Every outbound process starts with the same condition: you have to reach the person before anything else can happen. Reach is the first rung of the ladder we use to measure every process we run, and it behaves very differently depending on the kind of base you are calling.
Without reaching the customer, there is nothing else. – Dawid Wójcicki, Founder & CEO of Vee
That sounds obvious, and in a sense it is. What is less obvious is how often reach is treated as a technical detail of the dialler rather than as the first business result of the process. It is the rung on which a contact operation spends most of its effort, and the one where the type of base sets the realistic ceiling before a single call is made.
The ladder, in one paragraph#
We measure every process on four rungs. The order matters, because each rung is only reachable through the one below it.
- Reach. Did we get through to the person, within the agreed number of attempts and the agreed hours?
- Continuity. Did they stay on the line once they heard who was calling, rather than hanging up after the company name?
- Key information. Did the conversation produce what the process exists for: a promise to pay (PTP), a confirmed interest, a confirmed or cancelled visit, a completed survey?
- Business effect. Did it turn into the real result in the client's systems: a payment, a contract, an attendance? We influence this rung indirectly, and we say so.
A weak first rung caps all the others. A brilliant conversation design is worth nothing for the people it never reaches, and a high conversion rate on a small reached population can hide the fact that most of the assignment was never touched.
What we mean by reach#
Reach is measured on cases, not on attempts. A case is a record assigned to us for handling; it may take one attempt or many to resolve. The question is how many of the assigned cases ended with us actually getting through, not how many dials connected.
In regulated processes we go one step further and separate a connection from a right-party contact (RPC): reaching the verified person the case is about. In early collections, a call answered by a family member is connected but not reached. The AI closes it without disclosing anything about the debt and schedules another attempt. Counting that call as reach would flatter the first rung and confuse every rung above it.
Reach is also a business result in its own right, and its value depends on time. In early collections, every day without a verified contact lets a debtor slide further towards the next bucket, and for a bank the most painful part of delinquency is not the cost of the call but the delay that forces it to build reserves. In appointment confirmation, reaching the patient after the slot could have been refilled is worth very little. So we report reach together with when it happened, not only whether it happened.
Finally, reach is not the only route to an effect. In our experience, some customers pay simply because their bank called, even if they never picked up. That is why we also look for results among cases where nobody answered, rather than treating every unreached case as a loss.
Four bases, four kinds of reach#
Clients often ask what reach they should expect. The honest answer starts with another question: who are we calling? Across sales, cross-sell and collections, we see four types of base, and each has its own logic.
Cold: a fresh lead who has just asked#
Someone has just filled in a form on a landing page because they need a small loan quickly. They are probably looking for whoever will offer it first and actually put the money in their hand. Reach here is a race. The person is likely to answer, but only for a short while, and the first serious caller often wins. A fresh lead cools by the hour.
What decides reach in a cold base is speed and persistence. This is where a call centre struggles most: agents are busy with customers who are already interested, and the dialler stops when capacity runs out. We keep calling because capacity is infrastructure, not headcount. The voice matters as well, and in cold contact the choice of voice is a deliberate decision, something we learned in our very first outbound campaign.
Warmed: there is a reason to call#
A motor policy is about to expire. A customer has just insured a flat and might consider life or car cover. A contract is ending and a larger package might suit them better. These are cross-sell, upsell and renewal bases. The customer knows the brand and there is a natural reason for the call.
Reach in a warmed base depends on timing relative to the event. Call too early and the renewal is not yet on the customer's mind; call too late and the decision has already been made elsewhere. The brand name, which helps the customer recognise the call, also shapes the first seconds, so reach and continuity are closely linked here.
Known profile: customers with a history#
These are existing customers about whom the client knows a great deal: someone with a small loan for two years who pays more or less regularly, or a borrower who has just missed an instalment for the first time. Early collections lives largely in this world.
Reach in a known-profile base depends on using what is known. Which of the customer's numbers is actually answered? At what time of day? Did they ask us to call after work last month? Has a previous promise been broken? Each case may come back several times, so the history of earlier attempts is one of the most valuable inputs we have. The more we know at the start, the better the correlations and hypotheses we can build.
Worked-over: a base others have already been through#
Some bases consist of records that once applied for credit and have since been passed between several potential lenders before they reach our client. Reach here can look acceptable on paper, but the meaning of a reached, interested person changes. In such a base, the interested customer is more likely to be someone other lenders refused than someone they failed to reach. Miracles happen, but they are not a plan.
For a worked-over base, efficiency has to be measured differently, and expectations set accordingly, before the campaign starts rather than after the first report.
| Base | What drives reach | What to watch | Typical mistake |
|---|---|---|---|
| Cold | Speed, persistence and voice | Time from form to first verified contact | Calling too late |
| Warmed | Timing relative to renewal or purchase | Reach and continuity together | One schedule for every event |
| Known profile | Use of history: numbers, hours, earlier outcomes | Right-party contact, not pickups | Dialling the same unanswered number for months |
| Worked-over | Data quality and realistic limits | Quality of the interest, not its volume | Judging it by the yardstick of a fresh base |
Qualitative summary from Vee's operating experience across sales, cross-sell and early collections.
We cannot jump over a bad base#
We say this to clients in the first diagnostic meeting, and we mean it as a working principle rather than an excuse. No configuration turns a worked-over base into a fresh one. What we can do, if we know whom we are calling, is adapt the process: the intensity of attempts, the hours, the opening words, the voice and the definition of success. Where a base is of very low quality, that may also mean agreeing a different measure of success with the client before we start.
The same principle applies to comparisons. Reach figures quoted without the type of base behind them, by us or by anyone else, cannot be compared. A supplier with impressive reach on fresh leads may struggle on your known-profile collections base, and the reverse is equally possible.
What you control on the first rung#
Reach is not only the supplier's job. Several of its strongest levers sit with the client.
- Data quality. Current numbers, all of them, and a fast route for wrong-number findings back into your systems.
- Timing of assignment. A lead handed over the next morning, or a delinquent account assigned a week late, loses reach before the first dial.
- Context in the input file. First contact or repeat, broken promise, preferred language or hours: every field improves the schedule.
- A clear definition of reach. Agree whether you mean connections, right-party contacts or cases reached, and stick to it.
What to ask a supplier about reach#
- Do you measure reach on cases or on attempts, and do you separate right-party contact from any connection?
- How do you adapt the schedule to the type of base, and which base type is your quoted reach based on?
- What happens to a case after a wrong number, a family member answering or an unreachable number?
- Who enforces calling windows and attempt limits, and can they be overridden under volume pressure?
- How will you report reach alongside continuity, so that reach gained by over-calling is visible?
Reach is the least glamorous rung of the ladder and the one that decides how much the others are worth. It is where we start every diagnostic, because without it there is nothing else.
Follow The First Second
Occasional issues on the mechanics and economics of AI-powered customer contact. No noise, no vendor news.
Follow via RSS