Ask HN: How do you sell services, instead of a product?

23 points by domano ↗ HN
Hey guys & gals,

i am working for a IT-Consultancy & Agency typically involved in backend implementation & i am a consultant with a lot of freedom to do stuff i think makes sense.

Now, a lot of the stuff i read on here about selling / marketing is geared towards products (usually SaaS) . I've tried applying this to a consulting & development-focused organization, but i am struggling to do so.

Have any of you been in a similiar position and want to drop some wisdom? Cold calling / social selling on LinkedIn just does not seem to cut it :)

15 comments

[ 4.0 ms ] story [ 44.7 ms ] thread
I have built more than one consultancy successfully and the core sales/marketing methods were networking, referrals and brand exposure. Cold calling can work, but the numbers are just 99.99% against you. Direct marketing and ads is a way we experimented but found that they were more a way to promote things we did to gain exposure then to get new work. Consultancy can be a weird/hard thing to start. When you start you have no clients so nothing to point to and the thing you need to get new clients is a success to point to generally. Chicken/egg. Hence this is where the networking comes into play, or past employers etc. Get out, meet people (digitally is fine) and use existing clients as springboards.

My general rule was when you close a deal, setup a quick schedule to deliver 3-4 things as fast as possible but with as much quality as possible, then immediately ask for recommendations/referrals to other clients. Push this until you get a few referrals/names etc but don't be a pain point -- even if they just say well Jim over at Widget corp is starting a new project, that's good enough. The key is if you wait till a project is done (or accept that excuse from a client) the chances drastically go down that you'll get a referral for a variety of reasons. You can nail a project and get new work from the client and then they don't want to refer you because they are trying to lock you up for themselves, or they worry if you expand their project will suffer etc. It can be a success paradox when you are small and starting. Of course, small clients aren't as likely to do that so those you can sometimes wait till near the end and get some referrals. But either way, it is so important to get public statements (things you can advertise) and referrals during the contract and as early as possible.

Social media is good for running ads for exposure, but it won't pull in clients generally, at least not at a cost worth paying. So networking and referrals are the best. Referrals can be from existing clients but also from friends, past associates etc that you pay for the referral.

Marketing for consulting is about exposing company/yourself online, use blog posts, guest posts, forums, speaking engagements etc. Then use social media and ads to expose your blog, speaking engagements etc, this gets you exposure, branding and then everything becomes easier.

Thank you for the insight. We have some testimonials online and do lots of brand exposure in different kinds of media, but the way you handle it seems much faster paced.

Next time i will ask for a referral after the frist few successful delivered artifacts or successful sprints.

The "Get out, meet people" part is a tough nut to crack, at least it feels hard currently. Been to lots of meetups precovid but nothing came of it & online expos are just not really valuable networking wise.

I have really gained something from your post, big thanks!

What kind of things did you consult on?
To be honest, I went through a ton of trail and error over ~15 years. So it'd take a lot to fill in those blanks. What I can say in a more short form is that I started out thinking it was best to be a generalist and work on any software project I could get hired for. Not sure if I was ignorant, arrogant or maybe a little of both, but that is a quick way to fail.

In the end what I found worked best was to specialize in a niche area that is large enough to grow but specific enough you can find ways to stand out. Part of what made us successful over time was we defined our ideal clients, our threshold clients, set up minimum project values, moved away from hourly billing, and focused as "generalist" within the niche. The final key for us was we were selling our services to bring a product to market, not only write source code for a project. There is a huge difference in value to clients and profit to the company.

Happy to be more specific if there is something I can share my experience from that might be more helpful even if it was a failure for me. FWIW -- my primary niche areas when I exited consulting was marketing automation and distributed systems to solve large data problems (ecommerce, healthcare, ads and direct marketing were the "industries" we played in).

Great reply! Thanks.

You learn the most from your failures. And this is how you succeed.

How did you find your clients? Was it mostly by your networks?

I’ve been wanting to start my own consultancy, in a niche that I understand, but the question was always whether I could find those right clients. And any company that is big and profitable enough, would just build an internal department to handle such services themselves.

Hence, the problem. The clients that can pay for it, can handle it internally themselves.

Most larger companies are the first to outsource work if it isn't a core competency or a core strategic need. This is even where you can find great product ideas to solve problems for other companies while getting paid to build a solution. So I wouldn't say just because they can afford it they will bring it in house, generally I found that not to be true. Of course there are exceptions, one is large tech companies which have a tendency to bring lots of stuff in house even if they really shouldn't.

As for finding my first clients, I turned to companies I had worked for prior, turned to friends who still worked at other companies to get my foot in the door. I started meeting as many people as possible and started speaking at local and regional events as often as I could. I'd sign up at the chamber of commerce and see what new companies were coming in, opening facilities, plants etc and see if there was a vendor list I could get on etc. A hint I learned, being on the vendor list for a large company even if you have never delivered anything to them can give you assumed credibility. When I couldn't afford trade shows I'd volunteer for companies and go work their booth, do setup or whatever I could in exchange for getting me through the door and allowing me time to explore and network.

