Most founders who do their own selling are not running one sales process. A security consultant sells a part-time CISO service to scale-ups, packages the same expertise as a product for smaller companies, and takes the occasional startup through a certification. A developer-tools founder sells to engineering managers bottom-up while also trying a top-down deal with a platform team. Each of those is a different bet about who buys, why, and how they want to be sold to. This guide calls each bet a motion, and it is about running more than one without losing the ability to tell which one is working.
What a motion is, precisely
A go-to-market motion is one hypothesis you can test: a specific offer, sold to a specific audience, through a specific way of selling. Change any of the three and you have a different motion. “Our audit service, to fintechs, through outbound email and a discovery call” is a motion. “Our audit service, to fintechs, through partnerships with their auditors” is a second one, even though the offer and the audience are identical, because the way the buyer finds you and the cycle that follows have nothing in common.
The word is used more loosely elsewhere. Large companies talk about a “product-led motion” or a “sales-led motion” to describe their whole company’s approach. That is the same idea at a coarser grain. For a founder, the useful grain is the one where each motion has its own list of people, its own message and its own funnel, because that is the grain at which you can measure anything.
Why running them in one pipeline hides the answer
The default setup is a single CRM pipeline, a single spreadsheet of leads and one inbox. Everything is in there, so it feels organised. The trouble starts when you try to learn from it.
Suppose you send outreach for three motions in the same month and book six calls. If all six came from one motion, you have found something. If they are spread evenly, you have found something different. A single pipeline records six calls either way. Worse, the stages mean different things per motion: “qualified” for a five-figure service sold to a founder is not the same event as “qualified” for a product sold to a procurement team, so the conversion rate between stages is an average of things that should never have been averaged.
The second problem is contamination of the message. When you write outreach for three audiences from one set of notes, the pains and the proof points bleed across. The scale-up hears about certification deadlines that concern the startup; the SME hears about board reporting that concerns the scale-up. Each motion ends up with a slightly generic message, and a generic message is the easiest way to make a good hypothesis look bad.
Setting up a motion so it can be judged
Before you send anything, write the hypothesis down in a form that can be proven wrong. Four things are enough:
- The offer, in one sentence the buyer would recognise, not the internal name.
- The audience, as an ideal customer profile: the companies, and the role inside them that feels the problem.
- The way of selling: channels, the first meeting you are trying to get, and roughly how the deal closes.
- What would make you stop: the signal that tells you this motion does not work, decided before you have an emotional stake in it.
The fourth point is the one founders skip, and it is the one that makes the other three worth writing. Without a stop condition, a motion that is not working gets another month, then another, because there is always a plausible reason the last batch was unlucky.
Then give each motion its own copy of the working material: its own playbook with the personas, pains and objections of that audience; its own lead list built from its ICP; its own sequence; and its own stages, with exit criteria that make sense for that kind of deal.
Running them side by side
Two practical rules keep the comparison fair.
Separate the people. A person belongs to one motion at a time. If the same company fits two motions, decide which hypothesis you are testing on it and hold to that, or the replies cannot be attributed.
Give each motion a comparable effort. If one motion gets fifty carefully researched prospects and another gets a rushed list of ten, you are comparing your effort, not the hypotheses. You do not need identical volumes, but you need enough of each that a result is a result.
Then compare at every stage, not only at the end. Replies tell you whether the message lands. Meetings booked tell you whether the problem is urgent enough. Meetings that turn into a next step tell you whether the audience can actually buy. Two motions can have the same number of closed deals for opposite reasons, and the stage where they diverge tells you what to change.
Deciding what to keep
A motion earns more of your time when it moves people through its funnel with less effort than the others, and when the deals it produces are ones you want more of. Revenue per deal and how long the deal took matter as much as conversion, because a motion that converts well into deals you cannot deliver profitably is not the one to scale.
Killing a motion is not the same as concluding the audience is wrong. Look at where it failed. If nobody replied, the message or the list was the problem, and a sharper version of the same hypothesis may deserve one more round. If people replied, took the meeting and then stalled, the offer or the price probably does not fit, and rewriting the email will not fix it. Write down which one it was, in the motion’s playbook, so the next hypothesis starts from what you learned instead of from zero.
Common mistakes
- Too many motions at once. Every motion needs a list, a message, follow-up and calls. Two or three is what one founder can run properly; more usually means none of them gets a fair test.
- Changing the motion mid-test. Rewriting the offer halfway through makes the first half of the data useless. Close the motion and open a new one instead.
- Measuring only closed deals. Early on there are too few of them to tell you anything. The earlier stages carry most of the signal.
- Sharing one playbook. The objections a CISO raises are not the objections a founder raises. If the playbook is shared, it is wrong for at least one of them.