Documents tell you what happened. The seller tells you why it happened. A structured 60-minute call sits between a good acquisition and an expensive mistake — and most buyers waste it on small talk.
Deal Alert AI is reader-supported. We earn commissions from affiliate links at no cost to you.
This post is based on a video from our Deal Alert AI YouTube channel. Watch the original or read the full breakdown below.
By Sophal Lanh, Founder of Deal Alert AI
I have watched buyers spend eleven hours reconciling Stripe exports against a profit and loss statement, then hop on a 25-minute call with the seller and spend the first ten minutes talking about the weather in whatever city the seller happens to live in. That is backwards. The spreadsheet work is important, but it is bounded — you either reconcile or you do not. The call is unbounded. It is the only part of due diligence where you can ask a question nobody anticipated and get an answer nobody prepared.
The seller interview call is the most underutilized tool in online business due diligence. Financials tell you what happened. The seller tells you why it happened, what almost happened, and what is about to happen that has not shown up in the numbers yet. A structured 60-minute conversation with the right questions surfaces things no data room ever will — a supplier relationship built entirely on a personal friendship, a Google update the seller is quietly bracing for, an offer that fell apart in diligence six months ago for reasons that will also apply to you.
This post covers what the call is, when to run it, the 20 questions that actually produce signal, how to read response patterns, and what to do when the answers do not add up. If you are still at the stage of finding deals worth interviewing about, Deal Alert AI tracks new listings across the major marketplaces so you are not manually refreshing browse pages at midnight.
A seller interview call is a scheduled video or phone conversation where you, the buyer, ask the seller direct questions about the business, its history, its problems, and their reason for exiting. It is not a negotiation. It is not a pitch. It is an information-gathering session, and the more clearly both sides understand that, the better the call goes.
On brokered deals, the broker will usually sit in. That is fine and often useful — brokers have context on the deal history that the seller might not volunteer. But be aware of the dynamic: the broker is paid on close, so they have a natural incentive to smooth over rough answers. When a seller starts to say something uncomfortable and the broker jumps in to reframe it, note it. That interruption is data. Politely circle back: "I'd like to hear the rest of that from the seller directly, if that's okay."
The call typically runs 45 to 60 minutes. Sixty is the right target. Under 30 minutes and you are skimming. Over 90 and both parties are tired, and tired people give shorter answers. Ask for video rather than phone whenever possible. Not because you are going to run some amateur body-language analysis, but because video slows people down. Text and phone let sellers give you rehearsed lines. Video makes them think.
Key insight: Treat the call as a listening exercise, not a talking one. A good rule: if you are talking more than 25% of the time, you are wasting the most expensive hour of your diligence. Ask, then stop. Silence after a question is a tool — most people fill it with something true.
We scan Empire Flippers, Flippa, Acquire.com and Quiet Light daily — scoring every listing. Start free.
The right window is after you have read the CIM or listing prospectus cover to cover and before you submit your LOI. That sequencing matters more than people think. If you call too early, you burn the seller's goodwill asking questions that were answered on page four of the prospectus, and you signal that you are not a serious buyer. If you call too late — after the LOI, deep into exclusivity — you have already anchored your price and your psychology on assumptions that may be wrong.
Before you get on the phone, you should be able to state from memory: trailing twelve-month revenue and net profit, the revenue concentration by channel and product, the traffic mix, the multiple being asked, and at least three things about the business that look weird to you. If you cannot do that, reschedule the call and do more homework. You only get one shot at a first conversation, and sellers form an opinion about you in the first five minutes that shapes how candid they will be for the rest of the process.
There is a real competitive advantage in being the buyer who obviously did the work. On a good Empire Flippers listing, the seller may take four or five buyer calls in a week. The buyer who opens with "I noticed your subscription revenue grew 40% while one-time sales fell 12% — walk me through what changed in Q2" gets a completely different conversation than the buyer who opens with "So tell me about the business." Preparation buys you honesty.
These are ordered roughly the way I ask them. Start with motivation, move to problems, then to operations, then to the future. Do not read them robotically off a list — the follow-up matters more than the question. Every answer below should trigger at least one "why" or "walk me through that."
I am skeptical of buyers who claim to read micro-expressions over a laggy Zoom connection. What you can reliably read is response pattern — the shape of how someone answers, measured against how they answered other questions in the same conversation.
