Notes for owners · CRM, automation and RevOps
How to pick a CRM once your sales process is written down
Write your stages on one page, then test each CRM against them with a one-page checklist and a trial using your own deals. Process first, tool second.
The GullySales team · Updated 17 Oct 2026 · 5 min read
On this page
A software salesperson has shown you a demo with twelve colourful stages, and your own team uses five words on a whiteboard. Pick a CRM only after you can list your sales stages on one page, because the stages are what you are testing the tool against. Write the stages, mark which fields each needs, then run your own live deals through two or three candidates and see which one your team can update without a manual.
What must the one page contain?
Before you open any CRM website, put these on a single sheet.
- Your stages, in order. Five to eight. Use your own words: "site visit done", not "qualified opportunity".
- The rule for entering and leaving each stage. What must be true for a deal to move on.
- What you record at each stage. Quantity, site address, who decides, quotation number, the date of the next call.
- Who touches the deal. Owner, salesperson, the person who sends quotations, accounts.
- What you want out. Three questions you cannot answer today, such as "which enquiries have not had a follow-up in five days?"
If you have not written this page, our post on documenting your sales process shows how in an afternoon.
How do you compare tools against the page?
Make a table with your stages and needs down the side and the candidate tools across the top. Score each cell by trying it, not by reading the feature list.
| Test from your page | Tool A | Tool B | Tool C |
|---|---|---|---|
| Can I rename the stages to my words? | No | Yes | Yes |
| Can I add the fields from step 3 without paying for a higher plan? | No | Yes | Yes |
| Can a salesperson update a deal from the phone in under a minute? | No | Yes | Yes |
| Does the reminder for the next call reach the right person? | Not tested | No | Yes |
| Can I get the three questions from step 5 without exporting to a sheet? | Not tested | Yes | Yes |
| Does it work in the languages my team writes notes in? | Not tested | Yes | No |
| Can I get all my data out if I leave? | Not tested | Yes | Yes |
The table above is filled in with the invented results of the Kottayam firm described below. Tick, cross or a note. Anything that fails two rows is out. Do not count features you will not use; the sheet decides, and it only has your needs on it. For a longer list of what to weigh in the market, see how to choose the right CRM for a small business. Do not rely on the price on the first page: ask what the plan you need costs for your number of users, in writing.
How do you run a trial that means something?
Take ten real deals, a mix of won, lost and live, and enter them in each tool yourself. Then ask the salesperson who dislikes software most to update three of them on their phone while you watch. Count the taps and listen for sighs.
Two things to look for. First, whether a deal in the middle of your process fits the stages as you have named them, or whether you must bend them to fit. Bending is often the early sign of a tool that gets abandoned within months. Second, whether you can see at a glance which deals are stuck. That view is what you are buying.
Run the trial for a week of real use, not a day of poking. The point of a longer trial is that real salespeople forget things, and a CRM has to survive that.
For example: a lift-installation firm in Kottayam
Take an invented firm in Kottayam that installs and services passenger lifts in apartment blocks and small commercial buildings. Its process has seven stages, including two that a standard CRM did not have: "site survey done" and "committee approval". The owner had nearly bought the first CRM a friend recommended.
The owner tried the page against that tool. Two stages could not be renamed without a higher plan, and no field existed for the approval date. The salesperson tested it on a phone and gave up at the third screen. A second tool allowed renaming and free fields, but its reminders went to the account owner and never reached the salesperson. The third passed six of the seven rows. Notes in Malayalam displayed badly, so the team agreed to write notes in English for now and to test again after three months. That was a cost, but a known one, found by testing before paying.
What if no tool passes?
Then simplify the process before blaming the tools. A page with twelve stages and twenty fields will fail every CRM for a small team. Remove stages that are only ever a few hours long. Remove fields nobody reads. It is also a sign that a shared spreadsheet is the right tool for now: columns for your stages, a date for the next call, and a colour for deals gone quiet. Our post on naming stages helps trim the list.
What if my team lives on WhatsApp?
Some small teams do, and a CRM that ignores WhatsApp may be abandoned. Ask each candidate how a chat becomes a deal record: a copied note, a forwarded message, or an integration. Test it with one real thread. Do not assume any tool reads your customers' messages, and check what your own adviser says about storing customer chats outside your phone.
When is a CRM the wrong purchase?
If nobody owns the data, if your process changes every month, or if the owner will not look at the dashboard, a CRM adds a cost and a chore. Buy it when the same problem has appeared three times: a lost follow-up, a duplicate quotation, a salesperson who left and took the history. Ask about data protection too; where the tool stores personal data is a question for your own adviser.
Taking the page to your team
Take the one page and the filled table to your team and ask who will use the tool every day. Their answer matters more than the price. The stages themselves are described in the sales process hub. If you want a second opinion on the choice, see CRM consulting and selection, or tell us what you need.
