Reference Call

Nothing a vendor hands over contains a failure. The case studies in the deck are the accounts that went well, and the numbers come from the best month on record. That isn't dishonesty — it's what happens when you get to choose which examples to show. So buyers finish reading everything and still hold one question: does it really work that way.

That question never got answered in the meeting room. A team lead would ask whether companies actually running the product had run into trouble, and the rep would say that everyone is happy with it. A vendor vouching for its own product carries no weight at all, so the conversation ended with the buyer's confidence exactly where it started. This is where deals with every condition settled still sat unsigned for weeks.

A reference call changes who answers. The same question goes to someone at another company who pays for the product rather than to the person selling it. That person has nothing to gain from a sale, so they mention the awkward parts too — and the awkward parts are what create the trust. A buyer who hears "the first two months of moving our files over were rough" believes the "it works fine now" that follows.

There are six steps. The call itself takes 30 minutes; the preparation takes days.

1. Receive the request, or offer it. Prospects often ask first, but a seller will raise it when the terms are settled and the decision keeps sliding.
2. Pick a comparable customer. Size, industry and usage pattern all have to be close. Put a 200-person company on a call with a 300-person customer and it lands; put them on a call with a twenty-person team and it ends with "our situation is different."
3. Ask the existing customer. Explain the purpose, who they'll be speaking with, and how long it takes, then get their agreement.
4. Send material ahead of the call. A case study goes out first — a written account of one customer's before and after with numbers attached. Read in advance, it keeps the 30 minutes from being spent on basic explanation. A short list of likely questions helps too, sent to both sides so the customer isn't caught cold and the prospect arrives with something specific to ask.
5. Hold the call. The seller introduces the two sides and leaves. Stay on the line and the prospect won't ask the uncomfortable questions.
6. Close the loop. Ask the prospect what questions remain, and tell the customer who did the favor how it turned out.

The customer doing the favor is spending their own time, so asking the same account repeatedly wears out the relationship. Lining up three or four willing customers in advance and rotating through them is part of the preparation. Some companies formalize the arrangement by naming reference participation in the contract in exchange for a discount, which removes the awkwardness of asking but also narrows the pool to accounts that agreed before they had used the product.

One or two people join from each side. On the existing customer's side it should be the team lead who uses the product daily. An executive from that company can explain why they bought it but not what living with it is like. On the prospect's side, the champion joins — the person inside the account pushing for the product, who is collecting evidence to use internally. Sometimes the decision maker joins too, and the questions change weight when they do. Practitioners ask what's annoying to use; decision makers ask what support looked like after the contract was signed.

The timing is usually late. It belongs in the stretch after the product walkthrough and the pricing conversation, with only closing left. What remains at that point isn't a question about features but a question about spending two years with this company. It often overlaps with the weeks the customer's security reviewer spends on the security questionnaire.

Reaching for it too early costs something. Set up a reference call right after a first meeting and the prospect doesn't yet know what to ask, so the call drifts into a product introduction — and the same customer can't be asked again at the point where it would have mattered. A deal realistically gets one of these, which is a reason to save it.

"Everything checks out, but our VP is nervous. Could you set up a reference call with a customer our size?"

The internal case needs evidence that doesn't come from the vendor.

"Let's get agreement from three accounts for reference calls before we need them."

Manage a roster instead of scrambling for a volunteer under deadline.

"I'd suggest I drop off and the two of you talk for 30 minutes."

The seller's read is that an honest conversation requires their absence.

At Pagedeck this step shows up in nearly every contract with a company above 200 employees. When Sofia judges that the hesitation is about confidence rather than features, she offers a reference call. The roster is maintained separately: four customers across different sizes and industries have agreed in advance, and no account gets asked twice inside six months.

After the call, Sofia does not press the prospect on what was said. She asks only whether any questions are left. Trying to audit the conversation undoes the reason the call existed, and it also treats the customer who helped as a source to be checked on. What she does record is which customer was used and on what date, so the rotation stays visible to whoever handles the next deal.

The most common mistake is picturing a scripted call where a friendly customer says nice things. Since the seller picks the customer, the suspicion is understandable. But the choice ends at who joins — what they say is out of anyone's hands, and trying to control it backfires. If the rep stays on the line, or if the answers sound rehearsed, the prospect discounts the whole call as advertising and ends up trusting the vendor less than before it happened.

The other mistake is reading a scheduled reference call as a deal that's essentially won. It arrives late in the process, so it feels that way. But the call exists to verify something, which means it can resolve in the other direction. An existing customer who mentions slow support responses, or a feature their team gave up on, can cool a deal on the spot. Move the opportunity into the forecast as closed on the strength of a scheduled call and the number gets revised in the final week of the quarter.