It's Friday April 30, 2021.
\nAs of the time of writing, Basecamp, a small privately-owned technology company is in the middle of a stunningly disruptive pivot from legacy project management software and email service provider to an innovative publisher of founder-based opinions.
\nToday just one thing that caught my attention: a continuation of last week's episode. This one is... more conversational. Let's get to it.
\nLast episode, in Where Do You Start? I was writing/talking aloud about how, well, you get started when you know you want to kick off a new "technology project", but you know you don't want it to be done, broadly, "how we usually do technology projects".
\nThat post ended with "Yes, but what do I do next?"
\nI have to admit, the answer is complicated. And it's complicated for a bunch of reasons. And I don't really have a good answer right now, so what I'm going to do is talk through why it's been hard coming up with a good answer.
\nSure! Here's where I ended up:
\nBuy some consulting hours
\nThe consulting hours produce:
\n(Fun fact: BLUF is a term I learned while at Code for America, and I think came from someone at Code for America having spent time in the U.S. Department of Defense, where it's the functional equivalent of Too Long; Didn't Read)
\nOK, back to the main thing:
\nLet's recap: you're a director at a department that's responsible for food safety. You have an existing process for inspecting restaurants, which amongst other things involves coming up with a list every 2 years for surprise/unscheduled inspections. You think that you could probably have a better process for coming up with the restaurants to opportunistically inspect, but you don't know how to get it. Crucially, you don't want to do things the normal way.
\nIf we're thinking about concrete next steps, a literal "OK, I need to buy something", then I imagine that's going to involve filling in forms with codes in them and fields like "what is it, exactly, that you are buying?"
\nAnd this is where I get a bit stuck. Imagine you could be buying:
\nThe problem is that those bullet points all mean different things, or that they're a big messy ven diagram or, more bluntly, a consulting company can sort of deftly step from one definition to another in the process of making sure they extract as much value from you as possible.
\nAs an aside: I kind of hate how writing about this has become the sort of thing where I have "what is digital transformation" open in a browser tab.
\nBecause, are you buying digital transformation? Salesforce says Digital Transformation is
\n\n\n"the process of using digital technologies to create new — or modify existing — business processes, culture, and customer experiences to meet changing business and market requirements. This reimagining of business in the digital age is digital transformation."
\n
And already I have a bad taste in my mouth because I prefer plain language and also this is starting to read like a high school essay, "The Oxford Dictionary defines... ... in this essay I will..."
\nAnyway. The reason why I bring this up is because it's the framing issue: what you say you're buying will define not just the answers you get (in terms of responses to procurement solicitations, like requests for offers), but whom you get answers from, and also the results of that work. Which sounds obvious, but I want to think it through.
\nHere's an example. A good friend of mine was doing some work in foster care. One of the problems she found, after what you might traditionally call some management consultancy, I imagine, to identify bottlenecks and to improve processes, was that it was just taking a long time to get potential foster carers licensed. The forms, as you can imagine, were super difficult to fill in, and they needed help.
\nSo, what they did was they organized a form-filling weekend. Everyone come down. Come fill in the forms together. We're here to help you. Let's just get them all done in one weekend. A kind of hack-weekend, if you insist on thinking about things that way. Or, if you're potentially more rounded in your experience, a bit like the way libraries run sessions to help people find jobs or with tax returns.
\nWas this digital transformation? No. Can I imagine an Accenture or Deloitte or whomever saying "hey, you know what your real problem is? It's foster carer licensing, and instead of embarking on a 6 month to 3 year process of completely rebuilding your licensing process and making it Fully Phygital or whatever, we're going to start with getting 30-odd people in and helping fill in their forms."
\nI mean, do Accenture and Deloitte do that? If they do, I am impressed! I can't imagine it's great for their bottom line, though. Although I guess they do have junior associates or whomever working through the weekends, at any rate.
\nThe point being this:
\nThe way you frame something, and when it's framed, starts defining the solution or approach. It sounds really trite! Everyone kind of knows this! That's why in the previous episode, I wanted to be careful and use something that could be a real example, instead of something abstract.
\nI've talked to a bunch of people about this -- admittedly in the government/technology space -- and basically, the problem comes down to "isn't this strategy?" And at that point, you get to a somewhat distressing realization:
\nI mean, maybe?
\nWell, personally it's because I think most consulting is bad, and doesn't lead to great outcomes. I am handwaving a lot, but in my experience, consulting is an easy way to make something somebody else's problem. It's entirely reasonable to ask for help to do something when you can't do it (after all, that's what we teach children to do), but it's entirely different to continue with some sort of learned helplessness.
\nBut what are you asking for, and what are the bounds? The problem with something like digital transformation is if we use another definition, this time of "digital" from public.digital/U.K. GDS' Tom Loosemore, we have:
\n\n\nDefinition of Digital: Applying the culture, processes, business models and technologies of the internet era to respond to people’s raised expectations. [public.digital]
\n
This is super correct and also frustratingly difficult to pin down in practice, because it's foundational. How do you buy something foundational, but also bounded in scope? Remember our original motivation was "a better way of coming up with a list of restaurants to inspect than once every two years". And now, you're honestly telling me that you want to look at the entire business if you want to do this properly?
