It is half past ten and you have not dialled anybody. You have been in a spreadsheet since nine, pasting company names, hunting for phone numbers, and checking whether you already called this one last month. The calling gets pushed to the afternoon, and then the afternoon goes too.
Every cold call tool sells you the dial. Almost none of them sell you the list. That part gets handed back to you as a data provider login, a filter panel, an export and an import. Klesos runs it instead. Here is exactly what happens when it does, including the parts it will not do for you.
What happens when you start a lead run
You describe who you sell to in one sentence. Plumbing firms in Manchester with their own vans. Dental practices in Leeds that do implants. Then you start a run.
The run does the sequence you would otherwise do by hand:
- check_circlePicks the data source for the market you described, rather than asking you to choose one.
- check_circleRemoves anyone already in your workspace or already in your connected CRM.
- check_circleScreens every lead against your hard requirements and drops the ones that fail.
- check_circleFills in the contact details that are missing.
- check_circleWrites a script for each lead that survives.
The run is queued and works through that in the background. It is not a live lookup, so the list does not land the second you press go. What you get while it works is an activity log that narrates each step in plain language, so you can see the decisions instead of watching a spinner and guessing.
That log matters more than it sounds. A list you did not build is a list you have no reason to trust. Reading the decisions is how you start trusting it, and how you find out your description was too broad when the wrong names start getting dropped.
Where the leads come from
You do not pick a provider and you do not need an account with one. There is no API key to paste. The run chooses where to look based on the market in your description, and that is the part that normally costs you a subscription and an afternoon of reading filter documentation.
You describe the buyer. It decides where to look.
Finding local businesses when the databases are thin
If you sell to plumbers, gyms or dental practices, you have already found out that the big contact databases are built for software buyers. They are thin on the shop down the road.
So for local work you type a trade and a city, and those businesses are sourced from Google Maps. You get the business phone number and the website.
Two honest limits. What comes back is the business, not a named person, so do not expect an owner's name and a direct dial. You will be asking for the decision maker by role on the call, the same way you would if you had found them yourself. And results of this kind are metered per business returned, so a big local run spends data credits.
What you get in exchange is a call list in a market the databases are weak on, ready to dial the same afternoon.
Using the spreadsheet you already have
You probably have a list already. A trade show export, something a mate sent you, a tab you started in March. It is sitting there not being called.
Drag the CSV in. The columns get mapped for you, so you are not matching header rows by hand. Duplicates get flagged, and so does anyone already in your CRM. Then every row is screened, enriched and graded, and the best fits sit at the top of the queue.
The part reps notice is that your own columns survive. The note that says "met at the trade show, asked about pricing in September" comes through onto the lead card as a call ready callout, so it is in front of you when the phone starts ringing instead of two tabs away.
One thing worth being straight about on cost. A phone number that is already in your file is not looked up again, so it does not spend a data credit. That is not the same as an import being free. Rows that are missing details still get enriched, and enrichment is metered like anything else.
Why the list shows you what it threw away
Half the names on a bought list were never going to buy, and normally you find that out on the call.
Every sourced lead is checked against your hard requirements before it reaches your queue, and the run keeps a visible list of what it dropped and why. You can read the calls you did not have to make.
That screen is also your steering wheel. If the dropped list is full of companies you would happily have called, your requirements are too tight. If the queue is full of companies you would not, they are too loose. You fix the description, not the list.
Let your old customers describe the next ones
You know your best three customers by name. You have probably never written down what they have in common, so every new list is a guess dressed up as a target market.
Upload a CSV of the customers you already have, keep every column, and run reverse analysis on it. The app reads your own list back to you as a description of the customer you should be looking for, in plain terms, out of your data rather than out of a template.
Then you put that description into a run and let it go and find more of them.
What happens to the data behind the list
Worth saying plainly, because list building means handing over data.
Klesos is built on infrastructure that is SOC 2 and GDPR reviewed. Knowledge you feed the sales brain is stored encrypted and screened for personal data and prompt injection before a model reads it, which we covered separately in the post on the PII firewall. One inference step in the coaching pipeline runs in the United States under the Data Privacy Framework, so we do not tell you the whole thing runs inside the EU.
And none of this puts a machine on the phone. The AI sources the list, screens it, enriches it and writes the script. Then it stops. You make the call and you say every word yourself.
Start with three calls instead of a list
You will know inside a few conversations whether a list built this way is any good. That is faster than reading about it.
Sign up and your first three calls are free. A call only spends one of the three if it actually connected and lasted at least ten seconds, so wrong numbers and dead lines cost you nothing. Your own local number is free too, in every market we cover.