One platform vs many tools: an honest comparison

Re-typing: the tax nobody budgets for

Re-typing is the most visible cost and the one most often shrugged off, because each instance is small. A name and number copied from a chat into a sheet. A quotation's figures copied into an invoice. An appointment copied from a message into a diary. Thirty seconds each. But it happens dozens of times a day, by the people whose time is most expensive, and every copy is a chance to get it wrong.

The errors are the real tax. A number typed with two digits swapped means a reminder that never arrives. An amount copied wrongly means an invoice that has to be reissued and a customer who now doubts every invoice. A customer entered twice under slightly different names means two records, two histories and a conversation that starts from the wrong one.

The test is simple: pick one customer and count how many places their name is typed from first enquiry to final payment. If the answer is more than one, every one of the others is a place where the record can drift from the truth.

• Count the places one customer's details are typed across the whole journey.
• List the errors from the last month that came from a copied number or amount.
• Ask who does the copying and what else they could do with the time.

Lost context: the conversation that lives on someone's phone

Harder to see than re-typing, and more expensive, is lost context. A customer's history is spread across the owner's WhatsApp, a salesperson's call log, the accountant's invoice file and a staff chat group. Each person holds a piece. Nobody holds the whole.

The cost shows up at the worst moments. A customer calls with a complaint and the person who answers has never heard of them. A salesperson leaves and takes the only record of forty conversations with them. The owner wants to know why a customer stopped buying and the answer is in a chat thread that was deleted to free up space.

Context is also what makes follow-up possible. A reminder that references the last conversation is read; one that does not is ignored. A renewal offer that knows what the customer bought last time is taken seriously. Without the history in one place, every message is a cold message, however warm the relationship actually is.

• Where does a customer's full history live today, and who can open it?
• What happens to that history when a staff member leaves?
• Can the person who answers the phone see the last conversation before they speak?

What one login actually changes

One login is not about convenience, though it is convenient. It is about there being one record per customer that everything attaches to. The enquiry, the messages, the quotation, the invoice, the payment, the support ticket and the meeting all land on the same timeline, because they were created from it rather than copied into it.

That changes a few things at once. Anyone on the team can answer where are we with this customer by opening one screen. Follow-ups can reference what actually happened. Reports are built from the real records rather than from a sheet someone updates on Fridays. And the owner's morning starts with a list of what needs attention today, drawn from every part of the business, rather than a tour of five apps.

It also changes what staff do. The time that went into copying and reconciling goes into customers. The knowledge that lived in one person's phone lives in the business. That is the point, and it is why the switch is worth the effort of making it.

• One record per customer, with everything created from it.
• One place to answer where are we with this customer.
• Reports from the records, not from a sheet updated by hand.
• A morning list of what needs attention across the whole business.

How to evaluate a platform: a checklist

Evaluating a platform is mostly a matter of asking the right questions in the right order, and refusing to be impressed by a long feature list. The questions below are the ones that matter for a small business, and the honest answers are usually easy to find in a trial.

Start with your own numbers. Put ten real customers in, with their real history, and run a real week: an enquiry, a message, a quotation, an invoice, a payment, a support question. If it takes more than a day to get to that point, the platform is harder than it looks. If the people who will use it every day cannot do it on a phone, it will not get used.

Then ask the awkward questions. What happens to your data if you leave? Is there a way to export it? Which features are included in the plan you can afford, and which are on a plan you cannot? Is the pricing on a public page with every plan shown, or does it need a call? Who do you talk to when something breaks, and is that a person? A platform that answers these plainly is one you can rely on.

• Does it run on a phone, for the staff who will actually use it?
• Can you import your existing customer list without re-typing it?
• Does every part of the business attach to the same customer record?
• Is the pricing public, with every plan and what each includes shown?
• Can you export your data, and in what form?
• Is support a person, and how do you reach them?
• Does any AI act only with a person's approval, and can you see what it did?

Moving from Excel and Google Sheets

Most small businesses are moving from a spreadsheet, and the spreadsheet is not the enemy. It holds the only customer list the business has, and the move should respect that. The goal is to get that list into the platform cleanly, once, and then stop updating the sheet.

