Ask Where a Won Deal Goes Next

Most small businesses end up choosing between a CRM and a project management tool, or hacking one to do the other's job. At BookPrint we didn't have a good CRM, so we ran the work through Asana, and the customer lived in one place while the job lived in another. Every won quote had to be carried across by a person who remembered to do it. The software didn't know that the customer who signed on Tuesday and the job that started on Thursday were the same thing.

A lot of the companies on this page pay me if you buy from them, and they don't pay the same. One pays 80% of a customer's first year, one pays a $35 credit, and several pay nothing at all. You can't make incentives like that equal, so I've done three things instead. Every tool is placed by the job it does best. In every section I've joined, or applied to, the programmes of the tools that carry it, so no section is the one where I happen to get paid. And the last section of this page lists what each tool pays, including the ones I chose not to apply to and the ones that pay nothing. Read the recommendations with that table beside you.

Ask where a won deal goes next

The quickest way to judge a system for a small business is to win a deal in it and watch what happens. In a lot of tools the deal goes green and stops, a trophy in a pipeline report, and the actual work starts somewhere else. Where it goes next tells you what kind of business the tool was built for. I count six answers, and each one has good tools behind it.

It stays in the pipeline

Some businesses really are mostly selling. The work after the yes is small, or someone else does it, and what matters is how many conversations the team can keep moving. For them a sales-first CRM is the right tool, and the differences are in how the selling happens.

Pipedrive is the visual pipeline, the one most small sales teams picture when they hear the word CRM, and it sells a Projects add-on for the work afterwards at about US$8 a seat. Close is built for teams that pick up the phone, with calling, SMS and email inside the CRM. Freshsales starts cheap with phone and email built in, and comes from the same company as Freshdesk, if that is already your support desk. Nutshell is the simple all-rounder for small B2B teams, with human support included. OnePageCRM gives every contact exactly one next action, which suits an owner-operator who needs to be told what to do on Monday. Less Annoying CRM has one plan, one flat price per user and free human support, and it means the name.

Then there are the CRMs that live inside the inbox, for teams that will never open another app. Streak runs inside Gmail. folk CRM is a light relationship tool for people who work out of Gmail and LinkedIn. Copper is built into Google Workspace, and it is a poor fit for anyone on Outlook.

It becomes a project in the same tool

A won deal that becomes a project in the same tool is the BookPrint answer, and the one I find most interesting, because it removes the need for a second system without asking a small business to run an ERP.

In Capsule, closing a sale offers to create a linked Project straight away, and the sale details stay attached to it. Projects sit on kanban boards, so delivery has stages the same way sales does. On the Growth plan and above an automation does it without being asked: when an opportunity is marked won, create the project, attach a task list, assign people and send the customer a templated email. The emailing is the other thing Capsule does well, and it is classic CRM behaviour done properly. Templates live in the CRM, emails go out one to one through your own Gmail or Outlook, and they land on the contact's record, so the history of a relationship sits where the next person will look for it. In October 2026 it added Crunch invoicing, which sets up an invoice in one click when an opportunity is won, beside the Xero and QuickBooks integrations it already had.

It isn't the only one. Insightly sells CRM and project management as one product, and converts a won opportunity into a linked project. Copper does a similar handover inside Google Workspace. monday CRM sits on the same platform as monday's work boards, so a won deal can hand over to a delivery board, although the automation needs the Pro plan, the deal's fields don't copy into the new board, and there is a three-seat minimum.

It becomes an order, a job and an invoice

Further along is the mini-ERP: one system where the opportunity becomes a quote, the quote becomes an order, the order becomes a project with timesheets, and the timesheets become an invoice.

Odoo is the full version. A confirmed sales order can create the project automatically, time logged against it can be invoiced, and it all lives in one database. One app is free for unlimited users, which is a cheap way to start, but the breadth means real setup work. Odoo's own affiliate terms tell you how much: if one of its official implementation partners closes the sale, the referrer gets nothing, because the partners are expected to close most of them. Plan for a partner, or for a lot of evenings. Zoho One bundles more than forty apps, CRM, Books, Projects and Desk among them, on one licence at about US$90 per user per month on annual billing, and it is the cheapest way to own a whole suite, if you can live with the sprawl. Bitrix24 charges per organisation rather than per user, about US$49 a month for five people on annual billing, and throws in tasks, chat and an intranet, at the cost of an interface that takes a while to learn.

HubSpot is heading the same way from the other direction, by removing the walls between its own products and replacing them with a meter. That change gets its own section further down.

