Pipedrive for Agencies: The Third Plan or Nothing

An agency's CRM problem is not selling. It is that the same system has to hold new business and client delivery at once, and most CRMs are built for one of those.

Pipedrive sells a delivery add-on alongside its pipeline, which is why it keeps appearing on agency shortlists. Whether that works for you comes down to one number: how many client separations you need, and on which plan they exist.

OUR TAKE: Agencies belong on the third plan, not the second, and the reason has nothing to do with features in the usual sense. Teams, visibility groups and permission sets do not exist below it, and an agency without client separation is an incident waiting to happen. At that plan the delivery and document add-ons come included, which is most of the value.

## The client separation question

This is the item to settle first, because it decides your plan before anything else does.

Teams, visibility groups and permission sets all begin at the third plan, capped at fifteen of each, rising to twenty-five on the top plan^[1]. On the two cheaper plans they do not exist.

For an agency, that is not a nice-to-have. A freelancer who works on two accounts should not see the pipeline for a third. A client who asks for a user login should not get a view of your other clients' deals. There is no configuration below the third plan that produces that, and no amount of discipline substitutes for it.

Fifteen groups is enough for most agencies and tight for a large one. If you pass it, the top-up is bundled rather than sold separately: teams, visibility groups and permission sets go together at $100 a month for ten more of each^[2].

## New business and delivery in the same system

Projects, the delivery add-on, is charged on the two cheaper plans at $20 a seat per month and included from the third^[2][1].

That is the mechanic that makes the third plan the agency plan. A five-seat agency on the second plan paying for Projects and Smart Docs costs $273 a month. The third plan, including both, costs $295. Twenty-two dollars apart, and the third plan also triples the custom fields and the automations and adds the client separation you needed anyway.

Smart Docs matters here more than in most verticals, because agencies send proposals constantly and sign them. It is $39 a month for up to thirty documents or $59 for up to a hundred on the cheaper plans, and included from the third^[1]. Thirty documents is one busy month for a small agency.

A caution on scope. Projects inside a CRM is a deal-adjacent delivery tracker rather than a production system. It will hold the stages of a client engagement. It will not run a studio's capacity planning or time tracking, and agencies that expect it to are the ones who end up with two systems anyway.

Custom fields, which agencies exhaust faster than most

Every client relationship carries metadata that a product company does not have: retainer terms, billing contact, renewal date, scope boundaries, the account lead, the platform access you hold.

Custom fields cap at 30 on the entry plan, 100 on the second, 300 on the third and 500 on the top, counted per company rather than per pipeline^[1].

Thirty is not enough for an agency. A hundred is workable. Three hundred is comfortable, and it is another reason the third plan is the honest starting point here rather than an upsell.

Formula fields, for calculated values such as effective hourly rate or margin per retainer, start at the third plan, cap at ten, and remain at ten on the top plan^[1]. Plan to calculate agency economics outside the CRM.

## The seat question, with contractors

Agencies run on a core team plus freelancers, and this is where the billing model costs money.

Seats bill whether or not anyone occupies them, and releasing a seat takes effect at the start of the next billing cycle rather than immediately^[3]. Every seat sits on the same plan, so a contractor who needs access for six weeks costs a full plan seat for at least a billing period.

Two consequences. First, do not buy seats for freelancers speculatively. Second, think hard before taking annual billing: on the third plan the annual discount is 25.3 per cent and on the top plan 20.2 per cent, not the 42 per cent the page advertises, which is computed from the cheapest plan alone^[1].

An agency whose team size moves with the client roster should price monthly and treat the discount as the cost of flexibility rather than the other way round.

What agencies should leave switched off

Web Visitors, unless new business is your constraint rather than delivery. It identifies organizations visiting your site and bills by how many: $49 a month up to 200, $99 up to 500, $299 up to 2,000^[4]. For an agency with an inbound motion it can pay for itself. For one that grows by referral it is a line item with no return.

Campaigns, unless you are running your own marketing list rather than your clients'. It bills by contact count from $0.016 down to $0.003 per contact per month^[1]. Running client campaigns through your own CRM mixes their data into your contact count and your bill, which is both expensive and a governance problem.

Keep client marketing in client systems. Use this for your own pipeline.

## Retainers, which are not deals

An agency's revenue is mostly recurring, and a pipeline is built around things that close once.

