What Belongs in a Capability Deck
A deck you present and a deck you send are two different documents, and almost everybody sends the first one. Presented, your slides are a backdrop for you talking. Sent, they are alone in a browser with somebody who will give them ninety seconds.
This is about the second kind: the capability deck, credentials deck, or overview that goes out cold or after a first conversation, and that has to survive being read without you.
The reader leaves on slide three
The standard structure — who we are, our history, our values, our team, our clients, our process, our services, and finally something about the reader — is optimised for a meeting where politeness keeps everyone in their seat. Sent as a link, it loses the reader before it says anything they came for.
Assume the opposite of a captive audience. Whatever you need them to know has to be in the first two screens, and everything after that is for the minority who are still going.
Seven things, in this order
- One line: what you do and for whom. Not a positioning statement. “We build and run online stores for independent food brands” is the whole slide.
- The problems you take on, in their language. Three or four, phrased the way a client would say them out loud, not as service names. “Our site falls over every time we get press” beats “performance engineering”.
- The shape of working with you. What actually happens: how it starts, roughly how long, who they deal with, what they have to do. This answers the question that stops most people enquiring, which is not “are you good” but “what am I signing up to”.
- Proof — two or three results, with the method. Not a logo wall. Two specific outcomes with a stated measurement beat twenty logos, and a single case study beats both if you can send one that matches their constraint.
- Who does the work, named. Small buyers are buying people. If it is you and two contractors, say so — it is a strength against a large firm, and hiding it makes the reveal a disappointment.
- What it costs, at least in shape. See below. This is the slide most decks omit and most readers came for.
- What to do next. One action. A link to a calendar, or a reply-to address, and nothing else.
A deck with no price gets forwarded to nobody
The reasoning against including price is that every project is different, which is true, and that a number scares people off, which is half true. The number scares off the people who were never going to buy — the ones who were going to waste two calls first.
You do not need a price list. You need an order of magnitude and an engagement shape:
- “Projects typically run from £8,000, and most land between £12,000 and £25,000.”
- “Retainers start at two days a month.”
- “Most engagements start with a paid two-week discovery, at a fixed £2,500.”
Any of those lets a reader decide whether to keep going, which is what a sent deck is for. It also makes the deck forwardable inside a company, because the person who has to justify the conversation to a budget-holder now has something to justify it with.
What to cut
- The mission statement. It is the single most skipped slide in the format.
- The logo wall with no context. A grid of client logos says you have had clients, which the reader assumed.
- The process wheel with six unlabelled steps. If the steps mean something, write them as sentences; if they do not, cut the wheel.
- The forty-megabyte export. A deck sent as a heavy attachment is a deck read on a laptop or not at all, and a lot of first reads happen on a phone.
- Every slide you only understand when you are in the room narrating it. Read the deck cold yourself, in one sitting, and cut whatever needs you.
Reading a deck’s numbers carefully
Two honest cautions, because decks are the document where reading data is most often over-interpreted.
First, you cannot tell which slide somebody stopped on. A tracked link records that the document was opened, for how long, how many times and — where identity can be established — by whom. There is no per-section or per-slide map, and any tool that shows you one is inferring it. The limits are set out in full on the features page.
Second, the duration figure on a deck is only worth anything if it was measured properly. A deck is exactly the kind of document that gets opened in a tab and left there while somebody goes to a meeting, so a reading time computed from the gap between opening and closing a tab will hand you a forty-minute read of a document nobody looked at. Quixli counts time only while the document is actually on screen, which is why its numbers are smaller and usable — the honest way to measure reading time explains why the naive method can only ever exaggerate.