It becomes a job for the delivery team

Some businesses are mostly delivering. The sale is a handshake or a referral, and the hard part is running the work at a profit. For them the CRM is the small part, and the right tool starts from the job.

Productive is built around agency profitability: budgets, resourcing, time and invoicing per project, with a sales funnel that reviewers say doesn't act enough like a CRM. Teamwork.com does client work with time, budgets and billing built in. Scoro sells a quote-to-cash plan that joins estimates, delivery, time and invoicing, with a five-seat minimum. Projectworks, built in Wellington, does the same job for consultancies, from ten users up.

New Zealand has its own version of this story. Xero stopped selling WorkflowMax in 2024, and the brand was bought and relaunched by BlueRock, so WorkflowMax still exists, priced in New Zealand dollars and still linked to Xero, though its reviews on the Xero app store since the move have been poor. If you were a WorkflowMax shop, it didn't disappear, and the vendors in the paragraph before this one are where they say people went.

Then there are the tools most small businesses bend into shape, the way I bent Asana. Asana and ClickUp run tasks well and have no native time tracking or invoicing worth the name. Airtable and Notion will hold a CRM if you build one, and you will be the person maintaining it. For a solo or a very small studio there are all-in-ones that carry a client from proposal to contract to project to invoice: Bonsai, Moxie and Dubsado. HoneyBook does the same thing well, but is sold to businesses in the US and Canada, so for most readers here it isn't an option.

It goes to the customer's side of the glass

For some businesses the most important thing after the yes is what the customer can see. An accountant, a consultancy or an agency spends a surprising share of its week sending files, chasing approvals and answering "where are we up to?". A client portal turns that into a place the customer logs in to.

Assembly, which was called Copilot until September 2025, is a branded portal with a simple CRM, payments, tasks and contracts behind one login, so the customer sees one front door with your name on it. SuiteDash puts a portal, CRM, projects, invoicing and even courses into one flat price that doesn't grow per user, which is good value and a dense interface. Clinked is the secure option, white-label file sharing with a data room tier for firms that move sensitive documents. Moxo builds portals around step-by-step workflows for larger firms.

Trades are the same idea with a van. The customer gets a quote, a booking, a job card and an invoice, and the tool has to work on a phone in a driveway. ServiceM8 is the fast iPhone and iPad app, with a cut-down Android version for staff. Tradify is the simple New Zealand-built option for small trade businesses, from NZ$48 per user per month before GST. Fergus, also from New Zealand, goes deeper for plumbers and electricians, with supplier price books. Jobber has the most polished customer hub, where clients approve quotes and pay invoices, though New Zealand reviewers rate its Xero sync badly.

Between the sale and the work there is also the paperwork. Ignition, which started in New Zealand, turns a proposal into a signed engagement and then collects the payments automatically. PandaDoc does proposals, quotes and e-signatures that plug into most of the CRMs on this page.

It goes into a data model you design

The last answer is for businesses whose shape doesn't fit anyone's template. Attio is the data-model-first CRM: you define the objects and the relationships yourself, which is powerful for an unusual sales motion and a blank canvas for everyone else. Twenty is an open-source CRM you can host yourself, which is about as literal as owning the core gets.

The filing cabinet test

I have argued on this site that legacy CRMs are filing cabinets: software built to keep the current state of each record tidy, not to remember how it got there. The test is simple. Change a field, then ask the system what it said last month.

The answers vary more than I expected. HubSpot keeps a short memory: its export documentation says every property keeps its last 45 values on a contact and its last 20 on companies, deals and custom objects, with the source of each change, before the oldest drop off. Zoho's audit log covers thirty days and only admins can see it. Bitrix24 logs field changes. Capsule, as far as I can find, keeps none: its API has no history or audit endpoint, and its own data protection FAQ tells customers to run regular exports because backups can't protect against a user deleting or overwriting something. A self-hosted Twenty keeps whatever you decide to keep, because the database is yours.

For step one that is fine. The sequence I recommend starts with a simple system that writes down what the business actually decides, then improving it while you learn, then a data layer, then automation. Step one is supposed to be a filing cabinet. The mistake is buying a whole platform for step one and paying platform prices to learn what your business looks like.

It stops being fine the moment something other than a person starts writing to the cabinet. In October 2026 Capsule's ChatGPT connector gained the ability to log activities, set tasks and update custom fields on opportunities and projects, and almost every tool on this page is adding something similar. A confident wrong edit from a chatbot looks exactly like a right one, and in a system with no history there is nothing to roll back to. Whichever tool you choose, give an AI a read-only key until you have your own copy of the data.

