Have you ever heard the word “API,” and wondered exactly what it is? Do you have a vague idea of what it does or what it is used for?
This is something that you will encounter when you first start programming with any sort of web technology. Many newcomers struggle with this concept, and many get some vague ideas about it and assume that they know what it is.
Now APIs aren’t exclusive to web programming but it’s something that you use a lot when you are programming for the web or mobile. In the past, APIs were largely associated with computer operating systems and desktop applications. In recent years though, we have seen the emergence of Web APIs.
APIs can be as simple as 1 endpoint for use by 100s of users or as complex as the AWS APIs with 1000s of endpoints and 100s of thousands of users. Building them can mean spending a couple of hours using a low-code platform or months of work using a multitude of tools. Hosting them can be as simple as using one platform that does everything we need or as complex as setting up and managing ingress control, security, caching, failover, metrics, scaling etc.
What they all have in common are three basic steps to go from nothing to a running API.
Each of these steps has its own set of tools. Here are some I’ve used and popular alternatives.
REST is the most popular API interface and has the best tooling. Our design output for REST services always includes an OpenAPI specification. The specification language can be tricky to get right in JSON (how many curly brackets?) or YAML (how many spaces?) so a good editor saves a lot of time.
Four popular ones are:
I’ve only used Swagger and Postman but both Insomnia and Stoplight look interesting. All of them offer additional functionality like documentation, testing and collaboration so are much more than just specification generators.
#api #apis #api-development #restful-api #rest-api #development-tools #app-development-tools #developer-tools
If your business model involves selling to developers, you probably have already realized that much of the traditional processes and metrics applicable to traditional enterprise sales or consumer marketing don’t work. Specifically, selling to developers usually means attracting them to your platform and helping them succeed in building something, whether that’s a new app, integration, or automating an internal process. Getting developers to adopt your platform can be daunting and requires investment in product, onboarding and developer experience, and documentation. However, there are few tools product managers and developer relations leaders can leverage to measure and improve that experience.
Web and mobile analytics tools like Mixpanel and Amplitude can only measure activity on the website itself, yet most developer platforms price on what happens with the API, not via website activity. On the other hand, most API monitoring tools can only track infrastructure metrics like requests per minute and CPU utilization without any context of the user. To accurately measure your adoption and conversion funnel, you need to track usage across your entire platform from initial ad click and sign up, to what a new developer does with your API.
Most developer-first platforms have a B2D (business2developer) go to market strategy which is uniquely different from B2C companies which drive consumer adoption, but also different from B2B companies who maintain large sales forces to push their solution to other businesses. In reality, B2D sits somewhere in the middle between B2B and B2C.
B2D is like B2C in that:
B2D is also like B2B in that:
API product managers can focus on any number of initiatives at any given time, whether that;s API features, pricing and packaging, or top of the funnel acquisition. In order to know which area to focus, you should map out your _entire_funnel from very first ad click to a fully activated customer paying and referring other customers. The beginning of your funnel might start on your website as a visitor signs up to use your API. However, once they created an API key, much of the magic happens on the API side rather than the web side. Many developer platforms build their business model around usage-based pricing, which means your revenue is correlated to API usage, not website usage.
Funnel StageSourceDescriptionWhat to look forSigned UpWebsiteThe first step a developer makes is to show interest in your platform. For developer tools, this usually means signing up and generating an API key.Your analytics solution should be tracking which channel the user came from using UTM parameters, referrer tracking, etc along with which page drove the signup.Made first API CallAPIThis is a huge milestone as many signups never reach this stage. The developer was able to understand your API and give it a spin.Monitor TTFHW (Time to First Hello World) and conversion rate. You should continue looking at which channels drive the most activations.Made over 100 API CallsAPIAfter a 100 or so API calls, you could consider the developer “active”, in that they built a real integration rather than just testing with Postman or CurlA low conversion rate from the last step could imply difficulties with your SDK or unclear integration steps.Approaching Free LimitsAPIMost API products are priced on usage. In order to become a paying customer, they need to exceed some limit.If very few long-term “active” developers exceed their limits,then you may need to optimize your packaging and pricing.Converted to PayingWebsiteCongrats! You now have paying customers.Most developers have more than one value metric they price on. Keep an eye on which ones are driving paid conversions.Evangelized your APIWebsiteAre developers sharing and discussing your platform online?By adding a mechanism to track referrals, you can see which channels and mediums are performing the best. A complete analytics platform can track UTM parameters and things like referring domain. With a mechanism to generate unique links pere customer, you can also track which types of customers share the most.
