Choosing a CRM for a SaaS company

Every marketing number you will ever be asked to defend in a board meeting comes out of the CRM. Pipeline sourced, win rate by channel, payback by segment, the lot. So the selection question is not which interface reps prefer. It’s whether the system can hold the shape of your business without three custom objects and a nightly script holding it together.

Most SaaS teams get this backwards. They run a demo bake off, the reps like the one with the nicer inbox, and eighteen months later nobody can answer what a self serve trial is worth because trials were never attached to accounts in the first place.

Why default CRM objects break on SaaS

Standard CRM design assumes a salesperson creates a record for a company, works it, and closes it once. SaaS violates that assumption in four specific ways, and each one causes a different reporting failure.

Self serve signups arrive with no company. Someone signs up with a Gmail address, gets to value, and then twelve colleagues appear over the next month. If those twelve are twelve unrelated contacts, your CRM thinks you have twelve leads instead of one account with thirteen users. Domain based account matching is the fix, plus a rule for what happens with free mail domains. Decide it before launch, because retrofitting is painful.

Product qualified leads need a score the CRM did not calculate. Usage lives in Amplitude, Mixpanel, PostHog or your warehouse, not in the CRM. The mistake is syncing everything. Push four fields: PQL score, the event that triggered it, the workspace id, and the timestamp. A rep needs to know why this account lit up and what to open with.

Expansion is not a new company. If the same customer buys 40 more seats, that is a second opportunity on the same account with a type of expansion, not a fresh account. Get this wrong and your net revenue retention calculation quietly includes new logos.

Multi workspace customers break hierarchy. One enterprise buyer might run six workspaces across three business units, each with its own admin and its own bill. You need a parent and child account relationship, and you need to decide which level ARR rolls up to. Notion, Figma and Slack all sell this way, and all three had to solve it internally before their revenue reporting made sense.

The four realistic choices by stage

You are not really choosing from thirty vendors. You are choosing from four archetypes, and the stage you are at narrows it to two. This sits inside the wider SaaS marketing stack decision, and the CRM is the one component you should expect to keep for five years.

Option	List seat cost per month	Admin burden	Best for
Attio or Pipedrive	$29 to $59	A few hours a month, no specialist	Pre Series A, under 10 sales seats, founder led selling
HubSpot (Sales + Marketing Pro)	$100 to $150 per seat, plus contact tiers	0.25 to 0.5 FTE ops	Series A through roughly $100M ARR, marketing and sales in one system
Salesforce Sales Cloud Enterprise	$165, plus CPQ and platform add ons	0.5 to 1 admin per 30 to 50 seats	Complex territories, partner channel, quoting, multi entity
Close or Copper	$49 to $139	Minimal	High velocity inside sales, Google Workspace shops

Attio deserves a specific note. It models flexible objects natively, which means workspaces, trials and expansion fit without custom object gymnastics, and teams migrating off spreadsheets tend to be live in a week. The tradeoff is a thinner ecosystem. If you need a niche integration, you may be writing it yourself.

HubSpot is the default recommendation for most B2B SaaS companies between Series A and Series C, for one reason: marketing contacts and sales records share an object graph, so lifecycle reporting doesn’t require a sync layer. The catch is contact tier pricing. Every self serve signup counts as a marketing contact, and a product with 400,000 free users will produce a bill that shocks the CFO. Mark non marketable contacts aggressively. Our HubSpot vs Salesforce for SaaS breakdown goes deeper on where each one stops fitting, and the HubSpot for SaaS companies guide covers the configuration specifics.

Salesforce is the right answer less often than its market share implies. Buy it when the process complexity is genuine: split credit across overlays, deal registration with resellers, revenue across multiple legal entities, a quoting process with approval matrices. Buy it for none of those and you have purchased a platform plus a permanent staffing commitment.

The fields marketing needs before anything else

Build these before you import records. Retrofitting fields onto 40,000 existing records is a data project, not a config change.

Self reported attribution is the one people skip and regret. Platform attribution will credit the last click, which for most SaaS companies is a branded search that tells you nothing. Asking the human directly returns a different and more useful answer, particularly for podcast, community and word of mouth, none of which show up in any analytics tool. Run both and report the gap honestly rather than pretending one of them is the truth.

Lifecycle stage definitions cause more inter team friction than any other field. Write down what makes a contact an MQL, in a sentence a new hire could apply, and put it in the sales and marketing SLA rather than in someone’s head. Then audit it quarterly, because it drifts.

What the SaaS CRM market looks like and why you should care

CRM is the largest single category in enterprise software, worth north of 90 billion dollars a year, and it keeps consolidating. Salesforce holds roughly a fifth of it, with Microsoft, Oracle, SAP, HubSpot, Adobe and Zoho splitting most of the rest and a long tail underneath.

That matters for vendor risk in a practical way. A tool in the long tail can get acquired and sunset, or get repriced when a private equity buyer arrives. That has happened repeatedly in adjacent marketing categories. If you are betting a five year data foundation on a Series A vendor, check their funding, their logo churn and whether the API has a documented deprecation policy. The wider SaaS market by vertical picture and the marketing SaaS vendor map are worth a scan before signing a three year term.

The other consolidation pattern to watch is bundling. Every major CRM now ships AI features as part of a higher tier rather than as a separate purchase. Price your renewal assuming the tier you are on today gets squeezed upward.

Migration: what it really takes

Migration cost is almost never the record import. Here’s what a 40 seat B2B SaaS company actually spent, in weeks of effort rather than dollars, on a HubSpot to Salesforce move.

When switching is genuinely worth it, and when it isn’t

Switch when the object model blocks reporting the business needs and the vendor has no path to it. That is the only clean trigger. Everything else is expensive theatre.

Reps disliking the interface is not a reason. Reps dislike every CRM, and adoption problems are almost always process problems wearing a software costume. If activity logging is poor, the fix is usually automatic capture from email and calendar plus a manager who inspects it, not a new platform.

A new VP of Sales who used Salesforce at their last company is also not a reason, though it is the most common one in practice. Make them write down the three reports they cannot get today. If ops can build those in the existing system in a fortnight, the answer is no.

There is a real counter case. If you sell through partners, quote complex configurations, or operate across multiple legal entities, staying on a simpler CRM has a compounding cost: spreadsheets appear beside the system, and within a year the spreadsheets are the truth. That is the point to move, and moving earlier is cheaper than moving later.

What to do this week

Open your CRM and answer three questions. Can you report on self serve signups that later became sales assisted deals? Does expansion revenue sit on the parent account? Is there one system everyone agrees holds ARR?

If all three are yes, you don’t have a CRM problem and you should spend the money on demand instead, whether that is content or paid. If one is no, fix the model in the system you have. If all three are no and the vendor has no route to yes, start the selection process properly, with the object diagram first and the demos last. Teams running this at scale should read the mid market SaaS marketing playbook alongside it, and if an external partner is doing the implementation, run them through a proper RFP and scorecard rather than a reference call.