The product has subscription and forecast reports from the second plan^[1], which is the mechanism for this and is worth configuring properly rather than modelling retainers as repeated deals.

Two practical points. A renewal is a deal with a known date and a high win rate, so it belongs in a separate pipeline with its own stages rather than mixed into new business, where it will flatter your conversion rate beyond recognition. And retainer value changes mid-term more often than new business value does, which is exactly the case custom fields handle and formula fields would have handled better, had there been more than ten of them.

Agencies that get this right report new business and renewals separately and never average them. Agencies that get it wrong have a forecast nobody believes by the second quarter.

The contractor question, answered properly

The most common agency question about any per-seat CRM is what to do about freelancers, and here the answer has three parts.

Short engagements under a billing cycle: do not give them a seat. Share what they need outside the CRM. A seat released mid-cycle still bills to the end of it^[3].

Recurring contractors who work most months: give them a seat and keep it. Churning seats on and off costs the same as keeping them and adds administration.

Clients: see the section above. A client login is a full seat at your plan price.

None of this is a criticism of the product. It is how seat billing works everywhere in this category. It is just more expensive for agencies than for companies with stable headcount, and the quote should reflect your real pattern rather than your core team on a good month.

## New business against delivery, in one pipeline or two

Two. Always two.

The temptation is one pipeline with delivery stages tacked onto the end of the sales stages, because it looks tidy. It destroys every conversion metric you have, because a signed client sitting in "onboarding" is counted as an open opportunity.

Run new business as a pipeline and delivery through Projects, which is the add-on built for it and is included from the third plan^[2].

That separation also makes the visibility question tractable. Delivery is where client separation matters most, because that is where the client-specific detail lives.

Where it is strong for this vertical

Adoption, which is an agency problem specifically. Account managers are not sales people and will not learn a complex CRM. In the hundred reviews we read, setup averages 4.3 stars, the highest in the corpus, and complexity is raised in only four per cent of English reviews.

Support, the largest theme at 33 per cent of reviews, averaging 3.8 stars. Agencies rarely have anyone whose job this is.

And the API, which is included rather than surcharged: 150,000 tokens a day per seat on the third plan, capped at 100 million per company^[1]. Agencies that build client reporting dashboards use more API than they expect, and here that capacity grows with the team.

A worked setup for a six-person agency

Six seats. Four account people, two on new business. Projects and Smart Docs both in use.

	Second plan	Third plan
6 seats, annual	$234/mo	$354/mo
Projects	$120/mo	included
Smart Docs	$39/mo	included
Client separation	none	15 groups
Custom fields	100	300
Total	$393/mo	$354/mo

The third plan is cheaper. Not by a little.

Projects is charged per seat on the cheaper plans, which is the detail that flips this. Six seats at $20 is $120 a month for a feature the next plan up includes^[1][2].

An agency quoted on the second plan is being quoted wrong.

Running client logins

Some agencies want clients to see their own pipeline. It is a good retention mechanic and a bad idea without separation.

Every seat on the account sits on the same plan^[2]. A client login is a full seat at your plan price, not a cheap viewer, and it counts against your billed seats like any other.

At $59 a seat on annual billing, five client logins is $3,540 a year. Price that against what the transparency is worth to you, because it is not a rounding error.

And check the visibility groups first. Fifteen on the third plan covers a client roster of about that size once you account for your own internal groups.

Reporting across clients

Reports are capped per seat: fifty on the second plan, 250 on the third^[1].

An agency reporting per client passes fifty faster than it expects. Ten clients with five standard reports each is the whole allowance, before anything you build for yourself.

That is one more reason the third plan is the honest starting point here. Not a feature argument. An arithmetic one.

## The quote to ask for

Third plan, seats for the core team only, annual or monthly depending on how stable that team is.

Then four questions in writing: whether there is a seat minimum, since none is published anywhere we could read^[2]; what the meeting intelligence feature costs, since the comparison table shows it included on all plans while the page's own data prices it at $25 a seat per month^[1]; what happens to a seat released mid-term; and exactly how cancellation works.

That last one carries the most weight. Across the hundred reviews we read, every single one mentioning contracts or cancellation was a one-star review, with no exceptions. For a business whose own client contracts are the thing it argues about most, that is a conversation worth having before signing rather than after.