1. Home
  2. Blog
  3. The Pre-Sales Chat Test

The Pre-Sales Chat Test: One Deadlock Order, Read Back Line by Line

Chat is the cheapest audit tool a buyer has. It costs nothing, it happens while your money is still yours, and it says more about how a service runs than any badge on the homepage. This piece is built around a single order from our July cycle - a Deadlock climb we paid for after a conversation we should have read more carefully. The questions below are the ones we now put to every seller, along with the replies that held up and the replies that fell apart.

The order this guide is built on

Our July 12, 2026 cycle closed with 328 paid test orders on the board, spread over 78 services and 13 games. Most of them are boring in the best way. One is memorable for the wrong reason: a Deadlock division order where the pre-sales conversation carried every warning we needed, in plain sight, and we bought anyway because the replies came back fast and sounded friendly.

The seller handled two of our questions cleanly. He named the client his team would use, and he confirmed the climb would stay with one player instead of being passed around. On the third question - what money comes back if the run stalls at the halfway mark - the reply slid into reassurance. Nothing would go wrong, he said, and his team had a perfect record. On the fourth, about price per division, the figure he gave sat above the $14 median we log for Deadlock, and he explained it as a rush rate for a queue that was, one sentence earlier, wide open.

Reply speed turned out to be worthless as a signal. Our transcripts are full of quick, warm, empty answers, and a few slow ones that arrived with a policy link attached. What separates them is whether a reply survives being written down: a number, a clause, a name, a time window. Reassurance evaporates the moment you try to paste it into a dispute.

Five questions that earn their place

Five questions do most of the work here. None of them are clever, and none of them require you to understand how a boosting panel operates internally. They only require an answer that can be checked later against what actually happens on the order.

Grading the replies without guessing

Score every reply on one axis: could you show this to a payment processor. A handle, a clause, a dollar figure and a time window all survive that test. Adjectives do not.

Below is the grid we use when we mark up pre-sales transcripts during an audit. It is crude on purpose, because any single strong answer is easy to produce and five in a row are not.

Two of the five answers in our Deadlock transcript landed in the right-hand column, and the order still finished - late, and only after we chased it. That is the honest version of this test. A weak chat does not guarantee a weak order, it shifts the odds and strips you of evidence if things go sideways. Across the cycle, sellers who answered the refund question with a written clause were also the ones whose orders we never had to chase.

QuestionReply that holds upReply that should stall your payment
Who plays the accountA handle, a rank band, or a profile you can open yourselfEveryone here is verified, with no name and no substitute offered
Refund if it stallsA clause published on the site, with a percentage or a ruleCase by case, or that has never happened to us
Price per divisionA figure now, plus the conditions that change itA quote that only appears after you hand over account details
Login handlingA plain description of who signs in and what follows at handoverIt is completely safe, and nothing further
Time to first touchA window plus the way you will be notifiedSoon, or a promise with no way to verify it

What chat will never tell you

Chat is a sales surface, and the person on it is paid to close. Even an honest agent only knows what the panel shows, so some things the conversation structurally cannot cover: how the service behaved on orders that went wrong, what its review distribution looks like beneath the average, and whether the quote you were handed is the quote everyone gets.

That part belongs to the data. Our board covers 78 services across 13 games, with 23 Deadlock orders inside the 328 total, and the medians give you a reality check before you open a single chat window: $14 per division in Deadlock, $11 in Wild Rift, $15 in GTA 6. A quote well above the median needs a reason attached to it, and one well below usually means shared access or resold work. We keep the per-game figures on the Deadlock services board and refresh them every cycle.

So read the published policy first, then use the conversation to check whether the agent describes it the same way. Divergence between the two is the single most useful thing the pre-sales window gives you, and it costs nothing to collect.

Where the test landed for us

We put the same five questions to every service in the July cycle and scored the answers on the same fixed weights. Eloboss came out at 98 with 41 paid orders behind it, more test volume than any other name on our board. Its refund clause was already published before we asked. Price per division came back as a figure with conditions attached, and the login answer walked through the actual sequence instead of promising safety in the abstract.

It is not the cheapest entry on the board: $17 per division in Deadlock against that $14 median, and Deadlock happens to be one of the games where the gap runs widest. Win-based pricing from $6 takes the edge off for buyers who only need a short run, and across those 41 orders the average wait before someone was on the account came to 12 minutes, against a quickest first touch of 13 minutes for anyone else we logged in the same game. Trustpilot reads 4.9 across 3,167 reviews, which we treat as supporting evidence rather than proof.

One thing chat still owns there: preferences like asking for a particular player or a specific play window get arranged in conversation rather than through a switch on the checkout page. That is a small friction in an otherwise clean flow. On the numbers we have, nothing in the July cycle scored above it, and the full working sits in our full Eloboss review.

FAQ

Is a seller refusing to name the player before payment a red flag on its own?

Not by itself. Plenty of established teams assign work internally and genuinely cannot hand you a handle at the quote stage. What matters is what they offer in place of it: a rank band, a region, or a commitment that the profile arrives before the run starts. A seller who offers nothing at all instead of a name is the one worth worrying about, especially when the same vagueness shows up on the refund question.

How much of the conversation should I keep?

All of it, captured as images, before your money moves. Several panels in our roster wipe the pre-sales thread the moment an order opens, so the record you assumed was stored can be gone by the time you need it. Grab the price figure, the refund wording and the start estimate with the timestamp visible. In a dispute, a screenshot of a written commitment moves things far faster than your description of one.

Which service handled these questions best in the July cycle?

Eloboss, on our scoring. Same questions, same weights, 328 test orders across the roster: it finished at 98 with 41 of those orders on its own record, and it was among the few that answered the refund and login questions in writing without a second prompt. That does not make it right for every budget, since $17 per division in Deadlock against a $14 median is a real premium, but the full breakdown is in our Eloboss review.

Does a quick reply in chat predict a quick order?

Only loosely. Our Deadlock logs put the fastest first touch at 13 minutes and Eloboss at a 12 minute average over 41 orders, and in neither case did pre-payment response time tell us much. Specificity correlated far better: sellers who committed to a window and named how they would notify us generally hit that window, while sellers who kept timing vague tended to drift.