Where the big platform is heading

On 1 October 2026 HubSpot changed how it sells itself, and it is the clearest view I have of where all-in-one software is going. The change so far applies to new customers in Europe, the Middle East and Africa, according to every partner write-up I could find; HubSpot's own pricing supplement names no region or date. Existing customers stay on what they have, and new customers in New Zealand, Australia and the US still buy hubs.

For those new customers the hubs are gone. There is no Marketing Hub, Sales Hub or Service Hub to buy separately. You choose one edition for the whole account, Starter, Professional or Enterprise, and each person gets a seat. HubSpot's supplement calls them Core Seats, from €60 a month on Professional, and Front Office Seats, from €120, which replace the old sales, service and revenue seats; some partners report the names changing at launch, so treat the labels as unsettled.

That fixes the thing that frustrated me most. At BookPrint I used the free CRM and paid for the marketing automation, and I once exported my own form data through Zapier to import it back into the same product, because the form responses sat on the wrong side of a paywall. Under the new model that wall doesn't exist. What replaces it is a meter. Each edition includes a monthly pool of credits, 10,000 on Professional, that resets every month and doesn't roll over, and the supplement prices the work in credits:

Work it through. Professional's 10,000 included credits cover about 660 marketing emails a month, before a single workflow runs. A monthly newsletter to 5,000 contacts is 75,000 credits, which at HubSpot's list price of about a cent a credit is roughly €650 a month on top of the seats. When the credits run out the features that use them pause until you buy more. Records themselves don't spend credits, but every workflow action on every stale contact does, so a messy database stops being untidy and becomes a line on the invoice.

I think the direction is right for the customer, and it is also exact proof of the pricing leverage I wrote about in own the core. A walled platform charged you for access to the next room. A metered platform charges you for every step you automate inside it, which means the more of step four you do on someone else's core, the more of your growth becomes their revenue. Most of the smaller tools on this page still charge per user or per organisation, so the contrast is real today, but none of them is immune to the same logic once AI agents start doing the work. HubSpot already prices its agents per result.

Own the history, rent the cabinet

The prediction I'd make from all this is not that CRMs die. The 2026 direction runs the other way: Salesforce opened its platform to outside AI agents, HubSpot sells itself as the context those agents work from, and the new AI-native CRMs are betting on capturing every call and email automatically, not on replacing the record. The screens get less important and the record gets more valuable, which is exactly why the vendors are pricing it.

So the useful question for a small business is not which tool is the data engine. None of them is, and almost none of them needs to be. Pick the tool that matches where your won deals go: the pipeline, a project, an order, a delivery team, a portal, a van, or a model of your own. Then decide who keeps the history. The cheapest owner of it is nearly always the business itself: a nightly export, or a small sync through the API that writes each day's state somewhere you control, so that history builds up in your hands instead of disappearing from theirs.

At BookPrint the gap between winning a job and doing it was a person with a good memory. Any of the tools in the middle of this page would have closed it. Buy a good filing cabinet for step one, then photocopy it every night, and the day a vendor's terms change, the copy is the thing that makes the cabinet replaceable.

What each one pays me

Status as of 9 October 2026. "Applying" means I've joined or am joining the programme. "Not applying" means a programme exists and I left it alone, because the tool only gets a line on this page. "None open" means there is nothing I can join, and the tool is here anyway.

The spread in that table is the reason it exists. ServiceM8's 80% of a first year and Dubsado's $35 credit are not the same incentive, and if I let the column on the right choose the tool, a trades business would get a recommendation shaped by my commission rather than its van. The test at the top of this page doesn't care what anyone pays.

Sources: vendor pages, documentation and affiliate terms for each tool, read 6 to 9 October 2026, including Capsule's blog ("Projects redesigned", 30 June 2022; "Capsule Uncapped", 6 October 2026), developer documentation and GDPR FAQ; HubSpot's knowledge base on property history (updated 8 September 2026), its Limited Release and Nordic pricing supplements and US pricing page; Odoo's affiliate terms; Assembly's rebrand announcement (30 September 2025); BlueRock's WorkflowMax relaunch; Zoho, Pipedrive and monday.com documentation. EMEA scope and the 1 October date for HubSpot's new model as reported by HubSpot partners including Hacking Demand and Pricing Innovation.