Business2developer go to market models have elements of both B2C and B2B and involve both individual developers and also companies or accounts. It also involves cross-platform tracking across web and APIs. This can complicate accurate funnel and conversion reporting. Do you track an individual user sign up funnel or do you track companies who integrated and use the API? What happens if a developer clicks on an AdWord and signs up using his or her GitHub account but doesn’t do anything. If the person invited a colleague to do integration, then that person may only be attributed to “direct traffic” or “invite referral” depending on your marketing attribution. We need to still attribute the successful integration to an AdWord click.
One way to solve this is mapping out a 4x4 grid to track:
Developers active on APICompanies active on APIDevelopers active on websiteCompanies active on website
Not all API platforms will follow this model and may fill some boxes with a don’t care. For example, many APIs don’t care or track which developer is accessing the API, yet it’s important to understand which company is using the API. In this case, we want to track the following:
_Not Applicable_Companies active on APIDevelopers active on websiteCompanies active on website
In order to solve this, analytics platforms like Moesif link everything through user ids and company ids regardless of the sessions. Your API and website should use the same identities regardless of the platform for accurate reporting. By modeling companies as groups of users, the linking can be simplified. This provides flexibility in picking and choosing whether to track usage only at the user-level, only at the company-level, or both for the API and website separately.
While having permanent identifiers are great, sometimes we don’t have all information available when the API call or user action is logged. To solve for these cases, we leverage the website session token or API key to uniquely identify the person and company. An alias table that links both session tokens to user/company ids and also API keys to user/company ids enable end to end funnel tracking.
Tracking usage and retention accurately is critical in 2020 as leadership shifts from a growth at all costs to efficient growth driven from product optimizations. You can no longer just measure vanity metrics like pageviews and signups. Instead, you should be measuring the entire funnel and understand the inputs that impact each stage. How does changes to your pricing and packaging modify your conversion rates from active developer to converted paying customer.
#api management #developer marketing #developer relations #developer experience #api adoption #developer advocacy #developer advocate #api product #api program #developer evangelism
At Moesif, we primarily market to developers and other technical audiences like product managers and developer relations teams looking to make their own developer platform successful. Getting developers to adopt your API or tool is hard and requires heavy investment in experimentation and measurement in order to achieve the results you’re looking for. At the same time, much of the online advice you hear for B2C or B2B marketing simply won’t work when marketing to developers. This article outlines much of the findings we’ve seen as we grew Moesif from 0 to now over 5,000 companies including Deloitte, UPS, and Radar.
Before jumping into things you should be trying, let’s review what’s so special and hard about marketing to developers. Most developer-focused companies are founded and led by developers themselves, so you would think figuring out things like product marketing should be easy right? After all, you’re just marketing to yourselves. Just like any audience, there is no single developer persona.
Fast and iterative experimentation is critical to recoup any investment in paid ads. If you’re not diligently tracking your metrics, you can easily run up your cost per acquisition (CPA). However, what do you actually measure against for a developer tools company? Accurate attribution can be complicated since for many developer-tools, the majority of the value and magic your product creates is via an API or integration, not via a mobile or web app. This means a user who signed up and clicked around your website isn’t necessarily a converted user. Yet, there is some intent based on the content or areas of the product a user is looking at. For most developer-first companies, we recommend looking at a metric such as developers who made their first API call. Later, you can define additional funnel stages such as developers who made more than X transactions through your API or developers who used Y different features. Additional complication comes from attributing back to the marketing channel since now you have to instrument both your website to track the acquisition channels a developer went through and then linking that to what the developer did with your API.
Most social ad networks are designed to target general consumers based on personal demographic data, not developers. You can target consumers with these networks with simple Yes or No filtering. Such as Does the person live in San Francisco or Is the person aged between 18 and 24 years old. This means if you’re a pizza shop in San Francisco, it’s relatively easy to set up Facebook ads to target college students aged 18 to 24 who live in or around San Francisco. This demographic may be ideal late nite pizza customers.
