature toggle.
Where Scratchpad lands on AI-stack fit
Scratchpad scores 52 on Chief Revenue Buddy's AI-stack-fit scale, which puts it in the lower-middle of RevOps tools. The reason it lands there is the integration model. Scratchpad does not publish its own public API; it consumes the Salesforce API through OAuth and writes data back into Salesforce. So an external agent or your own code reaches Scratchpad data indirectly, by working against Salesforce itself rather than a dedicated Scratchpad endpoint. The upside is that everything ultimately lives in Salesforce, so your existing Salesforce automations and integrations keep working.
What holds the score back is the lack of an MCP server and the lack of a public Scratchpad API. An assistant like Claude or Codex cannot drive Scratchpad directly the way it can drive an MCP-native tool, and there is no documented Scratchpad developer API to script against either. For now, agent access runs through the underlying Salesforce layer, which is workable but a step behind tools that ship MCP or their own public API. Worth noting: Scratchpad's own in-product agents are genuinely capable at turning call and email context into clean field updates, so the AI value is real inside the product even where outside-in agent control is limited. It also connects to Slack, Gong, and Zapier or Make, and it is SOC 2 Type 2 compliant.
Pricing notes
Scratchpad runs a free plan plus paid tiers billed per user. The free plan covers individuals with a limited number of views, a monthly cap on call recording, basic AI field updates, and a small AI credit allowance. Paid plans unlock the full workspace, hygiene monitoring, advanced controls, and larger credit pools, with the team tier adding pooled credits and Gong support, and enterprise adding unlimited credits and white-glove agent setup. Annual billing carries the headline per-user rates and month-to-month costs more, roughly a 20 percent difference.
CRB has not hands-on tested Scratchpad, and the tier names and prices have shifted across sources, so do not lock in a number from this page. Verify the current per-user prices, credit allowances, and what each tier unlocks on the vendor's pricing page before you buy, especially if your team will lean on the AI agents, since credits are the variable that drives the real bill.
The verdict
Scratchpad is the fastest way to keep Salesforce clean without nagging reps, and the agent layer now does real work turning calls and emails into structured updates. If you are a Salesforce team and CRM hygiene is your bottleneck, it is an easy tool to justify. If you are not on Salesforce, it is not for you. And if you want forecasting and revenue intelligence across a wider motion, weigh a platform play instead.
Comparing options? Clari is the broader forecasting and revenue-platform pick, and People.ai leans into activity capture and pipeline data at enterprise scale. See the full best revops roundup for the head-to-head.