Also, networking outside your defined niche is still good as you build your network. I used to joke it was like the 6 degrees of Kevin Bacon, I didn't have to meet my exact client but I needed to meet someone who could introduce me to another company and keep that going. Along the way you pay the kindness back with your own referrals, help etc. It is way easier to type this out than to do it, but the only way to do it is just to do it.

Yes, I also find this to be true. Large companies have sometimes long and arduous procurement processes. Oftentimes, it works best when you can get on their preferred vendors list, as they will refer to those known vendors first.

How did you bill for your projects? Did you do fixed bids? Or did you do time and materials?

I did a few fixed bid, but tried to avoid them since so much is outside your control and a larger company can easily take advantage of you (even unintentionally).

2 ways typically (most all models are a form of one of these two):

1. Per resource type/experience etc, per week (per day for < week of effort) -- typical for software dev jobs where the change requests and evolution can destroy you otherwise.

2. Results based billing -- Used when we were handling more of the process, e.g. helping to bring product to market, marketing, content, ads, campaign management, product placements etc

We mostly used a combination of the two as we matured and wanted to start make more than just a percentage on time, which to be honest is a hard way to make money and doesn't scale nicely.

Results (revenue) based billing is awesome but can also leave you hanging a little sometimes, so structure and terms are really important. Results/revenue billing is the way to make much better margins and not be tied to a persons time, but recognize it isn't without risks. Terms are critical and I generally only did these with U.S. based companies because of legal & tax complexities.

I wrote a bit about it in a Twitter thread[0]. Context is machine learning projects for enterprise in the mid six figure range that can be tackled with one or two people. Contract, problem statement, dealing with people, and abstracting into a product.

It may be relevant to you. Questions are welcome.

- [0]: https://mobile.twitter.com/jugurthahadjar/status/13106682933...

Thank you!

Some things do apply. I am not an ML expert myself, but just today i helped during the initial requirements call for a ML opportunity.

Question about 2.7.0: Currently we have an offer for a ML PoC for a very cheap price. Do you do initial work completely free of charge?

Hey I'm a data scientist with 5 years of experience building predictive models. I'm trying to start my own side-consulting company specializing in scoping machine learning projects and building PoCs and MVPs. I think I may be able to help you here. If you are interested in connecting, please look me up on Linkedin: https://www.linkedin.com/in/acauthon
Sent you a request, thank you :)
>Some things do apply. I am not an ML expert myself, but just today i helped during the initial requirements call for a ML opportunity.

Nice. It is important not to promise what your ML experts said they can't deliver.

We have conversations with everyone about a prospect/deal. Everyone chimes in with talking points, potential problems, conditions of success [i.e: we'll need to have Y data, and we need to have their sales and data warehouse teams on board for this to work], ethical considerations, engineering constraints.

We fine tune this so that we know how to help the client succeed in that project. We do this so that the client organization knows who to include in the project.

>Question about 2.7.0: Currently we have an offer for a ML PoC for a very cheap price. Do you do initial work completely free of charge?

If you consider it's very cheap, you might want to make the client see how much value they're going to get from the completion of the project and price it accordingly. It helps to be aware of the client's context, their competitor, where they are struggling, knowing the sector and key metrics. You're not selling "ML". Your helping them solve a problem, and it helps framing that in terms proper to their sector.

For example, if you're doing a PoC for customer churn for a bank, you want to do your homework about that bank, read their financial documents and get information about number of clients and the problems they're voicing (what's growing, what's shrinking, why, competition). You also want to learn how much does a churner cost.

Framing things in terms of dollars and market share instead of of recall and accuracy helps.

We offer different modalities for value to be exchanged. We have the advantage of being very flexible and able to make offers on the spot; we therefore can offer multiple ways for the client to get value of this by adapting to how their own organization works.

This is why it's important to ask questions and listen/read what they're saying to discover the reasons of an apprehension/hesitation. Sometimes it's just that it's the end of a fiscal year and they really want to do this but they can't budget for this right now. How do you then construct a deal to make the project possible?

Sometimes it's an organization bound to do public tenders because of public procurement regulations. How do you construct a deal to make it possible for them to hire your services.

You want to avoid free of charge? Get them to agree that if the PoC meets certain criteria, they'll green light the project and pay you for what the PoC would have cost. You may offer a conditional discount tied to their subscribing to a maintenance offer later.

Ultimately, the way you structure things is up to you but you may be surprised to know that organizations are more than willing to pay you, and more than you think, if you structure things in a certain way, and make it easier for them to pay you in exchange of real value.

In the aforementioned meeting we came to the conclusion that we need to include someone from another project on their side, since we could not construct a use case generating financial value for them with the people that were present.

So we scheduled another meeting. The ML experts were present too, i would not claim something that is not feasible.

Been on the other side of the fence a few times as a backend developer so i avoid this.

Your comment supports the way i have done things after getting to the point of a requirements meeting, so this gives me some confidence.

Getting more referrals for more projects would be the No.1 priority then after we have closed this one :)

(comment deleted)