For developer’s there is no single “is developer” demographic that can be answered via a simple Yes or No question. You could attempt something like _Is their college degree Computer Science”, but software engineering is a career with a variety of educational paths and does not always require a computer science degree. Even if you could match on a person’s title, software engineering is an ambiguous role with varying responsibilities. Some engineering develop embedded systems for cars and drones, some software engineers are focused on building internal data analytics infrastructure, others are integrating billing APIs.
Many developer tools are in horizontal markets targeting a variety of markets rather than singling out a specific industry, but they solve a specific problem. Like enterprise sales, this means you need to target the right developer who will have a use for your devtool and champion it internally.
The art of selling to developers is similar to most complex enterprise sales processes which have a multitude of stakeholders involved. For devtools, the number of stakeholders can be even more than traditional SaaS since you may have architecture review, security sign off, implementation, etc. This is because many times developers are checking out your tool or product for a corporation, not just for themselves. Handling these complications can be rolled into your marketing plan, rather than waiting until last minute objection handling.
Technical audiences are by nature skeptical. They love to build things, not buy services. This creates inherent biases as they are exposed to third party solutions. Due to internal political struggles, some developers may have resentment or bias against sales and marketing teams. This makes direct outreach less effective. While it does work for certain top-down sales strategies such as for selling security software to CISOs, many devtools are trying to get an individual developer to first adopt the tool and then grow from there.
One of the most effective ways to attract developers to your platform (rather than filling their email box) is with inbound marketing. Inbound marketing is the art of attracting or pulling in developers to adopt your platform rather than forcibly pushing your platform on to them. There are a variety of techniques for inbound marketing from SEO to referrals, but the focus of this article is paid ads.
As mentioned, developers are usually skeptical of traditional marketing activities. Whether this is an email from a SDR that is two paragraphs too long or a huge banner that says “Sign up now.” Being numbers-driven, they would rather cut through the bullshit and explore how something works themselves rather than be told how something works by a non-technical person. Developers also love to tinker and discover. That feeling of discovery though is a strong emotion. Think of the last time you discovered some new life hack or website that became your favorite and told your friends. That’s what inbound marketing leverages.
Effectively inbound marketing makes it appear that the developer discovered a new tool or platform on their own when in fact they were marketed to.
When done right, paid ads can drive adoption with low acquisition cost, but it requires much more planning and support than simply dumping a bucket of money on Facebook ads. If done incorrectly, you’ll still be pushing your product. The key is to be natural.
First you should decide on the goals of these initiatives, such as:
Once you have your core objectives, you can lay out a content strategy to be used for social ads just like you would create a SEO strategy and keyword planner. Content is the number one item that makes your ads appear inbound. Content should be genuine and interesting to developers, but still relatable to the specific pain your tool solves.
It’s a fine line how specific you want to be. If you go too broad, then you’re not really pulling in the right developers that would adopt your solution. You can’t push out content that tells a developer to sign up when they don’t yet recognize they even have a problem. The way to solve for this is mapping the above goals to content that targets various stages of buying funnel:
#developer marketing #api best practices #developer relations #api adoption #developer advocacy #developer advocate #api product #api evangelist #api program #developer journey
We’ve conducted some initial research into the public APIs of the ASX100 because we regularly have conversations about what others are doing with their APIs and what best practices look like. Being able to point to good local examples and explain what is happening in Australia is a key part of this conversation.
The method used for this initial research was to obtain a list of the ASX100 (as of 18 September 2020). Then work through each company looking at the following:
With regards to how the APIs are shared:
#api #api-development #api-analytics #apis #api-integration #api-testing #api-security #api-gateway
Recently, I worked with my team at Postman to field the 2020 State of the API survey and report. We’re insanely grateful to the folks who participated—more than 13,500 developers and other professionals took the survey, helping make this the largest and most comprehensive survey in the industry. (Seriously folks, thank you!) Curious what we learned? Here are a few insights in areas that you might find interesting:
Whether internal, external, or partner, APIs are perceived as reliable—more than half of respondents stated that APIs do not break, stop working, or materially change specification often enough to matter. Respondents choosing the “not often enough to matter” option here came in at 55.8% for internal APIs, 60.4% for external APIs, and 61.2% for partner APIs.
When asked about the biggest obstacles to producing APIs, lack of time is by far the leading obstacle, with 52.3% of respondents listing it. Lack of knowledge (36.4%) and people (35.1%) were the next highest.
#api #rest-api #apis #api-first-development #api-report #api-documentation #api-reliability #hackernoon-top-story