Each experiment starts as a written question with an outcome that can be wrong. A hunch nobody can disprove is not an experiment, and we do not count it as one.
Why the name
The lab is nota metaphor.
Nobody knows yet how AI answer engines decide who to name. No rulebook, no documentation, no support line. So we run a team of AI research agents that test it every day, and build what they find back into the product.
The agents work on their own schedule, overnight and at weekends. A finding is written up the moment it holds, not when somebody gets round to looking.
Interesting is not the bar. A finding either changes the product, changes what we tell customers to do, or it is filed and we move on.
How a finding reaches you
Question, test, finding, shipped
The loop is deliberately short. Most of what the lab learns is in the product within days, and you never have to read any of it.
Something we genuinely do not know about how an answer engine behaves, written down so that a result can contradict it.
Run at a scale where the answer is not noise, and repeated, because an answer engine that changes its mind next week has told you nothing.
Written up with the working shown, including the ones that found nothing. A negative result stops us all spending time on a dead end.
It becomes a change in the product, or a change in what the agents do for your business. You get the benefit without reading a word of the research.
Being straight about it
What we publish, and what we do not
Findings about how answer engines behave in general, with the sample size and the working shown. Pooled, anonymous numbers from our own measurements. Results that did not go our way.
What is currently under test, and how we do it. Anything that identifies a customer. We will not hand a competitor the method, and we will not hand anyone a shortcut to gaming an answer engine, which would spoil it for everybody.
Some of what the lab turns up is not about SurgeLab at all. When a finding is useful to anyone building on top of AI search, we would rather it were known than kept.
Straight answers
About the lab
What is the SurgeLab lab?
A team of AI research agents that run continuous experiments on how AI answer engines decide which businesses to name, quote and recommend. What they find goes back into the product and into what we advise customers to do.
Why does a marketing company run research at all?
Because nobody has published the rules. Search engines were documented over twenty years. Answer engines are a few years old, change without notice, and tell you nothing. The only honest way to know what works is to test it and keep testing it.
Can I see what you are testing right now?
No. We publish findings, not the work in progress, and we do not publish the method. Handing over the method would hand it to competitors, and handing anyone a way to game an answer engine would spoil the results for every business using one.
Does any of this use my data?
Published research uses pooled, anonymous totals only. No customer is named, and no single business can be picked out of a published figure. Work specific to your account stays in your account.
Do I have to read the research to get the benefit?
No, and most customers never will. A finding is only counted as finished once it has changed the product or changed what the agents do for you. The research is there if you want it, not a job we have handed you.