Before importing, spend an hour on the sheet. Put one customer per row. Make sure there is one column for the name, one for the phone number, one for the email, and one for anything else you want to keep, such as the city, the source of the enquiry or the last purchase. Remove duplicates, or at least mark them. Fix phone numbers so they are all in the same format. The hour spent here saves days of untangling later.

Import customers first, before anything else. Once the customers are in and correct, the rest follows naturally: the next enquiry is logged against a customer, the next invoice is raised against one, the next message goes out from one. Keep the old sheet, read-only, for a month, so there is somewhere to check if something looks wrong. Then archive it. A business that keeps updating both the sheet and the platform has made the problem worse, not better.

• One customer per row; name, phone, email and a few useful columns.
• Duplicates removed or marked; phone numbers in one format.
• Customers first, then let enquiries, invoices and messages attach to them.
• Old sheet read-only for a month, then archived. Never updated in parallel.

The approval model for AI

AI in a business tool is useful for exactly one reason: it can draft the reply, the follow-up and the summary that a person did not have time to write. It is dangerous for exactly one reason: it can send them. The difference between those two is the approval model, and it should be the first thing you ask about.

The model that works for a small business is simple. The AI drafts; a person approves; the message goes out with the person's name on it. Everything the AI proposed, and whether it was approved, rejected or edited, is visible afterwards. The owner can see at any time what the AI did last week, and nothing reaches a customer that nobody looked at.

The same principle applies to calls. An AI that answers an incoming call when the team is busy, takes the caller's details and the reason for the call, and records it for a person to act on, is doing useful work in a way the caller understands. An AI that places calls to people who did not ask for them is a different thing, and this is not what Selfeey's AI does. The line between answering and initiating is the line to hold.

• The AI drafts; a person approves; the person's name goes on it.
• Everything proposed and every decision is visible afterwards.
• The AI answers incoming calls and records them; it does not place calls.
• You choose how much goes through approval, and you can see the activity at any time.

How Selfeey does this

Selfeey is built around the one-record idea described above. The Customer 360 timeline is the record: every message, quotation, invoice, payment, ticket and meeting for a customer sits on it, because each was created from it. Contacts and companies and the Customer list with lifetime value are the ways into that record, and Import from Excel and Google Sheets is how an existing customer list comes in without re-typing.

The parts of the business attach to it. Leads with score and source and the Leads inbox hold enquiries; the Sales pipeline, Quotations and Your own catalogue turn them into deals; Invoices with payment links and Payments, reminders and ageing handle the money; Tickets with response times handle support; Video meetings with a guest link handle the calls that need a screen. WhatsApp messages and approved templates keeps every customer conversation on the record rather than on a personal phone, and Message templates and content library keeps the team's wording in one place. Employee records, Staff attendance and Payroll sit in the same login, so the business is in one place rather than the customer side of it.

The owner's day starts with the Today overview and My tasks, which gather what needs attention from across the business. The Business overview report, Sales analytics and the Reports library with export answer the bigger questions from the real records, and the Reports library's export is how your data leaves when you want it to.

The approval model is a product feature, on the PRO Ultimate plan. AI employees draft replies and follow-ups; what they propose waits in the Approvals queue until a person approves it. Automations and Ready automation templates run the sequences you set up, and Activity and results shows what ran and with what outcome. AI answers your calls when nobody can pick up, and Call history keeps each call on the record; the AI does not place calls. Which features sit on which plan, and what each plan costs, is on the pricing page at /pricing, with every plan shown.

When separate tools are the right answer

An honest comparison has to say this: sometimes separate tools are right. A business whose whole value is one specialist function, such as a design studio living in its design software or a trading firm living in its accounting package, should keep the specialist tool and connect the customer side to it, not replace it.

The question is not whether a platform does everything. It is whether the parts of your business that touch the customer are in one place. If your accounting is elsewhere but your enquiries, conversations, quotations, invoices and follow-ups are on one record, you have most of the benefit. If your accounting is in one place and your customers are in five, you have none of it.

Start with the customer record. Get the list in, get the conversations onto it, get the next invoice raised from it. Judge the result on your own numbers after a few weeks, not on a feature list. That is the comparison that matters, and it is the one a trial lets you make.

Questions