Watch for the baseline shift. A seller who has been talking in specific, detailed sentences for 20 minutes and then suddenly goes abstract when you ask about supplier contracts has told you something. It is not that vagueness is inherently suspicious; it is that a sudden change in specificity is. Same with pace. Same with volume. The comparison point is the seller themselves, not some universal standard of honesty.
Also watch what happens with numbers. Honest sellers usually get ranges slightly wrong and correct themselves — "our margin is about 34%, actually closer to 32 after the shipping increase." Prepared answers come out clean and rounded every time. Neither proves anything alone, but if every single number is a clean round figure that exactly matches the CIM, you are hearing a script rather than a memory.
Warning: Four red flags that should slow your process immediately. First, vague answers to specific questions — especially when the seller was specific elsewhere. Second, defensiveness when you ask about challenges, as though the question itself is an accusation. Third, inconsistencies between what they say and what the documents show, even small ones. Fourth, and most serious: pressure to move fast without verification. Any seller who tells you there is another buyer ready to sign and you need to skip a verification step is telling you exactly what they are worried you will find.
Discrepancies are normal. Sellers misremember. Bookkeeping is messy. The question is not whether you find a mismatch — you will — but how the seller responds when you raise it.
Raise it directly and without accusation. "You mentioned the business takes about eight hours a week, but the SOP document lists daily inventory checks and weekly ad reviews. Help me square those." A good seller says "yeah, that SOP is outdated, I automated the inventory piece last year" and sends you proof. A bad seller gets irritated that you asked. Irritation at a fair question is itself the answer.
Document every discrepancy in a running list with the date, the exact claim, and the source. If you get to LOI, that list becomes your diligence request. If the seller has told you three things that turned out to be materially wrong, you are not dealing with a memory problem — you are dealing with a credibility problem, and the correct response is to walk. There is another deal. On marketplaces like Flippa and Empire Flippers, hundreds of listings go live every month. Deal scarcity is almost always an illusion manufactured by the person trying to sell you something.
Key insight: Send a written recap within 24 hours of the call. Summarize the material claims — hours per week, transition support, revenue concentration, known risks — and ask the seller to confirm or correct. This does two things: it prevents honest misremembering, and it puts every claim in writing where it can later be attached to reps and warranties in the purchase agreement.
The quality of a seller call is set before you dial. Preparation is not just reading the prospectus — it is building a specific hypothesis about where this business is fragile, then designing questions to test it. If revenue is concentrated in one keyword, your hypothesis is ranking risk and your questions target algorithm history. If it is an Amazon FBA brand, your hypothesis is account and supplier risk. Go in with a theory.
Build a one-page brief before every call: the headline numbers, three anomalies you spotted in the data, your top five must-ask questions, and your walk-away conditions. That last one matters. Decide before the call what answer would end your interest, so you are not rationalizing after the fact because you spent two weeks on a deal and feel invested.
This preparation is where Deal Alert AI earns its keep. We aggregate listings across the major marketplaces and surface the underlying data — asking price, multiple, revenue trend, business model, age — so you arrive at the call with a full picture rather than a marketing summary. When you can see how a listing's multiple compares to comparable deals in the same category, you ask sharper questions about why this one is priced where it is. You can browse current opportunities at Deal Alert AI and go into your next call knowing more than the person who wrote the listing expects you to.
After the call, give yourself 24 hours before deciding anything. Enthusiasm decays at a predictable rate, and any conviction that survives a day of reflection is worth more than one formed in the emotional afterglow of a good conversation with a likeable seller. Likeability is not a diligence finding.
Grade the call on three axes. Credibility: did the answers hold together and match the documents? Transferability: can you actually run this thing, or is the seller the business? Upside: did the seller describe growth levers they simply did not have the time or skill to pull? A deal that scores well on all three deserves an LOI. A deal that fails credibility is dead regardless of the numbers, because if you cannot trust the narrative you cannot trust the financials either.
One last thing. The seller interview call is a two-way evaluation. Sellers of good businesses choose their buyers, particularly on brokered deals with multiple interested parties. Being prepared, direct, and respectful about the seller's years of work does not just get you better information — it frequently gets you the deal at a fairer price than a more aggressive bidder. Sixty focused minutes, twenty good questions, and the discipline to walk when the answers do not add up. That is the whole method, and it is worth more than another week in the spreadsheet.
We scan Empire Flippers, Acquire, Flippa, and Quiet Light daily. The best sub-$500K businesses are gone within 48 hours.