\nI mean, kind of?
\nSo then we're back to a scoping question, and a question of: well, what do you want to do exactly? Do you understand the scope and implications of what you think you want?
\nThis is, at the best of times, a difficult conversation. Or at least it can be from a human point of view, because it kind of potentially admits that you don't know what you want, or what you want to do, which can be politically fatal if perceived in the wrong way, whether on purpose or by accident.
\nDo you see the disconnect here? It looks like you just want to figure out what sort of RFO you need to help figure out the best way to approach the "coming up with the restaurant inspection list", but now that you've done some research, people are telling you that this is going to involve examining and changing your culture, processes, business models and use of technology.
\nThat's not wrong!
\nSo here's where I get stuck, and it's a super dumb place to get stuck that just reveals the fractal nature of problems.
\nI'm sitting here typing and I want to show you a real example of what it looks like when someone has a "I want a better way to come up with this inspection thing", which means finding a real RFO, which means, in this case, checking out the State of Oregon's procurement website, ORPIN, and because I happen to work with them occasionally, the State of California's procurement website, eProcure.
\nReader, they are... well, let's just say they're difficult to navigate. I have yet to find a good example for you, mainly because if I tried much harder I would've thrown something at my monitor. One reason? For California's procurement site, you can search by UNSPSC code (The United Nations Standard Products and Services Code), which you can download for free as a PDF, or you can pay for as an Excel file, which is just another example of how broken everything is.
\nSo as an alternative, I take a look at what it is that Accenture and Deloitte and I must be dumb because I can't really figure it out? I mean some things are quite obvious, in that I can understand and recognize individual words on pages like Deloitte's Government Digital Transformation Services. I mean, I can read English and I know what most of these words mean:
\n\n\nAddressing complex organizational challenges for government by integrating AI, Cloud, Cyber, and Operate
\n
(But also I need to suppress my urge to laugh.)
\nBut look, here's where I see the problem and the need to just cut away a bunch of bullshit: none of these pages really talk about actual problems. They are vague.
\nI mean, I'm super interested in how a company like Postlight works, because a) I respect their work, and b) the way they describe themselves on their front page is like this:
\n\n\nYOUR PARTNER FOR THE NEXT BIG LEAP\nWe bring strategy, design, and engineering to deliver platforms and experiences that drive digital transformation.
\n
This is all true!
\nWait-
\nI'm sorry too. I really am.
\nBecause this is a two-sided question/conversation I'm having with myself, and got stuck. Let me try again:
\nFirst, you have a problem.
\nNext: find someone to talk about that problem[1]
\nAfter that: talk to someone about that problem
\nThird: be open to the fact that talking about the problem with someone might result in potentially distressing realizations that on the face of it have nothing to do with what you thought you were going to talk about, like:
\nAdmittedly, this is a pretty specific set of circumstances and contexts. But what might happen next is something like this:
\nWait, hang on--
\n\n\nThe Oxford Languages Dictionary defines strategy as a plan of action or policy designed to achieve a major or overall aim
\n
Sorry.
\nHere's what I think what happens next, with the caveat that I am making this up by typing it out. There is a very high chance that I am wrong and need to think about this more!
\nAlso, here's a disclaimer:
\n\n\nI've been thinking about how to do this in a way that isn't blatant, so I'm just going to acknowledge: a) this is my opinion and b) I do this work. It is not a coincidence that what I am recommending and expressing as my opinion happens to also be what I do. To be even more clear: you can hire me and a team I put together to do this kind of work.
\n
The context is something that describes both the initial problem you wanted to discuss (the restaurant inspecting thing), plus, well, the context. It's important because it's what's needed for both you and, crucially, other people in your organization to understand why and if the initial problem and approach has been reframed. The context makes sure it joins up the problem with the outcome, and the context describes enough of the environment and how-you-do-things that it includes the process. It helps set up and explain why the plan isn't just "some better software that produces a list in a different way" but instead might involve something like "also write notification letters differently" and "actually measure whether fewer people get sick".
\nThe plan is the usual part: it's a bit more discovery (more user research, technical research -- if it's needed!), some prototyping and the part that describes actually building software -- if it's needed! -- the kind of teams and roles involved, the expectations needed and so on. The big thing here to note is that the plan most definitely is allowed to not include not coming up with some new Digital Service or Digital Product, mainly because it's focussed on achieving an outcome.
\nI think the idea here is that you can take the context and use that as something that turns into your solicitation for a team that will do the kind of work described in the plan.
\nPHEW. This was another long one! I'll be back on Monday with more Snow Crash commentary, as well as some short things that caught my attention.
\nHow are you doing, as we go into the weekend?
\nBest,
\nDan
\nYes, I know. Now you're asking "OK, how do I find out who to talk to about my problem?" It's turtles all the way down. Let's leave this one out of scope for now. ↩︎
\n